CPUC Workshop - 2021-02-23
No chapters available yet
Legend
Transcript markers
- text Sentence being played
- text Search result
- text Intervention cited by the link you opened
Click a sentence to play the video from that moment. Select text to quote it with the speaker, timestamp and link.
More options
California Public Utilities Commission
Follow its sittingsSpeaker 2
Next, we'll have lunch, followed by the utility presentations on public safety power shutoffs, and then again another break in our Q and A session on that topic. We'll end the day with a brief wrap up and next steps. Next slide, please. So in framing our discussion today, Welfare Safety Division has asked the utilities to focus on each of the five questions listed on this slide. What progress have you achieved over the past year and how does that progress impact anticipated work in 2021 and 2022?
Question two. How do you understand your risk, and how does risk factor into decision making? Here we want to emphasize that utility mitigation initiatives should be prioritized based on the reduction of wildfire ignitions and PSPS events. It's imperative that the utilities move towards a model where mitigation activities are justified and risk reduction can be modeled and quantified. Question three, based on the modeling described above, what are your priority mitigation activities in the next year, and where are these activities being targeted?
What is the anticipated reduction in a risk as a result of these actions? Question four, how will these mitigation initiatives, in addition to reducing ignition risk, also reduce the scale, scope, and frequency of public safety power shutoff events? Finally, five, how will initiatives reduce the risk of ignition, wildfire spread, and impact the decision to use PSPS? The fifth question, of course, is essentially embedded in all of these previous four questions. Next slide, please.
So before I hand it over to the moderator, for the first portion of the workshop, I'll go over a few meeting logistics. Go ahead. Next slide, please. Most participants are in listen only mode. That means you cannot speak, but you're still welcome to interact and ask questions using the chat feature.
Please do not use the Q and A feature, as we will be monitoring the chat. The chat and Q and A tools are in the lower right hand corner of the WebEx screen. Again, please only use the chat feature to ask technical questions. The designated WebEx host is a technical resource and not a member of the wildfire safety division team. Please only direct a chat message to the host to resolve the technical issue.
If you wish to ask questions, please select all panelists in the chat function. Someone from the Welfare Safety Division will be monitoring posts. Again, please do not use the Q and A feature as we will not be monitoring that. We will route questions as appropriate to the panelists. Finally, note that we've given several stakeholder organizations an opportunity to serve as panelists.
They will be able to ask questions directly to the utilities. Next slide, please. So stakeholders designated as panelists may use the raise hand function in order to ask those questions, and the moderator will coordinate those questions. Everyone else, again, please use the chat function as described here on this slide. Depending on the number of questions received, we may not be able to cover all the questions, but certainly we'll do our best.
With that, I'll hand it over to Andy Biggs to moderate our first portion of today's workshop.
Speaker 4
Hi. Good morning. I'm Andy Biggs. I'm a utilities engineer in the mitigation branch of the Wildfire Safety Division. I'll be moderating today's morning session on grid design and system hardening, including inspections, mitigation choices, and emerging technologies.
Just a pretty broad range of important topics. Starting off, inspections are arguably part of the backbone for prioritization and understanding of the utility's growth and field. As we've seen, acid health play a major part to start up catastrophic wildfires in the past. Given the aging infrastructure throughout the state, some of which nears a 100 years old, it's imperative for utilities to conduct effective inspections and fully understand the state of equipment in order to accurately identify where to conduct repairs, replacements, and hardening projects. We've seen that traditional from the ground visual inspections have not been adequate, and advancements in augmentation to past practices would have been valuable in potentially preventing fires such as Camp and Kincaid.
From there, we heard a great deal yesterday about the efforts going into identifying areas of highest risk to implement grid hardening as a solution. From cover conductor, undergrounding, or even just the use of spacers, utilities have already been implementing grid hardening solutions as a major part of their wildfire risk reduction portfolios. While grid hardening may be an effective solution, utilities need to be demonstrating that implementation is occurring at areas of highest risk and also need to be validating the actual efficacy of utilizing grid hardening solutions, especially given the high cost and deployment time. Lastly, the utilities have all been exploring new promising technologies such as continuous monitoring sensors, rapid fall detection, and remote grid utilization. The utilities should be determining how to move forward with these solutions based on the results of research and pilots and continue to be leaders in wildfire prevention, not just throughout the state, but globally as well.
Utilities should be taking the opportunity to implement cutting edge technology that seems impossible previously in order to find optimal solutions and reduce catastrophic wildfires from occurring in the future. With that, I'll now turn it over to the utilities from the hard task that's hitting all that in twenty minutes, following the same pattern as yesterday starting with CG and E, SCE, then SCE.
Speaker 5
All right. So I'll start off. So good morning, My name is Mark Casguera. Thank you for the opportunity to present at the twenty twenty one Wildfire Mitigation Plan workshop. I'm the senior director of PG and E's Electric Asset Strategy Organization.
I'm responsible for the investment strategy and lifecycle management of our electric transmission, distribution, and substation assets. I will be presenting on PG and E's grid design and system hardening programs and answering any questions you may have on this program. So I'll go on to the first slide here. Am I there you go. Thank you.
So from a customer perspective here, you think about the wildfire risk across the PG service territory, approximately 9% of our customers are located in the High Fire Threat District Tiers 2 And 3, while over half of PG and E service territory resides in the HFTs. From an electric infrastructure perspective, nearly one third of our electric lines that provide power to our customers are now located in the HFTs. The observed risk has increased with the higher temperatures, extreme periods of dryness and extreme record winds we've experienced have increased kind of the wildfire risk across all our service territory, as well as the state. This last year was especially a challenging year for wildfires. So what you see here is that, you know, five, and we'll get into the next slide, five of the last six largest wildfires in California history occurred in 2020, and of which all were in PGC territory.
So catastrophic wildland fires, they're a major threat throughout our territory. They represent a major risk to our customers. And it's, our equipment has been the ignition source for a number of these fires. And hence, you know, really taking a really strong coordinated approach has been developed to reduce that risk. When you look at that chart here on the right hand side, you can see that 15 of the 20 most destructive fires in California have occurred within PG and CERT territory.
Furthermore, you can see that the frequency of these fires have been increasing, but also the severity or what we're calling the total structures damage from these events have also increased over the last ten years. Some additional complications that we consider is that our service territory that's classified as HFTD has grown from 15 to over 50% from 2012 to 2018. So it's been evolving. And so with this understanding, PG's focus on ensuring that our risk assessments and our operational work targets reduce the wildfire risk as aggressively as possible. So we really have to make sure that we are reviewing these and adapting and adjusting our models as well as our plans as we learn more and more about the risks in our service territory.
The next slide. So on this slide here, as we learn more about our system, you know, so what we're seeing is that we've had to make some adjustments to our wildfire risk model in a way to really take care of what we're calling as this risk buy down curve. And as a result, we've had to refresh a lot of our plans. So as we learn more about our system, we made some specific enhancements, three enhancements to the components of our wildfire risk model. Yesterday, Paul McGregor, our Director of Risk Management from PG and E, presented out on a lot of these details behind the risk modeling enhancements, which I will not be able to go into the detail that Paul went into, but I'll spell them out.
One was replacing the regression vegetation index likelihood with the twenty twenty one machine learning vegetation ignition probability, as well as replacing our equipment ignition likelihood with the machine learnings for the equipment vegetation probability as well for this year. And then notably, replacing the reacts consequence values with what we're using from the techno silva consequence values. And really the key improvements that you get out
Speaker 5
enhancements are one is our risk models is now available in absolute values. And what we're saying in come from a distance of 100 meter squares. And now those values are now additive where before they weren't. We've also increased the model and accuracy by doing some benchmarking with field folks and other utilities here to check into this. And then we've also addressed some of the overfit concerns that go along with machine learnings.
The model also highlights some of the importance of fast burning fuels. And then also we've tried to we've worked to try to capture more up to date prediction of fire behavior. So then when you look at the charts in the middle of the slide, they provide a view of what we're calling the risk buy down curves. And really, you're seeing here two curves that we're getting. And what is this risk buy down curve has changed.
And so when you look at the top chart, focusing on that top chart there, you'll see it's a view into the risk buy down curve using what we had in the 2018 model. And effectively, what you have here is each of our circuit segments, and here we're calling it circuit protection zones, we've done some work to calculate and quantify what we are seeing as the risk. And each one of those represents a dot. We've colored those dots. And when you line those all up, it forms this curve.
As you can see, this curve is pretty steep from the 2018 model, which we're on the top curve here. And what you're seeing here is that the top or what we're calling the highest wildfire risk circuit protection zones or circuit segments, you'll see CPZs from our initial wildfire risk model in 2018 are colored here in blue. And it addresses based on the way this curve is shaped about 40% of risk. As you can see that the color blue dots are on the top end of that curve, which is effectively translating to higher risk. And our work around system hardening was really focused on going after those top items, those top blue dots, those in the top of our curve, top high risk circuit protection zones.
Conversely, when you look at the bottom of the chart, which is a more refreshed view factoring in some of these improvements on our risk model, you could see that our buy down curve has a slightly different shape. And what you can see here is that the highest risk CPZ models, which are colored segments, which are colored in blue, when you factor in those improvements are no longer at the top. They're more towards the bottom half of that curve, which is really translating to a lower risk. And what we've done here is on the top end of the curve, the top 100 circuit protection zones, we've colored him in red, and we've highlighted them as, and you can see that they're on the top end of that curve. And conversely, when you compare the two curves where we had a much sharper amount, where a majority of the risks were in the top circuit protection zones, you see the curve is not as sharp as that before, where you could see it spread out a little bit more.
And so about 12.4% of that risk is in the top 100 there. So you can see that some of those enhancements really changed our risk model and how we think about what the top areas to system harden were versus what were not. And the other thing to point out here, there wasn't much overlap on what was considered the top 100 CPZs from both models. So as a result of that, we had to make some significant changes to our system hardening prioritization and risk buy down. And also mentioned before was
Speaker 5
the risk was available in absolute values now where we could apply like the distance factor where before it was not distance weighted, was more relative risk. Next slide. And so for this next slide here is our, as I mentioned, our system hardening portfolio shifting to be more of a risk informed execution strategy. And so our system hardening program for 2021 was based on this updated risk buy down curve. And we have about 180 miles that are planned for this year, where we're projected to reduce the risk by about 198 points.
So in order to formalize our plan here, we did evaluate various alternatives to system hardening that were based on risk and efficiencies, ingress, egress information. We were looking at trees and also factoring in some of the learnings from PSPS. We've taken a look at a lot of all our past events to try to understand, you know, where are the areas that are more commonly prone to PSPS, as well as our EC maintenance tags the timeframe that is allowed to execute that work here. And then so when you look at what we have in our plan from like a risk exposure and where we're going, we're really trying to count the circuit miles that are hardening our HFTDs and our high fire risk areas. We're working to, after this year, after this initial year, to really build up our portfolio and ramp up to reach a pace of about four fifty to 500 miles per year from 2022 and beyond.
And when you look at what's going to be comprised of those miles, we really are focusing on a profile where we are aiming for that 80 of our miles that we harden really have to be part of that high risk miles that are based off of our 2021 distribution wildfire risk model. And really what we're looking at in there is that we want to look at that risk buy down curve and really focus on the top 20% going after those, as well as miles that we are rebuilding as part of our fire impacts. And then also factoring in that we want to address, you know, are there miles that are going to mitigate the impacts of PSPS? From a risk effectiveness perspective, we're really looking at trying to prioritize those alternatives that provide the greatest risk reduction benefits. So really spending time with our teams to understand the opportunities on where we could do more undergrounding, as well as in the form of line removals where we could remove the risk by removing lines and going to a solution where we're looking at potentially remote grids.
Next slide. And so what we have here is, you know, when you look at our new plan on how the 180 miles are broken up, we've stood up an internal governance committee comprised of various PG and E officers that are part of our, what we'll call our wildfire governance risk committee. And they effectively reviewed our plans and was comprised of the 180 miles. And to give a double click into those miles, what we've identified is that about 80 of those miles, 81 of those miles are what we're calling in flight. So these are projects where we have work that we've received authorization.
We're further into the pipeline of having these projects engineered, procured, and constructed. And these are more likely to meet their timelines in 2021 due to their earlier start in the scoping and planning processes. The other category of projects here is what we're calling the new projects. As you can see, we've had to pivot our risk model to try to take on some of these higher risk miles. And these are starting with a little bit of a later start.
And we're calling these as new projects, which have not yet had kind of all the mitigation methods yet approved, but we're working towards that. And these projects may have a greater lead time to get the scope worked out, get the engineering and procurement worked out. So we do expect that there will be fewer miles, and that's been reflected here in this total with the 99. So that's what we've discounted already. And then when you look at the greater when you apply the discount rates of which miles are, you know, at risk of not being able to be implemented this year, it turns out to be 180 miles.
And so, again, as a result of adjusting our models and really reshaping our portfolio, we've had to pull projects into our queue for system hardening that weren't originally conceived to be queued for a few years. And some projects that were in the queue, we've moved them down into the process so that we can focus on these higher risk models. Next slide. And so as I talk about here, our plan is to build up this pipeline so that by 2022 and beyond, we're really hitting a pace of four fifty to 500 miles per year. And as you can see here, the way you read this chart is there's the table is really reflecting the different phases that go in through the lifecycle of a project.
So you've got your project scoping, the estimating, clearing out the various coordinating the dependencies of, you know, permitting and clearances, and then those that are going into construction and then really closing out. As you can see that we have a plan to really hit a pace greater than 500, but a lot of those miles are really in the early scoping phases. When you look at it, this is really looking at the projects. We have over 150 projects that are falling into this particular category here. Again, the balance of those are in the early scoping phases that we're really launching.
Some of those projects here can be scoped in time potentially for the 2021 execution and really, you know, if those can advance further, could increase our total from the 180 miles that we're targeting this year. And you can see where the miles are and how those could help influence and raise our 180 miles that we're targeting for this year. So there's more miles that's in our pipeline. But as we've mentioned, we've discounted some of them because of where they are in the process. But should we be able to kind of clear those phases and really move it into the pipeline, we could get more miles in there.
However, the way we're really looking at that, most likely they won't be scoped in time, which is why we're really building the inertia to get it into our pipeline going in for 2022 and beyond to really hit this sustainable rate. Next slide. And so what you have here is, you know, as I mentioned, the limitations for what's going on in our portfolio here is, you know, although that we're doing less miles, one thing that's really important to point out is that even with the less miles, when you compare it against the previous portfolio that we had planned, we're actually reducing more risk by targeting these higher risk miles than if we stuck to the original portfolio that we had identified that was based off an earlier model. And so although we're doing more miles, we believe that we're actually going to be clearing more risk here in the short term here. And just to kind of give a view of some of the constraints that we have when you look at it from a permitting perspective, you know, so we can see here that the days can add up in terms of the average amount of time that it takes for permitting.
So that's some of the constraints that we're working towards to try to address here as it pertains to hardening. And then when you look at the internal project scoping process, you know, a typical project can take a little bit over a year to a year and a half if things move as planned. And so kind of the change in work plan, you know, has to account for that additional lead time to do those different work from, as I mentioned, from the field and final scoping, the estimating, the dependencies and contracting and the different approvals and construction before we close out. And so we'd say that we've taken the information from this updated risk model to be able to really pivot our portfolio to try to address those highest risk items while factoring in PSPS impacts, construction constraints, and really, you know, from a risk and efficiency, what work is driving the highest amount of risk reduction in this timeframe here. At this point, concludes my presentation and definitely we'll hand it over back to the host that will be available for any other questions.
Speaker 4
Thanks so much, Mark. It's going on to SE next to hear about their grid design and hardening efforts.
Speaker 3
Yes. Thanks, Andy. I'm not seeing the slides on screen. Are we up? Are you seeing them?
Oh, there it goes.
Speaker 3
Good morning, everyone. And thank you for the opportunity to be here today and share with you what's in our wildfire mitigation plan update for 2021. My name is Russ Ragsdale. I'm a director of our asset and engineering strategy group. I have similar responsibilities as Mark does, responsible for our distribution, transmission, substation assets, including kind of our long term strategies, determining where we make capital replacements, determining how often we inspect and maintain them and developing several of the mitigations that I'll share with you all here today.
You can go to the next slide. So there's three kind of key topics that I'm going to touch on today. The first is grid design and system hardening, share with you some of the mitigations there, what we did in 2020 and where we're going in 2021. Next, I will speak to the asset management and inspection scope of work. Again, what we did in 'twenty and what we did in 2021.
And then in the third section, I will share some examples of some of the emerging technologies that we've included in our WMP. We can go to the next slide. So our grid design and system hardening mitigations, as it says here on the slide, are really implemented to maintain, strengthen, and upgrade our equipment and infrastructure, really with the focus of reducing the risk of ignition in our HFRA areas. The kind of underlying foundation for this is our risk model, which we follow a bow tie framework, at SCE, which really determines risk drivers, which is the probability of a risk event, as well as risk outcomes, which are the consequences of a risk event. And so our system hardening mitigations primarily are focused on reducing the probability of ignition, and thereby doing that, the overall risk is reduced.
When we look at our drivers of suspected wildfire initiating events, The primary drivers that we see are contact from object and wire to wire events. Those were associated with approximately sixty percent of those suspected wildfire initiating events. So that's why you will see that cover conductor is the largest mitigation that we're deploying because it is effective in mitigating those contact from object or wire to wire event drivers of potential ignition events. In 2020, we were able to complete nine sixty five circuit miles, which was above our target of 700. And in 2021, our target is a thousand miles with a stretch goal of getting up to 1,400 circuit miles.
We continue to believe that carbon conductor is the most effective way to mitigate these primary risks. And by deploying it across the system, we will be able to mitigate the risk of wildfire emission. We have several other system hardening activities that are included in our WMP, some of which are listed here. Continue to deploy fuses. In 2020, we deployed over 3,000 fuses, combination of new fuses and replacing fuses.
And in 2021, we expect to replace or install that three thirty that you see on the slide. Installing these branch line fuses is an effective way to reduce the risk, especially on branch lines where we don't have the fuses, because it allows you to interrupt a fault much more quickly than if you didn't have a fuse, which reduces the amount of energy that's delivered to that location. Circuit breaker relay hardware upgrades has a similar purpose. In 2020, we did a little over 100 of those. In 2021, proposing to do somewhere between sixty eighty six of these upgrades.
And then undergrounding and vertical switches are two system hardening mitigations that we didn't we primarily did the planning for in 2020 so that we can do these implementations in 2021. In undergrounding, we're doing four to six miles. Four miles is the target with up to six. And for vertical switches, planning on doing 20 up to 30 vertical switches. In addition, in our 2021 WMP, we are proposing some new mitigations that were not in our previous WMPs.
For C hooks, we're proposing to replace 40 C hooks across our system. C hooks are a piece of hardware that you will find on our transmission infrastructure that holds our insulators in place. The C hooks that are on our system are primarily there from when SCE acquired another utility many, many years ago, Cal Electric, and C hooks are not in our standard for normal transmission construction. We have done some inspections and assessment of our Seahooks, and we found that some of them are deteriorated because of the age of them. We've also found that it's somewhat difficult to identify the level of deterioration on COX.
And so because of those cases, because the deterioration and the difficulty to inspect and the fact that they're not in our standards and the relatively known number of these that we have, we are proposing to replace at least 40 of them, if not all of them in our high fire area, which is up to about 60. For vertical switches, we've done some assessments and we found that vertical switches that are installed on wood cross arms can become misaligned when that wood cross arm over time is exposed to changing temperatures, and therefore it warps. It can shrink. And then it can cause a vertical switch to be warped and the contacts to not be perfectly aligned, which when the switch is operated, could lead to incandescent particles, which could be an ignition risk. And so because of that, we have identified a vendor that would provide vertical switches attached to composite cross arms.
And we plan to replace the vertical switches that we have, 20 of them in 2021 with these new types that are not subject to the same failure. The third item on here that's new is our long span initiative. What we have found is that there are certain types of construction in the field that are subject to conductor clashing, either because the span is very lengthy or because the span happens at an angle where you're traversing one direction with your span of wire and then the next span is traversing another direction. Or you have sometimes where you have vertical construction and it transitions to horizontal construction. Those are some examples of scenarios in which the spans are subject to clashing, the conductors clashing during high wind events.
And so because of these, we would like to go out and remediate these. Remediations are a combination of installing spacers between the conductors. If that's not an effective mitigation, then you can change the construction at the cross arm to kind of widen the distance between the conductors or change the cross arm out itself to help widen the distance. And so we plan to do 300 of the highest risk locations up to 600 in 2021. Then the last item on here is microgrids.
This is more of a mitigation of the PSPS risk that we're facing and the impact of customers of de energizing circuits. And so what we are planning to do in 2021 is evaluate the installation of a microgrid to mitigate this. So we've identified a location in our high fire area that is subject to PSPS risk. And we are actively exploring installing a microgrid in that location to keep those customers up in case of a PSPS event. If we can go to the next slide, please.
Okay, this slide is really providing some examples of where we're using risk to prioritize the various mitigations. So on the left side here, we're talking about cover conductor. Similar to what Mark showed during his presentation, we're able to evaluate the risk of each span in our high fire area, again, based on probability of ignition as well as consequence of that ignition. And we are deploying our carbon conductor based on that highest to lowest risk circuit segments. Now we continue to evolve that risk model, as you probably heard yesterday during the risk presentation, and incorporate kind of new information and new technology into the model.
We also, as we're deploying our cover conductor scope, we look for operational efficiencies and being able to bundle the work together. So being able to do cover conductor along with some of the other mitigations that are out there and being able to help avoid some of those challenges that Mark spoke about regarding permitting and being able to do the construction. The third bullet here is one example of what I talked about of our risk model continually evolving. In 2020, we are using the TechnoSilva tool, which is a tool that helps us to better assess at a more granular level what the consequence of ignition would be. And so we've updated our risk model to incorporate that.
And again, are using that as a way to determine the kind of order in which we would deploy cover conductor in our high fire areas to make sure that we're deploying it in the highest risk areas. We've also incorporated the PSPS risk into our risk model. Again, looking at the probability of a PSPS event and the consequence to customers of that PSPS event. So as we move forward, we will look to deploy our cover conductor and our other mitigations based on that updated risk model, which includes risk of ignition as well as risk of PSPs. On the right side of the slide, we're showing our the long span initiative that I described thoroughly on the previous slide.
In terms of how we use risk for this, so the way in which we identify these at risk spans is we use LiDAR technology. So we fly and take a LiDAR scan of the spans, and then we process that internally to look at where the spacing is less than a certain threshold. And then based on the probability of conductor clashing together, as well as the risk of a given location, we prioritize which spans to mitigate and kind of work through that in that order from highest risk to lowest risk, similar to how we're doing. For undergrounding, we look at kind of multiple criteria to determine if undergrounding makes sense in a given area. Oftentimes, undergrounding is much more expensive than doing overhead work like a cover conductor.
So we're looking at the cost of the undergrounding work, but we're also looking at what are the risks in a given location, not only are they ignition risks, but as well as the PSPS risks. We're looking at the terrain, how constructible it is, considering the community, egress of the community, and that cost. And so based on that and the types of risks that are present in the area, in some cases it makes sense to do undergrounding in lieu of doing something like carbon conductor because of all of these factors. So in 2021, again, we're planning to do, you know, those four miles of undergrounding up to six miles. We are also evaluating kind of long term costs and benefits of undergrounding, including impact on potential inspections in the future, potential vegetation management, to really be able to do a better assessment of undergrounding as compared to other mitigations.
So continuing to evolve our thinking in that space. We can go to the next slide. So this is the second topic, asset management and inspections. So our plan in 2021 is to continue to do perform a three sixty degree distribution and transmission inspection. So what we mean by three sixty degrees that we're performing both a ground and an aerial inspection of the same structures.
And the structures that we've identified to inspect in 2021 are 59% of our distribution and 53% of our transmission structures. That covers structures that are responsible for 99% of the total wildfire risk. You'll see some more detail on that in the next slide. We will do ground and aerial inspections on 163,000 distribution and 16,800 transmission structures. In addition to these visual inspections, we're also deploying different technologies to be able to inspect other parts of our system.
So we're using infrared, Corona scanning, I mentioned LiDAR previously, And then these high definition images and videos, we're using a combination of helicopters and drones for aerial inspection, really looking for kind of all the different ways that we can identify hazards out on the system. We're moving towards being able to use machines to help improve the quality and increase the efficiency of our inspection process. And so we can train those machines using machine learning by capturing images on a consistent basis, we can train those machines and have them be able to identify hazards that maybe a human might miss, or be able to do it in a fashion that is more quick than a human can do it. Our initial deployments of this will be supplemental to what the human is doing and help to make sure that, you know, we're catching everything that's out there. In addition to our base inspections, we've put a carve out in our WMP for some optional inspections based on emergent risks.
So our wildfire risk model is primarily based on probability of events happening due to high winds, but there are other factors that can contribute to risks out on our system. And so these additional 30,000 distribution and 3,000 transmission give us the opportunity to assess kind of year specific conditions that are emerging based on fuel or weather patterns that we would determine the scope in Q2 of this year and then launch later on in the year. And lastly, we're implementing technology to help increase the effectiveness and efficiency of inspections, including mobile tools, data management and storage technologies where we can store the images and all the data that we're capturing, be able to access it, ultimately to assess the condition of our assets. We can go to next slide. So a little bit more detail about how we're using risk for our inspection.
So the chart in the upper right, this four by four, has probability of ignition on the y axis and consequence on the x axis. And so what we've done is we've placed kind of all of our high fire structures in boxes in this chart, and you can see the percentage of structures in there. And so all the ones that are highlighted in red there are the structures that we are intending to inspect in 2021. That represents the highest probability of ignition and the highest consequence structures, therefore the highest risk structures. This chart is particular to distribution, but we have one similar to that that exists for transmission.
As I mentioned on the previous slide, we're using other technologies to perform inspections as well. So for distribution, we're using infrared scanning. And we do 50% of our distribution high fire structures per year. So we did half of them last year or half of them in 2020, and then the other half will happen in 2021. And again, we organize them by the highest risk districts in the first year and the lower risk districts in the second year.
And then on the transmission side, we use aerial outfitted with infrared and corona scanning to scan our transmission lines. And we did a little over 1,000 miles in 2020. And again, our goal in 2021 is to inspect 1,000 miles. Again, that scope is prioritized based on operational constraints, as well as the highest risk circuit miles. And then, again, as I mentioned on the previous slide, we have areas of concern, which are areas that because of fire history or weather conditions or fuel type, egress, various other considerations that are beyond our welfare risk model, we will evaluate and assess and determine if we need to do those inspections within the year, because these are all conditions that change over the course of years.
And so assessing them within the year gives us the best information on to what extent we need to do these inspections. Okay, I'll move to the next slide. Moving on to our third topic, which is emerging technologies. So as part of our plan to mitigate ignition risk, not only are we deploying many of the kind of more commercially available system hardening mitigations that you saw in a previous slide, but we're also exploring new technologies and their ability to mitigate this risk. Our approach to evaluating and adopting these technologies is shown there on the slide, where we typically evaluate them through a desktop analysis evaluate what the potential risks that can be mitigated by them, what the costs are, what the availability of technology is, what the challenges are.
Then we pilot it out in the field. Once that pilot is successful and we feel comfortable with the technology and being able to operationalize it, we'll move to a small scale deployment. And then after that, it up and implement it as a programmatic application. So in 2021, we're evaluating various new technologies. We've highlighted three examples here just to give some examples of what we're deploying.
The first one here is the rapid earth fault current limiter, or REVCOOL for short. We're implementing three versions of this across our system. The first is a ground fault neutralizer. The second is arc suppression coil, and the third is an isolation transformer. The first two, the ground fault neutralizer and arc suppression coil, are installed within a substation.
And so that allows us to reduce the fault current that is delivered to an event across all the circuits in that substation significantly. We expect that both of these, the ground fault neutralizer and arc suppression coil, reduce the fault current to just a few amps, which really reduces the risk of starting an ignition when there is an event. And the isolation transformer has a similar effect, but this is installed outside of the substation fence on a single distribution circuit and allows you to kind of isolate a single circuit and provide the same type of protection where you're reducing the fault current delivery. Our system is designed in such a way where we have customers serve phase to neutral and because of the way that these protection devices work, in some cases, they're not compatible with that phase to neutral serving of customers. And so the isolation transformer allows us to find circuits that are more compatible with this type of protection scheme without having to completely change out how all the customers on that are served.
So by having these three flavors of the rapid earth fault current limiter, we can deploy them in different scenarios where the existing configuration of the circuits or the substations is compatible, thereby reducing costs. If we had to upgrade or change out all the ways in which customers are served to support these, we'd be talking about really high costs. So again, we're just piloting these. We expect in 2021 to be able to operationalize the ground fault neutralizer and the arc suppression coil. And actually at the 2020, we were able to get the isolation transformer installed and operationalized.
So we're looking forward to seeing how these devices react and respond during the pilots and their ability to reduce fault current. Second item on here is the distribution open phase detection. And so this is, again, out in the field, not in a substation, but out along a circuit. We install two devices, reclosers, that are within three miles of each other or even closer in many cases. And they have high speed communication between them.
And what they can detect is if a conductor becomes separated or becomes an open phase. And so when that happens, they can de energize the line so quickly that it can de energize before the conductor touches the ground. And so it's a really effective way to protect against an ignition or potential ignition. So we've piloted several of these out there in 2020, and we'll continue to monitor those and you'll see how they perform in the field. And then the last technology on here is our EFD, the early fault detection.
This includes this is devices that we install again on the field, but they're not necessarily detecting voltage and current as a traditional electrical device. They're detecting kind of radio signals that are sent when there's some kind of issue out in the field. So many times what they detect are, you might have a conductor that has strands that are being deteriorated and that causes a little bit of arcing and emits a signal. And so these early fault detections can discern what we call incipient faults that are existing out there in the field, allowing us to go dispatch folks to that location and fix that before it becomes a bigger issue or before it comes down or is damaged by a foreign object or during a high wind event. So this is an exciting technology as well.
We've about half a dozen events that the detection picked up and we've deployed crews to the location. And we found places where the conductor itself was deteriorated and we've been able to proactively fix it. So again, you know, this is a technology that we think can provide a lot of benefit, but still testing it out to make sure that, you know, there's not false positives and that it can detect kind of all the different events that are out there. I think the last slide is more details on these three that I dug into, so let
Speaker 3
Yeah, again, maybe I'll just summarize them here. Rapid Earth Fault Current Limiter really reduces the amount of energy that's delivered to a fault, thereby reducing the risk of ignition, being able to interrupt that fault before a fire can start. The distribution open phase detection is sensing for when that conductor kind of breaks apart and de energizes in the time it takes to hit the ground, thereby reducing that risk. And the early fault detection is looking for not events, but scenarios on the system where you have something deteriorated. And because it's deteriorated, it's giving off some kind of radio signal, allows us to proactively dispatch folks to the field to resolve that issue and make sure that doesn't become a bigger issue in the future.
So with that, I think I hit my time and be happy to accept any questions when we get to the Q and A portion.
Speaker 4
Thank you so much, Russell. And then we'll move into our final presentation from Eugenie with. Looks like it's getting pulled up now.
Speaker 7
right. Hello everyone. I'm Tyson Klinick, the director of Electric Distribution Operations for SDG and E, and today I'll be covering grid design and system hardening and particularly the goals around hardening and how we prioritize our hardening mitigations based on our risk modeling. Today, I'm going to walk us through some of the data that informs our risk models, and then Jonathan Woldemerian, our Director of Wildfire Mitigation, is going to get into our actual rate of hardening and inspection mitigation programs. Next slide, please.
So this is a system level view of how our risk event and emission data, consequence data, and overhead exposure all look at the high level. As you can see, a risk event that occurs in the HFDD is twice as likely to result in an emission based on our five year average, and it was actually four times more likely to result in an emission in 2020. On the consequence side, SDG and E's techno solid fire spread model show that the vast majority of impact for wildfires is located with it inside the HFTV versus outside. So even though we're reporting risk events at the system level, in WMP table seven, so that we can learn from every event on the system, the work we are doing is really about reducing risk events and admission percentage within the HFED as that's how we maximize risk reduction. The rest of this slide shows that we did reduce risk events in the HFED in 2020.
However, ignition percentage actually rose in 2020, we believe due to being an extreme weather year in terms of FPI and red flags. We're currently 25% hardening the HFTD, and as that percentage continues to rise, we expect risk events to continue to trend down with ignitions following, again, depending on weather adjustments. Next slide, please. So this slide drills down even more to show the risk drivers within the HFTV with the highest average admission totals and associated risk events. SDG has mitigation programs that target every single one of these drivers.
Hardening programs like under rounding and covered conductor mitigate every single driver on this list, while mitigations such as our enhanced veg management program that we discussed yesterday are very specific to targeting vegetation related risk events within HF TD. There are some nuances here like lightning arrestors, for example, they're a little bit different. We have a program to replace lightning arresters with a more fire safe version. These arresters won't reduce the number of risk events on the system as lightning arresters are there to prevent over voltage and will fail as designed to protect the system from overvoltage, but these are designed to fail in a way that's less violent, less prone to ignitions. So with lightning arresters, we're expecting to reduce ignition percentage.
Finally, while it's good to understand the individual risk drivers, SDG and E's hardening goals are really around reducing risk events in general. For example, just because animal contacts only has like a 1.87% ignition rate, You know, we're not exactly sure that they're less risky than vegetation contacts because of the problems we have with small sample sizes. Every risk event has the opportunity to lead to an admission and it really depends on the location and the weather at the time that it happens. And at this level, we should expect high variance due to the small sample sizes. That said, it is important to know the drivers that are causing risk events and ignitions on the system so that we know we have mitigations that are targeting the specific drivers.
Next slide, please. So in this slide, we wanted to talk about some of the studies we performed this year to either measure or model the risk reduced, which then went on to inform our RSC calculations and our risk reduction models that Mason and Sarah discussed yesterday. Today I wanted to highlight our barriconductor hardening studies for transmission and distribution. For transmission, we did see an 83% reduction in risk events going from about six faults per year per 100 circuit miles to one after hardening. And for distribution, we saw a 47% reduction, going from around 12 faults to six faults per 100 circuit miles per year.
When we look at the system level, it's a little hard to tell, how effective the mitigations are because it's mixed with all the unhardened infrastructure. But when we zoom in on the completed work, we did see significant reductions in risk events. To perform this study, we looked at over 200 miles of completed distribution hardening jobs where we had the completed poll numbers and completion dates. Utilizing the poll number to relate datasets, we were able to look at our risk event data from 2000 to 2019, look at risk events that happened to those structures before, risk events that happened to those structures after, and then we were able to normalize by operating years and mileage to get the final results. Next slide, please.
So this slide shows, our top 10 RSCs across all mitigations and the rankings of RSCs among our three primary system hardening. The two, the top two are controls, so we're not like mitigating any additional risk by doing them, but we believe they are saving us many ignitions per year for very little cost, which is why they're at the top. The third mitigation is kind of a key takeaway. It's something new. In our original WMP filing, we said that this mitigation had an RSE of zero because it didn't mitigate wildfire risk.
With the WINGS model quantifying PSPS impacts, we can now quantify and compare the benefits of PSPS specific mitigations and sectionalizing can be very good. A well placed switch that isolates underground customers from overhead customers can save impacts to hundreds of customers from PSPS for relatively low cost. It is interesting to see that equipment mitigations rose to the top, but it does make sense. Something like capacitors causes an ignition just once every five years within the HFTV. But because we only have 100 capacitors to target, we can mitigate every one of them to reduce the risk down to once every twenty five years based on the effectiveness we believe we'll get from the mitigation.
So it doesn't reduce a lot of risk overall, but it's very efficient. As you can see, hardening mitigations do show up lower, but they are critical to complete because the hardening mitigations are the only mitigations that address kind of our top risk drivers, the drivers that are causing the majority of our, of our, risk events and emissions. These are all the foreign object in line and equipment failures such as wires and wire bounds. Next slide, please. So, I'll go through these next two slides quickly because they were covered in yesterday's presentation, but they represent how we prioritized our previous work and how we're going to prioritize our hardening work in the future.
All of our previous fair conductor hardening projects have been prioritized leveraging SDG and E's WORM model, you know, and output visually that you can see here. This strategy of replacing the highest risk assets led to lots of like 10 to 40 structure hardening projects across the HFTD, but was like truly asset risk based and not tied to the way we operate our electric system. The updated model, which leverages the same data as the WAR model, can you go to the next slide, please, reflects how we operate the system. So it's going to prioritize hardening in a way that all the way to the segment level, which is, you know, the switches that we actually operate in PSPS events, which is going to allow us to realize both grid hardening or both wildfire risk reduction benefits along with the PSPS benefits. And as Jonathan goes through, his updates on the hardening projects, we're gonna see the shift in strategy going forward, as we shift away from kind of our bare conductor hardening programs and more towards covered conductor and undergrounding because they provide the dual benefit of risk reduction along with PSPS mitigation.
Next slide.
Speaker 6
Thank you, Tyson. So my name is Jonathan Old Vironment. I'm the wildfire mitigation and vegetation management director with STG and E. I'm going to continue from where Tyson left off with the information that allows us to rank and assess as well as prioritize our hardening as well as designs. And it's important to note that with SDG service territory, 64% of our service territory area is within the HFPD.
And that means that we not only leverage the risk ranking and the RSE scores that Tyson shared with you, but also leverage our extensive weather data and fire science expertise that we have internally and that we've developed over the years, including the mapping of our areas, highlighting the highest wind areas so that we understand how to design in particular known local conditions. That was an important early step that we took in order to design programs that were more effective. An additional tool that or tactic that we use and highlights our efforts is that we use lidar and PLS data, as mentioned by Eugenie and Addison as well, in our design tools so that we can then also be more effective in understanding a more comprehensive program rather than kind of an asset by asset approach. And then finally, the post construction QAQC checks. So the quality assurance that we do post construction using LiDAR and using other tools has also been very effective to make sure that our construction meets the design standards and addresses the high wind speeds that we're trying to mitigate against in our hardening.
So with this first slide here, we have these various programs that have been in place for many years in SDG and E, the FIRE Risk Management Program, and the PRIME program, which is mostly focused on poll voting. And then our Cleveland National Forest Focused Hardening, which went in and looked at our transmission and distribution lines and is going to be almost completed. It was almost completed in terms turning in for some additional work that we're gonna do in 2021, first quarter, to complete that. And that really helped us to harden one of our highest risk areas within the Cleveland National Forest in San Diego. And so this was a really important project.
In fact, we have taken out and removed from service a transmission line that goes directly through the middle of the Cleveland National Forest as a result of this project. And then finally, our transmission hardening effort has been in place since the February, which has led to us being able to be more surgical with our PSPS events and not take out as many customers with over 400 miles of hardening of transmission with an HFTV. So those are some key steps. So in 2020, we achieved over two twenty miles of hardening. We'll continue to use the bare conductor of the traditional, as we call them, traditional hardening approach for 2021 and 2022.
Beyond 2022, we're going to shift our hardening from the bare conductor to more of the covered conductor and undergounding, as we'll see in the coming slides. So we'll go to our next slide now, please. And I'd like to highlight additional things that were part of our program. Inspections, of course, that we do for compliance and our regular maintenance programs that we do, as well as some specially focused programs that we have for inspections in the Hi Fi Threat District. So for example, annual patrols of over 86,000 poles each year.
We have a five year cycle for detailed inspections that we conduct for over 17,000 poles each year. And then we have the HFDD specific inspections that we conduct every year for a three year cycle, and we do close to 12,000 poles each year. And this will continue into 2021. We also have a wood poles and shoes inspection program that does over 14,000 polls last year, and will continue that. Here, we're just highlighting two of our newest programs that we added in 2020.
One is the drone investigation assessment repairs, which is the drone inspections program over our tier three distribution system, where we inspected over 37,000 poles. And what we did there was we took in more detailed photos using drones, and sometimes it led to over 40 pictures per pole where these photos were then taken and reviewed by qualified electrical workers and assessed to identify any damage and issues with public. That has been a very effective program, and we will continue that in 2021 and 2022 to also inspect all of tier two within those two years. The distribution infrared inspection is also a newer program. We've always conducted transmission infrared inspections, but now this distribution focused infrared inspection that began last year will continue into tier two.
We did about 17,000 poles in tier three. Now we're going to move to tier two and see what differences we can identify and what benefits or things that we might learn different from the material key inspections. And so those are things that I wanted to highlight as some new tools and technology that we're using. One more thing I'd like to express about the drone inspections is that we're leveraging machine learning in order to use it to identify or assess the photos, instead of having the person go through thousands of photos, sometimes millions of images, then having the machine learning and being able to identify more quickly will lead to faster remediation as well. So that's how we're highlighting the use of AI or machine learning as well.
So on our next slide, we'll talk about some of the asset based specific programs of hardening that we have. Expulsion fuses, we've conducted replacements for those for several years now. We expect to complete that in 2022. Hotline clamp replacements, these are connectors that have been a problem for us in the past, where the connectors to put wires together have failed in causing emissions. So we've progressed quite a bit over the years, we continue with this program in 2021 and 2022.
And then skater capacitors. Tyson mentioned those were fewer events, but they were high risk. So what we're doing by going to more remote controlled and state controlled capacitors and HFTE is being able to learn earlier for any issues that may come up, but also safer operations for our employees and then less emissions as a result of better monitoring and better equipment in the system. Lightning arresters is something new that we're going to start this year, and this is a result of our risk ranking and recognizing the Latin investors creeping up in risk. And so we're going to do a lot this year, over 900, and then continue that program until we resolve it.
Now, something to keep in mind with these assets is these are highlights of some of the asset specific programs that we have. But then we also have technologies that we are utilizing, like the volume conductive protection, right, and enhancing our communications platform to enable high speed communications, as well as the early fault detection that Edison shared with you. We're also pursuing these high technology solutions along with the equipment that we're installing and making sure that we're staying up on the technology to advance safety, wildfire safety in particular. So with all of this to say that we're evolving into the technology, with the technology, as well as our approach as different risks creep up. And the one final thing I want to mention regarding asset specific programs is that we're not just addressing these assets as part of specific programs, but also in the comprehensive overhead hardening that we talked about in the earlier slide, the traditional hardening, and as well as the covered conductor and undergrounding that we will address here in the next few slides.
But here in the next slide, we'll talk about the covered conductor. Next slide, please. And so our shift from the bare conductor, which we also refer to as traditional hardening, over to the covered conductor, started in 2020 with only about two miles worth of covered conductor. And we took a slower approach because we were learning the work methods and the standards that we wanted to apply and the tools that we needed for our field employees to make sure we got the job done right. And so in 2021 and 2022, we're going to leverage the learnings from 2020 and escalate our application of the covered conductor.
The covered conductor helps to reduce risks, we agree with benefits for a covered conductor. Avoiding animal contact, blinding the balloon contacts definitely will help with that. And then we're going to learn more as we deploy more of the covered conductor in future years. In our next slide, we'll talk about the undergrounding strategy that we have. And the strategy for undergrounding here really tries to address two risks, right?
The wildfire risk, which we've always tried to address through our Harvey program, but also the PSPS events risk. What we're trying to do is see if we can reduce the number of impacts from the events by undergrounding strategically in the areas that have highest exposure. So with this year in 2021, we're going to focus on Alpine Boulevard, Cannon, and Wisconsin. These four areas are areas where we're scoping out some undergrounding. We completed 30 miles, the highest completion of undergrounding in the HFTB, specifically for wildfire mitigation that we completed from previous years.
And then we're going to continue in 2021 and 2022 to grow the program even more. And then beyond 2022, we'll scale up as well. One of the things that we're doing, and we have learned in the first year in 2020, is that the workings with the counties, the other jurisdictions that are within the HFD area, to make sure that we work out some processes to increase our productivity and undergrounding and make the processing faster. But we're also looking at the easements that are needed, right? We can't really follow the overhead lines to route to the undergrounding.
So most of the undergrounding will probably go to roads. And so the routing and easements that you have to acquire takes longer. So we're planning ahead and putting in that to replace earlier than we would with the overhead construction and planning. So with that, let's go to the next slide. And here I want to highlight the microgrid technology that we're using.
As I mentioned, we're not only shifting on our strategy for the wildfire hardening, but also in making sure that we address the SPS attached to our customers and our community. Four microgrid sites were put into service this past year. The technology that we used and that is displayed here is different. The technology that we used to deploy the generators for these sites was conventional technology. Going forward and this year, we will put in service the renewable solutions for each of these sites.
The the four sites on the left here, the Ramona Air Tap Base, Cameron Corners, Butterfield slash Alcaliente and Shelter Valley are the four sites that I'm talking about that were put in service in 2020 with the conventional technology. But now, in 2021, we'll have the solar and energy storage for both sites except Ramona, which will have energy storage. And just to share with you the benefits, these communities, especially Buck Cameron Corners, Shelter Valley, and Butterfield Ranch, are in the desert areas of San Diego County. So the transmission lines and distribution lines that feed these circuits are traversing through the high fire threat district and high risk areas. So we de energize these distribution lines, especially during PSPS events.
Therefore, the microgrids will keep these communities in the desert energized through the PSPS events. So with that, were able to keep 119 customers, for example, on Butterfield Ranch and two nineteen customers in Shelton Valley energized during these last PSPS events over December 2 and December 7. The Ramona Airtide Base is a strategic microgrid site because of the cal fire air support importance and resources that are flown out of river and fires. So we thought that it was a key area, a critical area for energy storage, lack of power. New sites that will be reviewed and implemented by the 2022 are the two sites on the right, South Tampa and Charlottesville Valley.
And some of these sites are not just to keep customers energized during PSVFTS events, but also to keep the convenience in the essential stores and convenience stores and essential services hot or powered during the PSBS event. So people can get their cell phones charged or go to get water or buy essential services from stores. So those are some of the goals for some of the sites and not just keeping intact communities powered during PFAS events. Now on this next slide, we'll talk about the generators that we have also deployed to customers. Next slide, please.
So our customer generator program, starting from the left, it's a generator grant program that provides customers with Goal Zero Yeti solution, which is a battery that helps customers, medical baseline customers in particular, keep their equipment energized, very important for them in the home. Next to that, on the right, mobile home park project. Did not complete any projects in 2020, but in 2021, we have a few sites identified. And this is to energize essential community centers within the mobile home parks so that communities can have a place to go during TSPS events. And then the generator assistance program is a rebate program that was provided to customers where over twelve seventy four customers were able to take advantage of the rebates and get generators to help them light the power of PSPS events.
And then finally, the fixed backup power program is a program that provides customers with a fixed solution. So it's a generator that is installed permanently on their property with their permission, and they will then operate and control that generator going forward to help them also like the USPS events. And so these four solutions are part of our grid hardening, not just because we're looking at it from a system hardening point of view, but also providing solutions to customers for TSPS events, as I mentioned, the dual solution that we're looking at. So I'll wrap up with just saying that our risk ranking and efficacy programs have informed how we do our grid hardening and how we are planning for future grid hardening as well, but also the programs that have evolved due to not just having a look at our assets. Tyson talked earlier about a lot of what we were doing was hardening our spans that were smaller wire or longer spans.
Now we're shifting to more of a programmatic approach where we're doing full segments, as Sarah and Mason also highlighted through our risk modeling approach. And then finally, the dual effort that we're trying to do and achieve by reducing wildfire risk as well as PSBS risk. So thank you for your attention, and I'll conclude there. Look forward to your questions.
Speaker 4
Great. Thank you so much, Jozin and Tyson. So with that, we're gonna take a short break and meet back here at 11AM so we can go into the q and a portion. I'll start off with a question from wildfire safety division, move on to the panelists. And then once we get through the first round of panelists, we'll move on to the chat and kind of repeat the pattern.
So with that, the first question I had was directed mostly at PG and E since their presentation focused primarily on the grid hardening aspect and not so much the asset management or emerging technologies. But some of these questions also get into more specifics for us, Eugenie and Nessie. I wanted to know how the utility is using risk modeling to inform and prioritize its increased inspection efforts, and also wanted more details on how those increased inspections differ from routine inspections both in scope and content. And also sort of building on that, was curious how mitigations were being implemented as a direct result of improved inspections.
Speaker 5
Hey, Andy. This is Mark Casguera. So, there's a lot to unpack there in that in that questionnaire. So I'll try to answer kind of the things that I heard, we could have some follow ups on that. So from an inspections perspective, our approach for where we're at right now is like in Tier three, with our higher tier.
We have Tier two, Tier three. Our Tier three, we are still continuing to inspect that annually. So we're doing a detailed inspection of all our T and D and assets in those areas. For Tier two, we're moving you know, we had a cycle that looked at all the assets in Tier two. We're moving to a one third cycle on that one.
And what we're looking at there is, from a risk based view, we're trying to understand where are the highest risks in those areas and really trying to prioritize those to be inspected. But factoring in that we just inspected a group of those assets the prior year. So trying to get the top 33% each year to assess, then moving to a three year cadence from that perspective. How are inspections informing our asset management and other cycles there? Definitely what we're doing, you know, as part of like some of the work that we are doing in terms of scoping out for PSPS events is that we actually factor in the condition of the asset based off of some of the non conformances that are identified on our facility.
So facilities that have higher risk tags, we definitely prioritize to have those repaired before fire season. And if we're not able to get to all those, those are items there that we factor into our decision making for when we conduct our PSPS. As well as those non conformances that we identified that have a longer lead time, we really try to bundle all that work into a larger system hardening project, particularly if it has more time before it has to be addressed. We really try to optimize our resources, our views, our inspection information to develop our system hardening plans and really try to prioritize those in an optimized fashion. So that's kind of what I heard generally with some of the questions that you brought up, but I know I didn't cover all of them.
So if there's any others that I didn't cover, feel free to kinda restate those.
Speaker 4
I was wondering if you had any details on how the inspections differ from routine that you just described apart from timing.
Speaker 5
Yeah. So I I think for for our areas, we're moving more to using the the detailed enhanced inspections in general. I'd say our substations have kind of the regular routine where we're checking our substations out on a monthly basis. And then the other type of inspection we have is on the patrols. And the patrols are not as enhanced as the details, where we're really just trying to capture things that are kind of visible from the eye.
And the other kind of our detailed inspections is really looking at, you know, what are all the different failure modes that we're concerned about from an ignition risk perspective. And then for our non tier, non wildfire risk from a public safety perspective. So it's a more enhanced view of items that we look for. And then so you have your enhanced inspections, and then we have our patrols, which are really trying to capture those things that could be visibly seen. And then what we also have is for those locations where we have an open maintenance tag that is projected to go close to its due date or its timeline, we'll do a field safety reassessment of those assets for that particular structure to see if the reason for that asset has degraded in any fashion.
And if it is, then it gets escalated to be dealt with, whether through a repair or replacement. Or it gets reassessed that we can continue to monitor it.
Speaker 4
Got it. Thanks so much, Mark. I guess with that, can we move into SE, Ms. Russell?
Speaker 4
no. It's the same question. Going through the difference between the enhanced inspections and how results from those inspections are prioritizing work being done in certain mitigation.
Speaker 3
Yeah, so in our high fire areas, our enhanced inspections and our, I'll call them compliance based inspections are really done the same way. So we're now capturing more information when we do the inspections. There's a mobile application that we've deployed to our inspectors and they have a survey in there. It's kind of a smart survey. So based on what they see on the poll, the survey, they'll answer a question and it'll kind of give them different questions.
And it's really intended to capture not only the condition of the infrastructure and identify any hazardous conditions, but also to capture information about our assets so that we can use that information later to assess the condition of the assets and make some decisions on maybe proactive work that we need to do accordingly. Typically our compliance based inspections in overhead systems distribution are every five years and transmission it's every three years. So these enhanced inspections now are increasing the frequency, at least for the higher risk assets. Remember I showed you that kind of four by four table. So for the higher risk assets, we're doing them annually.
So increasing the frequency by which we see, by which we get our eyes on the assets. We're also combining ground with air. So prior to deploying these enhanced inspections, we were only doing ground inspections, but now we've combined that with aerial to get that three sixty degree view. And it has allowed us to identify some things that you cannot identify from the ground. Deterioration on the top of a cross arm or on the top of a pole.
So we've seen some areas, actually many instances where that's been an issue and we've been able to catch that from aerial inspections. In terms of how we
Speaker 3
risk to the priority. So as we identify hazardous conditions in the field, we categorize them as priority one, priority two, priority three. Priority one is an issue that needs to be addressed immediately as typically addressed within twenty four hours or at least made safe. And then priority two conditions typically like twenty four months to address, but in high fire tier two, we address them in twelve months and then high fire tier three, we address them in six months. And then the priority three conditions we address in sixty months.
And so, as I described, we are determining what we're going to inspect based on that probability of failure and the consequence chart that I showed earlier. And then throughout the course of the year, we're looking to inspect the higher risk assets earlier in the year and kind of work down that risk curve so that when we identify those hazardous conditions, they're remediated earlier in the year ahead of the fire season. Did I address everything or was there any other parts of it?
Speaker 4
I think that covers it for now. Thanks so much. And then moving on to SDG and E, either Jonathan or Tyson.
Speaker 7
Sure. Yeah. So for for our enhanced inspections and how they differ from our compliance inspections, first, it's it's really the they differ based on essentially the cycles and the technology used to complete the inspections. So our compliance based inspections are, like the other utilities, detailed inspections once every five years. We inspect our wood poles intrusively once every ten years, and we do an annual patrol once a year for the entire system.
So our enhanced inspections or the inspections that are additional or incremental to the compliance inspections include a ground inspection of the tier three structures once every three years. So it's essentially increased cycle times to make sure that we're looking at those poles more frequently. And also as part of that, they're specifically looking at condition codes that are related to risk events. So it's a more specific inspection, looking at the types of conditions that could lead to failures that would result in a risk event, which we define the same way as like a spark on the system. In addition, we have our drone inspection program, which is leveraging the technology to talk down three sixty views of the structure.
We've had a lot of, we've seen a lot of success from this program, a lot of conditions turned in that had not been seen previously given the different perspective. So a lot of value there. And then we leveraged infrared as well to be able to seek things like flat connections. So leveraging different technologies to see things that you can't see from the normal inspections. We also then have a priority system in how we go about repairing what has been identified on these various inspection programs.
We have emergency where we believe that the failure is imminent based on what we have found, and we set a, you know, essentially a same day, a zero to two days timeframe to get that issue repaired. We have priority, which is a goal of replacing or remediating the issue within one month. And then we have what we would consider normal, which is we do believe that, you know, if repaired within one year timeframe, it would not fail until the next, you know, it wouldn't fail before that timeframe. So we allow one year for all of the normal type findings, unless it's in the tier three HFTTD and then six months.
Speaker 4
Thanks so much. Think that should cover it for now. Moving on to the panelists, I'm going to give it over to Tern with either Marcel Howager or Katie Marconi.
Speaker 9
Yes. Okay. Thank you very much. So I have a quest I have a few questions, and I'll just start with one. Although it's about the scope of work for the covered conductor grid hardening work, but before I ask that, I like to get I need to get a clarification from the panelists about the units that are used in their WMPs so that we know how to look and compare their work.
Table 12 that showed, for example, the historic and forecast spending and the work, it looks like PG and E used line miles and SDG and E used not line miles, but Edison used circuit miles. Specifically for the item on the covered conductor installation, I wonder if the panelists could just explain whether those units are comparable or not. And more specifically PG and E for 2020 showed four sixty million for three forty two line miles and Edison showed five forty six million for nine sixty five circuit miles. I'm wondering if those are comparable or if how one can compare those units.
Speaker 3
Yeah, I can start and explain what a circuit mile is for us and then maybe PG and can explain what a line mile is for them. So a circuit mile, so typically when you run a circuit, you may have one to three phases running down the circuit on your main line, you'll typically have all three conductors, three phases running, and then you may branch off and have, you know, two phases going or one phase, you know, neutral going. And so we use circuit mile to represent a mile of that conductor that includes the multiple phases. Now you could break it up and say conductor miles, that's another term that we use. And then if for example, you had a one circuit mile, but it had three phases, then you would call that three conductor miles.
But for us, a circuit mile is that one mile of the circuit that includes the multiple phases.
Speaker 9
Fine, a one circuit mile would involve three miles of conductor, that would be one circuit mile in your table. Is that correct approximately?
Speaker 3
If it was a three if that portion of the circuit was a three phase portion of the circuit, then yes. Yes.
Speaker 5
So can you I think I've got my audio back on. Can you hear me? Yes. Okay. All right.
So the way we're looking at it here is that there is the miles of the distance of just the stretch of the run that's where folks are looking at some line miles. When we're looking at it more from a circuit perspective, it is looking at all three conductors that are going to be placed into service there. I know that we are also preparing a data request response on this. So one thing I need to do is I actually want to get aligned also internally to make sure that we're consistent on that. So just to kind of get back to you there on Marcel, is that we'll have to validate on how we're counting that in our data request, because see the question that you have there and I just, I've got to go back to the WMP and understand how we reported out on it.
Speaker 9
Okay. And I guess, I'm sorry, know SDG and well, SDG and E, do you have anything are your line miles similar to the actual linear distance of the entire span?
Speaker 7
Yeah, it's the same as how Edison described. We're going the distance between two structures and all the equipment replacements, conductor replacements within that span. So we're measuring NSR circuit miles.
Speaker 9
Thank you. Well, with that explanation, my actual let me ask my question, which is that looking at those numbers, it seems that Edison's circuit covered conductor work is much less expensive per the same total line mile than PG and Es or SDG and Es. Very roughly, it looks like it's about a half to a third the cost. And I'm wondering if you could all sort of discuss a little bit what might be any differences in the scope of work for covered conductor installation that explain the difference in price per mile?
Speaker 5
I can start off for PG and E. Maybe the bigger difference is that as we put on the covered conductor, there's a lot more replacement of poles that we're going to have to include because our poles aren't necessarily sized to handle not only the weight of the cover conductor, but when you factor in the cover conductor and the expected winds in those high risk areas, adds a lot more pressure on those poles and the necessary equipment holding up the conductor. So I believe our costs will be a little bit higher because it factors in more of the retrofit and replacement of the poles.
Speaker 6
Yeah, on the SGD increase, this is Jonathan Holmgren. Just like to add that, you know, we agree that the whether you replace the poles or not could make a difference on the cost per mile, but also the type of structure that you're replacing with. If you're going with steel poles versus probably glass or others, that could make a difference as well. As well as, you know, terrain and other, you know, permitting and other issues that may come along depending on where you're doing the work.
Speaker 3
Yeah, just echoing the comments of Mark and Jonathan. I think the assumption around pole replacements and what types of poles are replacement makes a big difference on the unit cost. There's probably other factors, but I think the number of poles that are being replaced is a big factor.
Speaker 9
And if I may follow-up on that, I guess, we continue to sort of try to investigate why is there such a difference in the number of poles that need to be replaced between the utilities. And I confess the data we've seen regarding the existing poles is confusing. And with respect to the issue of supporting the weight of the covered conductor, I guess we've been unable to obtain sort of pole loading data from PG and E that would support the need to replace all the poles just due to the weight of the conductor, and I'm wondering if you have any additional information or thoughts you can share about that.
Speaker 5
Yeah, we're willing to share also our standards for system hardening that get into how we're sizing holes in our high fire risk areas, as well as, you know, the kind of the pole types that we're including. So we're welcome to show that information if that'll be helpful. And we also included some of this information in terms of poll loading data in our GRC. So if that'll be helpful to package that up and then share with this group, we'll gladly do so.
Speaker 4
Thanks so much for our follow-up. Do you think that covered your question for now?
Speaker 9
Well, I think I still have some additional questions, but I think that is sufficient for now. Thank you very much. And I have some questions on microgrids, but please just come back to me later when after other people have asked questions.
Speaker 10
Hi. This is Joseph Mitchell, MGRA. Just a follow-up on the covered conductor. It seems that SCE has really relied on covered conductor as a hardening technique, whereas PG and E and SDG and E are significantly further behind on that. And it's rather striking in that there's an overlap of service territory between where conditions are very similar between SCE and PG and E areas in certain areas in the North of SCE, South of PG and E.
And SDG and E service area is very similar to the southern parts of SCE's service territory. Why is it that a covered conductor is so compelling for SCE, but not so compelling for SDG and E and PG and E? Why is there such a difference? And a related question, because I think that's going to be part of the answer for SDG and E is what fraction of hardening in the high fire threat districts is already complete. And I'll just present that to all the utilities.
And this is sort of historical and going forward in the sense that how did we get to this particular place where a covered conductor is compelling for one utility but not for the others, and is it going to be compelling for all utilities going forward? Thank you.
Speaker 3
I can start. I can't speak for the other utilities because we do have different configurations, different service territory, different voltage levels, but for Southern California Edison, our decision to use cover conductor is really tied to the risk analysis. As I mentioned in the presentation, when we look at the major drivers of events that could potentially lead to ignition, contact from foreign object and wire to wire represent approximately 60% of the events. And cover conductor is, you know, a great solution for that. By covering the conductor, you mitigate that impact when something else hits your lines, as long as it's not like a tree falling on the lines.
And you also mitigate that wire to wire contact. And so because it's highly effective at mitigating our major drivers, that's why we have identified it as the primary mitigation alternative.
Speaker 5
Thanks Russ. And I'll jump in here for PG and E. I'd say it is a compelling mitigation for us. It's definitely something that we are working towards. In 2020, we did a number of system hardening.
Think we reached us slightly over 300. We exceeded the target that we had set out for ourselves. This year, the target's a little lower because we've made some changes to our risk model to improve it and really try and target the highest risk. And so covered conductor will play a role in the 180 miles from there. And, you know, when you look at our plans going forward, we're actually looking to spend
Speaker 5
Okay. I'm back, sorry. Sorry for the connection here. One of our reasons, some of our challenges that we're working through is really getting through the ramp up phase of getting our cover conductor and our system hardening program into the pace that we were talking about, which is about four fifty to 500 miles per year, so that we can harden our areas. And some of the challenges we've had, as I mentioned, is that a lot of our infrastructure, we can't move as quick as a pace as we wanted in earlier years.
One, it was a new program that we were launching. There's an understanding that there would be a significant amount of retrofit with our poles so that our pace has not been as quick as we'd like. But we are working to improve upon that. And also, you know, so cover conductor is definitely something that we are moving towards and really want to embrace. But also, some of the lessons that we've learned from the last two cycles of PSPS is, you know, we also want to factor in how we could make the solutions work for us to not only address the wildfire risk, but also mitigate the impacts of PSPS.
And in some cases where you have heavy tree density, very heavy strike potential, covered conductor is not going to prevent some of those incidents that you have in those areas. And we're really taking a closer look to see, are there other areas where we could further increase the amount of undergrounding that we could accomplish? So, I'd say cover conductor is a big part of it. We're really spending we're planning to spend significant resources in there, but we also are really trying to understand how we can address the PSPS mitigations and, you know, are finding more opportunities where we could underground our facilities. And that'll take a little more time, and we want to be very strategic.
As you've seen San Diego, we're really taking a look to see how they're managing things. And I think they've got a good approach on being very targeted and strategic to not only address the wildfire risk, but mitigate the PSPS impact.
Speaker 6
This is Jonathan. I'll take a stab at answering the question about the cover conductor from Joe. So for SDG and E, it was just like Mark mentioned, it was an effort to take it slow with covered conductor to make sure that we understood how much undergrounding versus covered conductor we were going to do, as well as the, bare conductor ramp down. So we have projects in flight for bare conductor that had been already identified in flight. So making sure that we understood how to, you know, ramp that down and shift strategy as well as how what's the balance, the right balance of public electric versus undergrounding and figuring out where to best apply that using our risk modeling, using our risk ranking and understanding the whole picture versus just directly going into covered product replacement for the conventional hardening that we were doing.
Coupled with that, though, we were also developing work methods and standards internally to figure out the best way to go out and install the cover conductor. It wasn't just a matter of purchasing the cover conductor and applying it in the same way that other utilities would, but really figuring out what is the most efficient, best way for our employees to install the covered conductor. One, for example, technology that we were able to leverage is the fact that we're not we don't have to strip the cover connector. We have a piercing connector that can make the connection of the cover connector to our structures easier, right, using the right connectors. So those types of things helped us to enhance our approach a little bit better.
And we feel we have a balanced approach on the whole strategy of unraveling versus cover collecting. And then finally, you asked about the percentage of hardening that we have so far. With over 800 miles of distribution hardening within the HFTV, HFTV is about 3,500 miles. We have 23% hardening along distribution. On the transmission side, it's much higher.
It's over 65% in transmission because we have been at it longer on the transmission side, but also have been doing there's less mileage on the transmission side. So already done that.
Speaker 10
And what do you see as your long term goal as far as the are you going for 100% in the HFTT, or what do you see as final target?
Speaker 6
Yeah, so for the HFDD, we're still using our risk modeling, but if not 100%, a high percentage of hardening, whether it be covered connect or underground.
Speaker 11
Hi. Thank you. So my question for all the utilities is how is egress and ingress included into the vocational prioritization for the grid hardening activities?
Speaker 3
I can start. One of the ways in which we're looking at egress is when we're evaluating underground. So we have some parts of the territory where there may be only a single road in or out of a community. And while, you know, the risks for that scenario may be still may be contact from object or wire to wire contact, which would identify cover conductor resolution. We consider that egress when kind of comparing that mitigation with undergrounding as a mitigation.
There's many other factors to consider costs being one of them. Also whether or not there's large trees in the area that could lead to events, potential ignitions. So there's a lot of pieces that go into analysis, but that is one place in which you're using egress. Another place in which we've looked at it was when we did our areas of concern inspections in 2020. We identified 17 areas of concern.
As I mentioned earlier, we looked at fuel conditions, weather conditions, wind conditions, history of fire, but we also looked at egress as part of that to determine if those were areas that rose to the top in terms of where we should be doing inspections.
Speaker 5
And for P and E, just want to add on the egress side. We also factored in more on the scoping side. Once we have identified our high risk areas, as part of our alternative assessment, do factor in. We bring in our public field safety specialists to come in and others to give their view perspective on what the egress ingress is in that area to help us better, you know, select the appropriate alternative, whether undergrounding would be a better alternative or relocating some of our facilities out kind of major highways or roadways is one of them. A future improvement, which is something that we want to improve on is how do you build that into our risk model?
Our earlier risk models had an egress model in there. And after we did a deeper dive with it and brought in some additional experts in that area, we found out that there was a lot of opportunity to improve in that. We're now looking to try to build, bring in a more informed egress ingress model to be part of our model, to not only be part of the risk model, but also part of the scoping process when we're developing our projects.
Speaker 7
Yeah. SDG is the same way. It's considered in the scoping as we're doing kind of the alternatives analysis on planning actual projects. It really gets factored in two ways. The wildfire spread impact analysis, there's gonna be more impacts near communities.
So a lot of the prioritization, like you'll see it's a lot of the hardening work we're doing are around our rural communities, Julian, Ramona, Descanso. So a lot of the work is kind of focused there anyway. And when we look at specific ingress and egress, it's the same type of conversations that we have as far as one more data point to potentially push something towards undergrounding versus covered conductor, especially typically when we're talking about ingress egress concerns, we're also talking about more concentrated populations and some benefits around GSPS mitigation as well, by going with the underground mitigation.
Speaker 11
Great. Thanks everyone. So it sounds like it's not integrated into the initial risk model. And then it's basically integrated in specifically for undergrounding or different mitigation after the risk model kind of determines the higher highest risk circuits. Is that correct?
Speaker 5
Yeah. For PG and E, it's similar. It's once we've identified our high risk areas, and then we start developing our alternatives and scoping out the project, egress ingress issues are factored in there and informs the preferred alternative.
Speaker 4
Great. Thank you. I have a clarifying question for PG from what I remember, I think egress used to be a factor within calculating risk, that was sort of decoupled. And so sort of curious as to why that was the the choice or why that occurred?
Speaker 5
Yeah. Earlier risk model did factor in egress ingress. And what we did is we started doing some benchmarking and brought in other folks, other experts in that field of egress ingress. And they took a look at and reviewed the model and gave us some really good feedback that there are really a lot of opportunities that we could take on improving that model. And so that's been kind of we took it off of our models while we work to get a more informed model that better reflects the egress ingress issues.
So that's kind of the reason for it. We took it out because we're working towards trying to find some additional help in that space to build up an egress ingress model.
Speaker 4
Great. Thank you so much. With that, we're going to move on to Henry Burton with Cal Advocates.
Speaker 12
Okay, good morning. Everyone hear me okay? Okay, thanks. I think the first question I want to ask is for your asset inspections, how do
Speaker 12
determine which asset inspections to perform in the earlier part of the year before the wildfire season? So let's say January through July before the really peak wildfire risk period. How do you sort of sequence and prioritize your asset inspections and determine which ones should come first? Let's start with PG and E.
Speaker 5
So I think one of the things we look at, I think you touched on it, is we really want to get all our inspections done before the peak fire season. And I think we are aiming for, you know, by before the July or sooner. This allows us to address any of the non conformances that if there's an immediate repairs, we have enough time to deal with them before peak fire season occurs. So definitely from that perspective, we've talked about for the different tiers, we're looking at 100% of our assets in Tier three and the top third for that year for Tier two. And I think as we start to prioritize that, I think we factor you know, what are the assets that have kind of the highest worst conditions.
But it's also working with our work execution teams of, you know, where we could, you know, efficiently deploy our resources out to do the inspection. So we want to make sure when we're sending our teams out there that if they could look at more than one set of assets at that time, we want to make sure we factor that in there so that we're not just going purely risk based, we're in a way that it sends resources to a from a really better utilizing your resources and time. So really looking at what are all the assets in that particular area, trying to prioritize kind of the highest risks for those assets, and then assigning them all to kind of coincide. So it's kind of a blend of all those different items there. And I think the key there is that we want to accomplish all those before July, or if there's assets that we know have a higher risk profile, really trying to attack those areas of the map sooner than later.
Speaker 12
Okay, but you are doing inspections throughout the year, right? So I mean, not able to complete all of your inspections by July. Am I wrong? Did we lose Mark?
Speaker 12
Mark, it may help if you turn off your background effects because I think we're having bandwidth issues.
Speaker 5
Okay. I'll just turn off the camera in general. I think it's easier for me to manage it that way. So all our high fire threat areas, our tiers two and tier three, will be done. Our plan is to complete all those before the July.
So that one's pretty clear. But to balance in the non wildfire areas, tier one, I think those we have the balance of the year. So those get prioritized in a different timeframe. I don't know if that helps. More about the tier ones.
Speaker 12
No, that's helpful clarification. Thank you. Great. SCE, could you answer the same question?
Speaker 3
Yes. I took off my background effects. Hopefully that helps. In a similar way, we're focused on high fire parts of the service territory early in the year. So our plan is to finish all of our high fire inspections in August.
And then after that, then we move on to the parts of the service territory that's outside of high fire areas. But even within the high fire, we're also look to prioritize, you know, showed you that four by four that we have that has probability and consequence on the axes. And so we're looking to do earlier in the year, the ones with higher probability and higher consequence in the upper right. And then kind of ride that risk curve down as we do inspections throughout the year. Now that is subject to operational efficiencies.
Know, we want to make sure that when we send inspectors out, they're not like jumping around the whole territory and can be inefficient, but you know, looking to do the highest risk early on while accounting for those operational efficiencies.
Speaker 7
Yeah, so for our inspections, again, the compliance based inspections, we're just gonna doing throughout the year on our normal schedule using the interval process. But for the specific tier three inspections that are specifically designed to reduce wildfire risk, Those are done by threethirty one every year. So again, with the goal of getting them done very early in the year to have time to complete all of the emergency or priority findings that come out of that before fire season.
Speaker 7
Yeah, the tier three HFTD ground inspections, the additional incremental ones we do every year.
Speaker 7
Tier two areas, we have the drone program this year, but no, we don't have any specific programs that are parting the Tier two in the same way that we're doing the Tier three.
Speaker 13
Thanks very much. Very helpful presentation, very informative. I had just two points I was hoping to follow-up with PG and E on, if you could just clarify it within your WMP. And then I had a question for all panelists. So for PG and E, in your WMP, you described relative risk prioritization scores, but in your presentation, you identified a risk reduction of 198 over 120 miles.
So just trying to understand that because usually I see risk reduction stated as a ratio, and just wanted to see if you could clarify that. Along with, in your WMP, it says, there's a section on 106 that says finding fault with PG and E's historical equipment maintenance. But I didn't see in your presentation where you were identifying infrastructure that caused problems in the past that you fixed in the future. As an example, with the Kincaid fire, I understand that there was infrastructure that was abandoned or decommissioned. And just trying to understand if from that you went out and sort of made sure that other infrastructure that was decommissioned was not still energized.
And then the question for all the panels, a follow-up on Doctor. Mitchell's question, which is around the covered conductors. Just wanted to know specifically after a catastrophic wildfire, if it is your practice to make sure that covered conductors are installed throughout, or after a wildfire do you install non covered conductors and then sort of later assess where a covered conductor might be situated. Thank you very much.
Speaker 5
All right. Thank you, Mr. Abrams. This is Mark Eskara from PG and E. I will, I think I got all three questions there.
So the first question on the relative risk versus what I reported out on in terms of absolute risk. So one of the changes that we made in this year's WMP was on our distribution risk model. I think I talked about the various enhancements. But one of the improvements that we made is that we were able to move away from a relative risk ranking, basically ranking our facilities relative to each other. And then, you know, so that was more of a relative ranking where you didn't really have an absolute score, a risk score for those facilities, but it was more relative to, you know, this facility is higher risk than facility X or Y.
Where this year, our model that we've done is that we've actually been able to enhance it to a point where now we're moving away from the relative risk and getting into an absolute risk score, where we can actually have, you know, a point score where you can actually add them or subtract them for that matter. Where before, using that relative risk ranking, it wasn't quite as precise as that. So that was why this year we were able to, with a lot more information, be able to show the absolute risk score than we were in previous past where it was not quite as precise back then. So I'm not sure if that helps, mister Abrams, on on that first question.
Speaker 5
That's based off the multi attribute value function units. So information, we could provide that as a follow-up. And then the oh, can you hear me now?
Speaker 13
I just caught the last part, but I understand. I guess we can follow-up after. That's fine. Thank you.
Speaker 5
Yeah, so just to restate, there's a formula for the MAVF that we can provide and you can see how that is computed. On the second part here to your question on, I think you were referring to the items that you brought from the Kincaid fire. So for those, when we look at these investigations, as well as when we have failures of our assets, do work to try to understand, you know, what failed, how it failed, what's the extent of condition. And for that particular scenario, it was dealing with idle lines, as well as a configuration of how we've set up on our structures, a dead end on our structures. And so we've identified, you know, where are all the idle facilities within the service territory within high fire threat areas and have developed our plan to not only ensure that they're de energized, but also making sure that they're grounded appropriately and a plan to effectively remove those facilities to address, you know, risks that are out there with those facilities.
So those were kind of like
Speaker 5
then the third question, I'm sorry. I I forgot what the third question was, but I know it was general to all panels. If you could restate that?
Speaker 13
Sure. Sorry about that. So trying to understand following up on doctor Mitchell's question, if after a catastrophic wildfire occurs and you are getting the lines back up, Are you installing covered conductors right away? Or are you just installing regular conductors and then later going back and identifying whether a covered conductor is appropriate?
Speaker 5
Oh, yeah. No. Went through this last year when we had the lightning induced wildfires in a significant amount of facilities for damage. And we have a fire rebuild standard that we deploy. So it's provided to our teams.
And in fact, when we, this past year, we set up a dedicated branch to support the team providing restoration on the rebuild of the facility. Because some of those times we've already prepared job packages that align with kind of our system hardening plans. And in a lot of ways, we provide that to help the field move forward. But to answer your question, in the high fire thirty years that have been damaged by actual fires, we're going back with our fire rebuild standard, which would call for covered conductor, more resilient poles, the change out of the various equipments to make them fire safe. So the answer is yes for PG and E.
Speaker 3
Thank you. For Southern California Edison and similar, our standard in high fire now is to use cover conductor. So after a fire event when we're rebuilding, we're using cover conductor. There may be some scenarios where there's like a small event or outage that happens and to get customers back up or you know, replacing with bare wire in a small scenario, but the standard is to go with cover conductor. So for large rebuilds or after fire, that's what we're doing.
Speaker 6
For SG and E, this is Jonathan, we would be similar, where if it was depending on the scale of the fire, if it was extensive enough where it burned down much of the structures and much of the wires and equipment with it, the full rebuild would take place in the new standard using power connector. If it was a much smaller impact where the poles, the steel poles, know it was already hardened previously, And then we might put back the conductor just because the loading may not take the cover connector, However, we would also consider how long the customers have been out, So the priority would need to be balanced with the fact that customers need to have power back on. And then we would come back and revisit the area to do the covered conductor down the road. So depending on how fast service had to be restored and how much obsessive the damage was.
Speaker 4
Okay, thanks so much. Going into some of the chat questions, this one's from Jacqueline Eyre with the Ashton Town Council. So this is directed primarily to SCE to begin with, dealing with the 41, 46 mile per hour threshold for wind blown debris. Apparently, for a covered conductor circuits, you're still using that threshold and not taking advantage of the benefits. And so I wanted clarification as to why that threshold is not being raised since covered conductors should protect against wind blown debris.
And then moving on to PG and E and SDG and E, just wanted to kind of lead into the question of how are thresholds for PSPS events changing due to the installation of covered conductor?
Speaker 3
Yeah, thank you, Andy, and thank you, Jacqueline, for the question. So I think the section in WMP that's referenced says that we consider national service wind advisory levels, which is the 3146. However, when we deploy a cover conductor, the threshold we use is the national service high wind advisory level, which is forty, fifty eight. And so that does get bumped up. Of course, there's other considerations when we set the thresholds, based on that particular circuit in question and the factors around that particular circuit, but generally speaking, it moves up to the forty, fifty eight level.
Speaker 15
Hey, Russell, before we move on, Andy, to the other questions, I think there was one other lead in question to Jacqueline's, which was, I believe in your presentation, you had indicated a SCE microgrid project. And I believe the question was where that project was located.
Speaker 3
Yeah. I'm not a liberty to share exactly where it's located, but I can say it's it's in tier three. It includes 189 residential customers, 26 low income customers and 16 non residential customers. We're not ready to share because we're right in the middle of the RFP and we're about to make the final award. So once we do that, I think we can be more public about where it's located, but it is in a tier three area, an area that has been impacted by PSPS.
When we looked at the microgrid location, we looked at PSPS history, we looked at circuit configuration, planned wildfire mitigation work, customer profiles in the area. So all of that factored in costs, all of that factored into kind of the location that we wanted to try and pilot this. And like I said, soon we should be able to kind of go public with location, but we're right in the middle of that RFP.
Speaker 4
Yeah. Thanks so much for catching that, Coco. I guess going into PG and E and SDG and E, curious how thresholds would be changing due to the implementation of cover conductor for PSPS purposes?
Speaker 6
This is SDG and E, Jonathan. So just wanted to highlight the fact that, you know, we're still early in our experience with the cover conductor, and but we do expect that our thresholds would be higher as we gain more experience with the covered conductor and through PSPS events see how it performs as well as you know depending on the PSPS event and area and other considerations, right, with, you know, what may happen, what may come up with, you know, during PSPS events, we just want to, you know, make sure we learn and then apply the learnings carefully. So, yeah, but we do see probably an improved, you know, TSPS impact.
Speaker 5
Yeah. And this is Mark for for PG and E. Very similar. We're in our early stages of seeing how the covered conductor performs. We all recognize that directionally it will raise the threshold.
And, you know, have a lot of our models and information that we've compiled over time. We have our outage producing winds threshold, but it was based off non covered conductor. And so we're trying to understand how that'll change, which we believe it'll directionally go higher as they perform. Some of the learnings that we did get through this past year, particularly when we had the major winds events, what we did notice here is our facilities that had the covered conductor versus the facilities that didn't, We definitely saw that, you know, the ability to restore the ones that had covered conductor and the damage that the hazards and damage that were inflicted upon those were significantly less than what we saw for other areas that were not did not have the system hardening. So, you know, it's just more qualitative, and we're working towards quantifying that to be able to build that in our risk models, as well as how we incorporate those thresholds into our PSPS decisions.
Speaker 4
Great. Thanks so much. Touka, are there any other questions in the chat to
Speaker 15
is one specific question for PG and E that I think would be hopefully a quick response. It comes from Marty Kurdevich of the Safety Policy Division. And the question is, how does PG and E's WMP relate to its current FERC proceeding and the proposed wildfire project and PG and E's application in that proceeding?
Speaker 5
So I believe this dealing with the TO rate case. If it's dealing with the TO rate case, I'd say that the investments we have in our WMP are captured in our going forward TO rate case. Not sure if that answers the question.
Speaker 3
There's another question in there for SCE that I can address quickly is about the vertical switches, because I had them on the slide in both ongoing and new. I think it was a little confusing. The reason we had them in ongoing is we actually initiated the inspection efforts for vertical switches back in 2019. And then in 2020, that's when we were kind of selecting which scope we wanted to do in 2021. So ongoing and that we've been assessing them and kind of planning to do the work new and that we haven't actually replaced any until 2021.
So sorry for that confusion.
Speaker 4
Yeah. Thanks so much for covering that quickly. So going into more of those wildlife safety division questions, this is again, mostly directed at PG and E, but was hoping to get sort of a brief explanation as to the status of any pilots on emerging technology that are being utilized, and what PG and E is moving forward to in that regard. And then also had a question for SDG and E about, the lack of continuous monitoring sensors being used, and wanted more information as to why that sort of technology wasn't being utilized at this time.
Speaker 5
Okay. I'll start off for PG and E. As far as emerging technologies, we have a few. The one more notable one we're working towards is the Refcol emerging technology. I think you heard Edison is also exploring that.
We have it deployed. We're at one station feeding a few distribution circuits in our Calistoga area. So we have the ground fault neutralizer installed, as well as the suppression coil. And it's interesting, in SE, they also had the isolation transformers as well. And we have that all kind of wrapped in one pilot.
We had those isolation transformers installed on the distribution circuits. The kind of status where it is right now, the equipment's all been installed. We are doing our commissioning and
Speaker 5
Yeah. I'm not sure what's going on. Yeah. Sorry for the bandwidth issues here. So we're in the middle of commissioning the rep call system.
And, hopefully, pretty soon, we're gonna be able to stress test it with applying, you know, test faults onto the system. And then we can see how the devices are responding to get better learning. So that's moving along pretty well. We're hoping to have that all up and running, and have a final report later on this year, but really start to get results, you know, this first half of the year. And then the other pilot that we are engaging in is the DTS FAST pilot that we've listed in our WMP.
And effectively, that pilot is really trying to catch objects that are flying into the lines, or even if the line is to fail, trying to catch it before it actually hits the ground or before the object hits the lines. So it's working off of laser technology to kind of see if it can catch those concerns, and then quickly de energize the line. So we've we're in early stages of doing engineering and design. We have a pilot location that we've tested out in a non wildfire area, and now we're moving it out to an area that's in our higher wildfire areas, and we're piloting a transmission and distribution circuit. I think for this year, we're looking to get maybe more of the engineering and planning done this year, and really have those deployed for next year.
So that's kind of an emerging technology that we're really focusing in on.
Speaker 6
Thanks, Sandy. For SDG and E, I wanted to highlight the fact that we are continuing to evaluate these emerging technologies and some of the continuous monitoring technologies that are available because there are various stages of development and we want to make sure that we understand what, the technologists can do you know, allow climate change. But we have the following conductor protection and other protection equipment that we use to identify and monitor. And this is continuous monitoring of our electric system and the attributes and characteristics on our electric system, like the voltage and current and faults that we would monitor and trip accordingly. And especially with the quality of electric detection, being able to de energize the power line before it falls to the ground is something that we're leveraging.
And we're also trying to make sure that we wait for our high speed communications initiative, our project distribution, communication, reliability initiative, our improvement project that will have a wide area network that will help us or a wide area, high speed network that will help us to put on more devices for higher throughput, but also higher speed in communications and responding in the field. So it's important to have that platform and being able to leverage technology. And then we have the cameras that we have for situational awareness as well as weather stations that are providing us that continuous monitoring. So we feel like within the territory that we have and the ability to collaborate with our community safety partners, with our different fire agencies, we have that collaborative effort that can keep an eye on our system, on fires and missions and internal resources as well as community resources that we can rely on. But also the continuous monitoring I mentioned that we have from our system protection system devices, as well as the sectionalizing switches that are smart devices out there at the various key locations, mid circuit and at the substations that can help us with the monitoring.
Speaker 4
Got it. Thanks so much. Circling back to Tern, either Katie Marsoni or Marcel Howager, if you wanna ask another question.
Speaker 8
This is Katie for Turn. I I just I was hoping that Mark could walk us through, there was a two different risk charts showing risk curves on slide four of the PG and E presentation. And from my perspective looking at it, it was showing basically that the risk model previously used and the risk model used
Speaker 8
WMP prioritized completely different circuits for mitigation. And I was hoping that Mark could talk a little more about that divergence between the two models and kind of how that impacted their thinking, this WMP filing, you know, given that maybe it was undermining the previous findings or risk findings from previous WMPs?
Speaker 5
Yeah. Well, thank you for the question. And definitely, when you it is that is the what you what you've identified is is actually, you know, it's you're spot on. When you look at a couple of things here to point out again, when you look at the, one, the shape of the risk model versus the, you know, the shape is different. I mean, they look the same, but, you know, the risk model in what we're calling the 2018 model, the top curve, is a lot sharper where it shows, you know, that a fair amount of the risk is only on kind of the top portions of circuits or circuit protection zones.
And then when you look at the shape on the bottom, it's not as steep. It's still steep, but not as steep as before. We're not that same low.
Speaker 5
I I can hear you. I can hear the the folks loud and clear. I guess it might be on my end where it's, creating the, the bandwidth issue. But, definitely, this this, change has, was definitely something that caught, a lot of our team. We we we didn't wanna take this lightly.
We took a closer look at all these. And as you can see, the colors have been shifted, meaning, you know, the items that were blue, the top risk miles based on these enhancements on our risk model have shown that, you know, they're really we're not really dealing with the highest risk. And that wasn't something that we wanted to go to take lightly. So we did a lot of work to pivot our portfolio prior to the filing to really, you know, be able to deal with the highest risk items. Because then what we're really focused on is trying to limit, mitigate the highest risk and mitigate, you know, catastrophic wildfires.
So we had to do a lot of work to pivot that over to our portfolio. And that actually was some of the challenges that we had, which is why we have a lower production output that we're aiming for this year. I think we prior year, we had over 300. We had to tamp that down to 180 miles. But our plan is with this enhanced risk model is that we'd be able to build that pipeline to get it in for the future years.
But we believe this model is much more of an improvement factoring in a lot of the different enhancements that we're learning from our equipment and in feedback from other models and benchmark with other folks that we believe we're capturing the risk much more appropriate than we were before in our earlier model.
Speaker 8
Is it the machine learning that caused the big shift? I mean, I guess I'm trying to understand why we're seeing such a change now.
Speaker 5
Yeah, I'd say, and I know we unfortunately, I wasn't part of the presentation from yesterday's discussion. Know that Paul McGregor and team, they kind of went through a lot of the details of the risk model there. But I would say one of the big changes there was on the fuel burnings on where we were expecting the biggest risks. And so this had more of a much more the revised model factoring in a lot more of the input and the modeling is focused more on where the fuels are more likely to burn and kind of create this latter effect and really cascading to other fuels. So I'd say that was probably one of the bigger differences that really shifted our risk model.
Speaker 8
I have one other quick follow-up on the PG and E, and this is just a clarification. On slide six, you referred to does, in flight authorize the in flight projects being authorized, does that mean that it's regulatory? Like, the CPUC has authorized the project, or or is it authorization by the WFRG?
Speaker 5
Yeah, it was more authorization with our Wildfire Risk Governance Committee. And typically, maybe how that could also be said is that we've had authorization, and so our teams that actually build out the pipeline for system hardening projects, they have it in their hands. And they're able to work on those, they've gone through the engineering, or they're going through the engineering design, execution, the coordination process to make it happen, to get the projects built in the field. So we call it those are in flight, that they're ongoing and they're moving forward. And those have like the highest likelihood of making their timeline in 2021 this year because they've cleared some of those early stages where the new projects were really starting early on.
We're in those early phases. Feeding that pipeline of projects through our scoping, engineering, design, kind of like the permitting review processes. They're just starting that. And so we know that it's going to take time to build the inertia to get those projects in there. So we actually have a higher volume than the 99 miles in there, but we've discounted some of those miles because of just based on our understanding of how long the process can take.
So then what you're seeing here is 180 miles, which reflects the miles were discounted for the primarily under the category being new projects because the likelihood of those all getting built without having some risk that'll slow it down, we felt that would be kind of over too optimistic. So what we're showing here is one that reflects kind of the discounted miles of what will go through that pipeline.
Speaker 4
So we're out of time. I do have one quick clarification question involved with the in flight projects, and then we'll take off for a lunch break. Do the in flight projects consist of the original risk ranking, the top 100 CPC, or is it based off of the new risk ranking?
Speaker 5
So the in flight projects were some of those projects that were in between. And they also were also projects that were maybe previously identified in other models that were we've had a fair amount of engineering that or work that was done. So when the new risk model came in, it didn't take much to kind of package them up and move them forward. So it was a lot of work that was already done as part of the previous model, as well as other programs that had worked that maybe were paused and we picked them up to advance them over as the new risk model showed that they had a higher priority.
Speaker 4
Got it. Thanks so much to all the panelists for questions and to the utilities for the responses. For all the remaining questions, unfortunately, we won't be able to get to it now, but we suggest sending it and encourage sending data request to utilities so you could get your responses in. And with that, we'll meet back up at 01:15, and hope everyone has a great lunch. Thanks.
Speaker 16
February sunshine, one more round to go, and we're through what hopefully has been a productive and beneficial two days. My name is Kevin Miller, I'm an analyst with the Wildfire Safety Division of the CPUC. Our remaining topic area today is public safety power shutoffs, reducing the scale, scope, and frequency, which is an important one as demonstrated by how PSPS is interwoven through previous discussions. Because they touch so many people, there are many aspects to consider about de energizing and the issue is certainly a primary concern in the minds of many Californians. Last year discussions of PSPS were somewhat distributed through the WMPs that were submitted.
They included intent to continue implementing without fully discussing how initiatives could mitigate and need to de energize. They counted PSPS itself as a wildfire mitigation tool without fully focusing on PSPS consequences to vulnerable and broader communities, as well as response partners. In the 2021, WMPS utilities were asked their directional vision for PSPS, their protocols, projected changes to PSPS impact, engaging vulnerable communities, and PSPS specific metrics. We know that PSPS has evolved significantly since 2019 and even 2020, in terms of decision to de energize, implement notification practices, understanding of implications, and acknowledgement of PSPS as a last resort. This has come about through a combination of use, additional rulemaking, and lessons learned from utility, regulatory, and customers perspective.
In today's PSPS discussion, we're hoping to narrow in a little, focusing in on how utilities intend to diminish usage and consequences of PSPS centered around three areas, potentially buckets of how each utility is improving its accuracy to predict PSPS events, what utilities are doing to reduce scale, scope, and frequency of individual events, and how WMP initiatives are driving toward eliminating PSPS as a mitigation tactic. So with that, we have from Pacific Gas and Electric, Aaron Johnson today, from Southern California Edison, Eric Takayesu, and from SDG and E, Brian D'Agostino and Jonathan Voldemariam. And we'll go in the same order we've been going. Do not surprise any of our panelists, PG and E, SCE, and SDG and E. So with that, I'll hand it off to Aaron from PG and E.
Speaker 17
Thanks, Kevin. So this is Aaron Johnson from PG and E. Thank you for having me. I'm a vice president in our electric operations organization and I have responsibility for the PSPS program in 2021. I've been involved in this program for the last three years from when we first stood it up in 2018 and have worked every PSPS event we have.
Very familiar with details of this program and look forward to presenting some of our improvements this last year and where we look like we're going forward. So next slide, please. PSPS is a tool of last resort for us to reduce the risk of major wildfire fires during severe weather. We recognize that losing power disrupts lives, especially for our customers who rely on electricity for critical life saving equipment. So this is why we've been continuing to focus on learning, listening, hearing feedback on the program, learning how we can improve and really more narrowly target that risk and ultimately reducing the impacts on customers of these events.
So in my presentation today, I thought I would cover plan to cover four areas, just a general overview of our program outcomes in 2020, how our plan is evolving in 2021, the feedback from customers and communities that are shaping our go forward plans, and some of the key areas of focus for 2021 in this program. I will say that I will cover a fair bit of material that goes into some additional detail beyond what is in the wildfire mitigation plan As, you know, there are continues to be a fair bit of activity in the company and not all of that is all of the detail of that is outlined in the wildfire mitigation plan. So know that this is obviously the subject of other regulatory proceedings where there is additional guidance and detail being provided. So next slide, please. So starting with 2020, we had an event series that was an event season that was significantly reduced from last year.
We know based on a lot of the feedback that we've received this winter as we've gone out and asked for feedback that the program performance was significantly enhanced from last year. That said, we understand that we're still receiving a fair amount of constructive feedback, and we know that we have a fair bit of ways to go on this program in terms of improvements. So we had six events in 2020 and one already in 2021. You could say, I guess, that that is really part of the twenty twenty wildfire season from our perspective. Compared to the events in 2019, these events impacted far fewer customers, and our average restoration times were significantly lower.
We really think some of the key drivers of this were our ability to really narrow that risk to the areas of the greatest fire danger. And it is that limitation in the size and scope of the event that really helped us shorten the duration of those events as well as we were able to deploy resources to really restore power more quickly. As you can see on this slide, even with that narrowed tailoring of the events, we did end up with a significant number of hazards still that were found as we did our patrols post event, any of which of course could have led to a very large potentially catastrophic wildfire given the weather conditions on the ground during these events. Next slide please. Pause for one second.
I just need to I'm having a little trouble seeing on my screen, so I'm going to call this up on secondary screen for myself. Bear with me one second. Thank you.
Speaker 17
So this slide provides a little bit of an insight into where we made progress in 2020 in reducing the scale, the size, if you will, of a PSPS event. So when we run these events back through the models that we were using and the infrastructure we had in place in 2019, we see that these events were reduced in size by approximately 55%. Our goal had been to reduce the size of events by approximately onethree. And we were able to significantly surpass that goal. Some of the key drivers, as you can see, were enhanced meteorological guidance and our ability to scope and segment the transmission system and really narrow the risk that's identified there.
And the third biggest contributor was being able to provide impactful item there was being able to island the Humboldt Bay Power Plant station that it was able to provide almost 60,000 customers, over 60,000 customers in Humboldt County with power, almost that entire county that had been put out of power. So some engineering solutions were allowed us to operate that power plant as an island up there, and that was the big driver of that. Obviously, we had a series of infrastructure related investments that we made going in to the season that were expected to drive a lot of that one third reduction. And those were some of the items like the substation microgrids, the segmentation devices that help divide up the grid into smaller pieces, switching mid feeder microgrids, a series of initiatives around infrastructure. Those improvements that we expected from those were really eclipsed by those changes in meteorology and transmission.
And it really was that narrowing of risk that made some of those devices, rendered them less impactful in the end than some of the meteorological and transmission scoping for us. It's not called out on this slide, but I will say we were also able to reduce our average restoration time by 40% from seventeen hours to ten hours. Each of these areas that's identified on this sheet, there are initiatives around these going forward to continue to seek improvements in 2021 in each of these areas, to continue sort of a cycle of continuous improvement, if you will, to add additional capability in each of these areas to further sort of limit the impact of PSPS in 2021. If we could go to the next slide, please. So our PSPS program has really evolved over the last three years.
This slide depicts sort of some of the key highlights from each season and what we've learned from that. As we saw in 2018, we think the scope was too limited for the program. As we expanded it in 2019 to include transmission lines in all of the high fire threat district areas, we failed to scale up our notifications, the website, the customer support, and our customers paid the price for that. Those events were far too impactful. So in 2020, we've improved on, you know, there were over 45 internal initiatives as a company and we've been able to improve, you know, most all elements of the program.
The size and severity of these events going forward will really be driven, continue to be driven by the severity of the weather. And as we've seen, certainly in 2020, that wildfire risk continues to grow in California, the lengthening of the fire season, those fire conditions will really ultimately dictate how we're able to limit this program going forward. When we look out of 2020, we think there's still continues to be an opportunity to make sure that this program is capturing all of that catastrophic wildfire risk. So as we look for some of those improvements in 2021, I'd really call your attention to that first bullet where we're trying to improve distribution scoping analysis to further incorporate tree overstrike potential. One of the reasons we were able to be successful this year in limiting the transmission scope is we got much more granular in assessing transmission risk.
We had things like tree density rather than we had individual health of assets looking at individual structures on our transmission system. And we were able to look at things like the slope of those trees and where they are in relation to our transmission assets. And so that gave us a lot more granularity and we were really able to narrow and target that risk to the transmission lines that were most in danger of potentially being a source of ignition for a catastrophic wildfire risk. The challenge that we are undertaking right now is to look at in 2021, how would we add similar levels of granularity to the distribution system? Right now, certainly have tree vegetation as it contributes to outages factored in.
We have the density and the fuel, the moisture content of vegetation in the area. But we haven't gotten it down to a structure by structure basis. So there's approximately 50,000 structures for us in the transmission level. There's about 700,000 on the distribution side in the high fire threat area. And so the intent is to get a lot more granular on the distribution system going forward.
And so that's been a huge focus and look for us. We have not identified a target for a reduction in either the scope or the size of the events or for the duration, I should say, or the size of the events this last year. That is largely driven by this analysis that we are undergoing. I must acknowledge that our federal probation officer, Judge Alsup, has issued a number of queries to us. And we are in a discussion there about adding potential scope to this program.
And I know a number of agencies are weighing in on that topic as well. And other folks are weighing in there. And so that is a very dynamic situation. And the outcome of that will really and some of the modeling that we're working on right now will help us understand how to incorporate that tree overstrike potential into PSPS and what the implications are for the size and the duration of events going forward based on that. So that's why you don't see a specific target for us.
That said, all the various initiatives that were launched last year to really push for both smaller and shorter PSPS events continue. And those are a lot of the same initiatives you saw on the previous slide where we will continue to add sectionalizing devices. We will continue our microgrid work. We will continue additional islanding opportunities and switching opportunities, as well as improvements to both our transmission scoping and our meteorological guidance this year. Our feedback that we've received has been we've been on a process of gathering up feedback as we got to the end of PSPS season.
I would highlight sort of three areas. There's feedback that we've gotten from our customers, from our agency partners, and then a variety of other sources, not the least of which is our own staff that work these events and provide us feedback about what they need to be successful. Three areas that I'd really highlight. In the customer area, we continue to hear the ask from our customer base, especially our most vulnerable, that they need additional help navigating these events and support among our most medically vulnerable. Among our agencies, while they gave us a much constructive and positive feedback about our improved information sharing and tools for doing that and quality of those tools themselves.
There is still opportunity to improve information sharing timeliness and accuracy, consistency, and additional refinement available on those tools. That will be a big area of focus. And we continue to hear just a general desire for that consistency of information across all platforms. As we're sharing information in different channels, Can it stay a little more aligned during events? So I'll finish now on the last slide, which is some areas of focus for us.
There's a number of initiatives highlighted on this page. I'd call out some that I think are most critical for us. One focus area is just providing that better information and resources to our customers, especially for those with medical needs. We think we have further opportunity. The battery program that we launched last year should be fully deployed this year, will be fully deployed this year.
So that will help. We continue to partner with cities, counties, tribes, regulatory agencies to drive joint solutions. So the level of local outreach that we pushed very hard for last year and that I was personally involved in driving, we want to continue that level of engagement so that we really have this continuous feedback loop with our local partners that are managing these events for their communities. We really need to continue to enhance our state of the art technology and our predictive wildfire risk models to inform the PSPS program measures and make sure that we are focused on those areas of highest risk. And we can eliminate as much as possible any collateral impacts that come from not having the right level of granularity or focus in those models to really target those most narrow areas of risk.
We have to continue to make improvements to the electric grid so that as those models refine, that we're able to turn those models into a mirroring of that so that we're really isolating the wildfire risk in those areas. And sectionalizing devices in particular really help us to match those increasing granularity of that weather model. And then finally, we have to continue to improve the emergency management training for our staff, refining our operational protocols, and just ensuring consistency across and repeatability across that entire emergency management operation for us during an event. So with that, I'll conclude. Thank you for your time today, and I look forward to answering questions at the appropriate time.
Speaker 16
-Thank you, Mr. Johnson. Now we could turn to if Eric is on. Hopefully I'm not butchering everyone's name too much. Thanks.
Speaker 14
Thank you, Kevin, and good afternoon, everyone. My name is Eric Takayasu, and I'm the Vice President in our Operations Department with responsibility for PSPS readiness. I've been an incident commander for the last few years in multiple events, including some of our largest ones. So I've been involved in all the work, not just getting ready for 2020, but getting ready for twenty twenty-twenty twenty one. I appreciate the opportunity to participate in today's panel and look forward to discussing our strategy for reducing the scope and frequency of PSPS.
So if you can go to the next slide, please. And so just to introduce this topic, reducing the scope of PSPS is a fundamental component of the action plan that we recently filed on February 16. So I'll be discussing the approaches that SCE is taking in this presentation, including a brief review of our 2020 performance and commitments made in our recent WFP filing that preceded our action plan. Next slide, please. So we'll be talking about the improvements that we've found in the WMP, but also leveraging the work from the action plan that we recently filed that is referenced in the WMP for a more complete view.
The concerns highlighted in President Batcher's letter, feedback from our partners and our customers are all part of the plan to not just reduce PSPS, but to mitigate its impacts to our customers, as well as improving our engagement and communication during events. The focus of today's discussion is on reducing PSPS, which means designing, building and operating our system to reduce the need by leveraging both their past experience, as well as making improvements that will ultimately result in a more resilient system for our customers. Next slide, please. So last year's fire season was one of the worst with record dry fuel levels experienced throughout SE service territory during the year from heat spells and lack of precipitation and a series of more significant Santa Ana wind events towards the end of the year. We saw about 70% more red flag days in 2020 versus 2019.
And two thirds of our activations occurred from November through December, noting particular hardships experienced by our customers over the holiday season. And for that same November to December period, we saw four times more circuits being monitored for PSPS due to a number of those being repeat occurrences. So while more circuits and customers were de energized as a result of those weather events, we did see some positive improvements as indicated on the slide. Most notably, when we compare the distribution circuits de energized in 2019, we saw fewer of those same circuits and customers de energized in 2020. And we removed around 25,000 customers in scope for PSPS.
The majority of these improvements came from expanding our segmentation plans and our distribution system. But net overall, we saw more circuits and customers come into scope. 230,000 customers impacted, which also include repeat de energizations. And this highlights the concern around the frequent impacted circuits and communities and represents even more the need to advance our mitigation measures more aggressively as we look to get ahead in advance of this year's peak fire season. It's also worth noting that the vast majority of our de energizations occur on the distribution system versus our bulk and sub transmission levels that are 66,000 volts and above.
Hence our strategy to reduce and the mitigations that are part of our action plan that I'll be covering in more detail are specific to the local distribution system that serve our customers directly, which represent about 1,100 circuits in our high fire risk area. Next slide, please. So coming off of our experience in the last two years, our initial emphasis to reduce PSPS was through better segmentation of our system primarily. And in 2020, SE developed a segmentation plan for each of the 1,100 circuits that traverse our high fire risk areas, identifying the specific devices that would be remotely operated in the event that de energization becomes necessary. This allowed us to more quickly and efficiently de energize our system using real time information.
And as we develop more operating experience with segmentation, we learned more about our circuits, the characteristics of the geography they cover and how that is overlaid on our segmentation plans with particular focus on those areas that have been de energized repeatedly. So as you compare distributed circuits from one area to another, each one covers a unique topography and each have varying configurations. And so by taking a more micro view of individual circuits and segments, we can compare the physical and operational characteristics that determine the most appropriate solution for the area. And as a result, there's not a single technology or method that will reduce PSPS cost effectively, but rather a combination of possible measures that can be applied. So on the top of this slide, I've listed three principal forms of mitigation for 2021 that we expect to have the largest impact in reducing PSPS.
These include circuit exceptions, followed by targeted circuit automation and the more extensive grid hardening measures. So circuit exceptions refer to localized situations on our circuits where the wildfire risk is significantly less than what is represented by tier or consequence risk models. Examples of this include recent burn scar areas where there is no longer burnable fuel that could ignite a wildfire or where types of developments in the area have changed the fuel loading assumptions. So one of the enhancements we've made to supplement our incident management teams is having permanent operations folks that we call the advanced circuit evaluation team working during events as well as blue sky conditions to gain an increasing knowledge of opportunities for reducing PSPS by identifying these areas and working to reconfigure our circuits to accommodate. So currently there are about 22 circuits where we've identified these opportunities and we're working to build an additional queue for 2021.
Now, in most cases, these reviews involve relatively small sections of an overhead circuit, the physical characteristics of the areas around the section and various options for reconfiguration or even changes to field operations. And where deemed appropriate generally does not require capital investments to deploy. In addition to the benefit of reducing de energization, this may further lessen the need for more costier solutions to mitigate. And implementing these changes require a thorough and ongoing review of our fire science team, our engineers, our operators, and our enterprise risk management folks. The second mitigation is targeted distribution circuit automation.
And as we prepare for 2020, we were largely able to leverage our existing automation devices, remotely operated reclosures and switches located throughout our high fire risk area. However, not all remote devices used for de energization are naturally optimally placed, and our operating experience has identified additional devices that could be automated or even relocated to minimize the number of customers that are impacted. It's also important to have adequate weather station coverage aligned with our ability to de energize in the right locations to ensure that we are best using real time information. So the more targeted automation we're reviewing where additional weather stations are needed to better increase the granularity of our segmentation and better optimize how we de energize and reduce the scope of PSPS by further isolating additional customers from where the problems exist. Unlike circuit exceptions, this may also allow us to reduce the scope of grid hardening if we're able to isolate parts of the system from the areas that are exposed to hazardous conditions that may require PSPS.
And having more options to reconfigure the system with automation enhances our overall reliability, even under normal outage conditions that are not related to wildfire mitigation. Next slide, please. So thoroughly grid hardening is replacing our bare overhead conductor with covered conductor primarily. I know this has been talked about throughout the session. One of the drivers behind our wind thresholds is the need to mitigate against the contact of foreign objects, which isn't just trees and vegetation, but also includes man made debris such as metallic balloons or other things that can cause a risk of ignition.
So by installing covered conductor, this will allow us to raise wind thresholds given the significant removal of one of the leading causes of ignitions. And by raising wind thresholds, the frequency of PSPS will be reduced. So we're still relatively early in the deployment of covered conductor, and we have used our wildfire risk models to prioritize the locations given the number of circuits we have in our high fire risk areas. But given our recent PSPS experience, we recognize that those circuits impacted haven't fully coincided with our models. So we are utilizing in our action plan additional criteria for accelerating cover conductor.
The goal is there to reduce wildfire risk, but also to make sure that we have the biggest impact in the shortest amount of time in our heaviest hit communities. So while there are other mitigation measures such as undergrounding and microgrids that we also consider and will continue to evaluate, we expect that those would only enhance our plans to implement the three options presented today and to reduce the need for de energization with similar benefits. So for the circuits we are reviewing, we evaluate each of these three options between exceptions, automation and hardening to arrive at what might be the best and most feasible way to reduce PSPS and to provide significant benefits this year. Now, is also an iterative process that we can perform each year, providing a pathway for continuing to reduce PSPS more broadly beyond 2021. And by also recognizing that in addition to reducing the need for PSPS and working on our most impacted circuits, even the process of activating and notifying our customers of a forecasted event, whether results in de energization or not, is something we need to better address given the disruption in everyday life.
So if you go to the next slide. So we're working to improve our situational awareness, and this means improving the accuracy of our weather forecasts as well as what happens real time. So today we forecast PS events on a circuit by circuit basis. We have a two kilometer resolution in our forecast, which allows us to connect to specific circuits. We also strive to have at least two weather stations per circuit to use the best real time weather data as a basis for the decisions to de energize.
However, because we forecast for activation and make decisions for de energization using data from our weather stations, this often leads to notifying more customers of a potential de energization than experience. So while we believe it's better to notify and not de energize than de energize when it's not substantiated by actual conditions, we recognize and acknowledge the issues that arise as customers need to prepare for something that may not actually happen. So as part of our action plan, we're also working on additional data and enhanced modeling to improve the granularity to one kilometer resolution. We believe that this should result in a more accurate forecast on our circuits. And in addition, we're developing a machine learning model for each weather station based off of historical data.
Now we can only do this for stations that have at least six months of information. So we're going to build models for 50 weather station locations, starting with the highly impacted circuits and work forecasts in our experience have been the least accurate. And then we'll begin later this summer and to next year developing models for all of our weather stations, representing some longer term efforts to train and retrain our models over time and as each station accumulates more data to produce a more accurate result. So we expect that these efforts will continually close the gap between customers who need to be notified of a potential event and those that actually experience de energization. We also intend to enhance, to leverage advancements and real time fire spread modeling as another factor in the decision making process for de energization.
While there is more refinement and verification that's needed, we do expect that using fire spread consequences is a factor in de energization decisions by November year. So lastly, we also intend to install about three seventy five weather stations to supplement our automation to provide better situational awareness for de energization decisions. In addition, we're also working to improve the weather information by our first responders using portable devices. It's worth noting that when you take reads from the ground versus the air at the height of our conductors, we've often seen about a 10 mile an hour difference on average. So hence I thought it would be interesting to mention that we started to get some early success in the next generation of low cost mobile weather stations modded on vehicles to supplement the analysis, measuring and transmitting data to our emergency operations centre, and to provide data to our incident management teams where we have gaps in weather station coverage or need additional validation and verification for what's actually happening on the ground.
Next slide. So based on the strategies I've outlined in this presentation, we cite in the WMP these reductions in scope, frequency and duration as shown in this chart. However, as part of our action plan, we're also further accelerating additional work on frequently most impacted circuits. So we expect to have incremental benefits more than what is shown here, which we're still in the process of determining. So in addition to the reductions identified from our exception process, increasing thresholds due to cover conductor and any additional benefits from providing backup generation.
The action plan outlines accelerating work on frequently impacted circuits that have experienced four or more de energization since 2019. So this amounts to about seventy two of the two twenty circuits that have experienced at least one de energization. And more importantly, having a hardened system that will reduce PSPS is expected to be a more reliable and resilient system overall, including weather events from more traditional storm conditions throughout the year. This should also help our customers in the face of more extreme weather threats and to ensure that we're continuing to provide the reliable service that our customers expect. So lastly, if you go to the next slide, I wanna close by providing some links to PSPS information and resources on sc.com, and and I certainly look forward to answering any questions.
Thank you for your time. And Kevin, back over to you.
Speaker 16
Thank you, Mr. Takneisou. Next, we'll go to SDG and E, Brian D'Agostino and Jonathan Valderramarium.
Speaker 18
Yeah, great. Thank you very much, Kevin, and good afternoon, everybody. I appreciate the opportunity to be here to really talk about public safety power shutoff and reducing the scale, scope, and frequency. As Kevin mentioned, I'm joined here today with Jonathan Muldemeriam, our director of wildfire mitigation. And I'm meteorologist Brian Bagastino.
I'm the director of fire science and climate adaptation. And I joined the SDG and E team back in 2009 as the first meteorologist to really start bringing the situational awareness and the science into our original shutoff program. So I'm here to really talk a bit about, you know, what we experienced in 2020, because it was a it was a very eventful year across Southern California. And then we'll talk about ways that we did mitigate and reduce the impact this year. And then we will get into ways we're gonna continue to do that moving forward.
As we head into the first slide, we are starting with some weather data. When the first steps that we took back in 02/2010 was we started putting weather stations on all of our high risk circuits, a practice which is now spread across the state, even across the country in many cases. But what we have seen over the last decade is we're seeing an increase in occurrences of these high wind events. So I'll point out just below the text there, Hellhole Canyon, it is a wind prone area in Valley Center. The graph there is actually showing peak wind speed every day back to 2013.
So the last eight fire seasons. And that red line is our trend line in terms of the wind speeds that we've been seeing in these areas prone to Santa Ana winds. So it is in fact that the data is showing more events, stronger events impacting the backcountry. And this has been especially since 2017, where we have seen a pretty sharp increase. And it's not just across this one community.
As we, look at the graphs on the right, whether we're looking at our foothill communities in Alpine, or we go all the way up to the Cuyamaca Mountains, wide scale across the SDG and E network, we are seeing this increase in wind, right, which does impact the implementation of public safety power shutoff. And the next slide, what we're gonna get into is, especially in 2020, and we heard this Eric, was that we had a very late season impact. And this is unusual. We've had a program in place going back now, as I mentioned, to 2009, And out of the last twelve years, this is only the second instance where we've surpassed Thanksgiving and kind of moved through that Thanksgiving to Christmas point, and we haven't seen the rainfall that really starts bringing down that fire potential. So now this is two out of the last four years, and it's also a clear trend, which is consistent with the climate science that we are seeing, that less reliable late season rainfall across the region.
So but I did wanna highlight this period that you're looking at here between Thanksgiving and Christmas. Alright? So you'll notice the graph or the bar chart up at the top. It starts on Thursday 11/26, that is Thanksgiving, and it ends on December 24. The red lines across the top are when we had red flag warnings issued by the National Weather Service during that time.
And the bar underneath with the yellow and the red, that's our internal fire potential index that we share with community stakeholders, the National Weather Service. So I wanted to point all this out and go through a few events because the event that started on Thanksgiving, point that I wanted to make there is, you know, we did not have any PSPS customers affected in that event. Right? So that was positive. But part of the reason you'll notice our fire potential index remained elevated, and we take all of these factors in.
So this was an area that the fire potential wasn't quite as high. We were monitoring conditions very closely, and in that case, we were able to avoid public safety power shutoff impacts. The next event was very different for us. It was described as the National Weather Service as extremely critical, as we moved into that event. And to this day, in the twelve years, it is the windiest event that we've seen with the top 20 average weather stations recording 71 miles per hour and a peak wind gust of 94.
So that event came in, and that was the largest and most impactful public safety power shutoff event that we have had. But that being said, we'll hear from Jonathan here more. There were great implementation of our different generator programs. Our microgrids really helped keep core community resources energized in that event. So despite the larger impact, there were there were some benefits that we've seen this year, even our community resource centers extremely popular during that event.
And as we move then into the next event, which was just several days later, that event was not quite as windy. So part of that event, we'll talk more about how we've implemented some of our new high speed weather data technology during that second event to try to decrease the impact. And what that is now is we can set our weather network to give us weather station updates every thirty seconds. And if we start seeing areas that get windy and it triggers an alert for us to start watching it closely, then we start looking at it in thirty second reads. And a lot of times, it will start to trend down, and we found that additional information and that additional situational awareness really helped us save some customer impacts during that event.
And then the the very, impactful, event just because the timing coming in right at Christmas, on the December 23, yet another red flag impacted our region. But you'll notice again, looking at the the bar chart up at the top, we didn't have that extreme FPI. We were in an elevated buyer potential, but lacking those extreme FPI, it really did help us start to minimize and be much more targeted and leverage those weather stations and leverage the sectionalizing devices to really start to bring the impact down during that extremely critical time around the holidays. But overall, as as I've mentioned already, a very eventful year for us with a total of 11 red flag warnings and public safety power shut off events for the region. But out of those 11 red flag warnings, I'd like to go to the the next slide.
Because out of the 11, there were five of those red flags that escalated to actually implementing a public safety power shut off event. And a couple of those, early season events, September, and October, we were really able to implement our weather station and our sectionalizing devices. On September, we kept it down to 49 customers. Really being able to just leverage everything we've learned over the last decade to isolate and keep those numbers small. And then we did talk about the damaging wind event, widespread wind event that we had December.
That, again, that was our largest damage that we found across the system with over a dozen different instances where we had impacts to the system. And then we've already talked about December. But the other thing that I wanted to point out is and I I think we've done a much better job this year on really keeping that outage duration as low as possible. And there's a few different ways that we're able to do this. One of them is our additional weather stations.
And the more that we segment the system, it doesn't only help us minimize the impact of PSPS, but in some of these larger events, as soon as the winds begin to come down, we can find out right away that the winds have decreased and we start patrolling earlier, and it helps speed up the restoration as well. So as you see those first two events, we kept the average duration to less than ten hours during those events. So there's been really significant improvement in that area this year, something that we're planning to continue to build on moving forward. With this ongoing impact of PSPS and the trends, again, of longer fire seasons with windier conditions, the next slide starts to get into, you know, how do we work with the customers. Right?
And how do we work to mitigate the impact that public safety power shutoff is having on our communities? We heard yesterday from Mason and Sarah that we're really integrating PSPS risk into the wings model moving forward, and we'll continue to assess it that way and prioritize it that way. But at the same time, we're looking at who are some of these most impacted customers and what can we do right now to help them. So this year we did have thirteen thirty four portable battery powered generators. That went to medical baseline customers this year.
So anybody who anybody who's registered in medical baseline who had experienced a PSPS outage in 2019 qualified for this, and we're expanding this as we move into 2021 because it's been very well received. In terms of the generator purchase rebates, this is a program that is targeting our care customers, those where we can provide some financial assistance and rebates to those who want to go and purchase a generator. So again, that was also a a very successful program this year. The picture you see on the right there, that was the first install of some happy customers who got their whole home generator. And we've gone in using our weather data and we've really targeted those areas that are most prone to these extreme weather conditions.
And those are the places that we're prioritizing first for the whole home generators. And then we're still through our emergency operations center finding ways to implement any community needs. And one example was the Alpine Mental Health Facility during our large event in December. Earlier today, you did hear Jonathan mention the microgrids, a big piece of this is it's not just the hundreds of customers that are being served here, but it's the type of customers. It's very specifically targeting those resources that can serve a community while they're experiencing a PSPS to mitigate the impacts.
So this is an ongoing project which we'll hear more about, and I know you heard about it in detail this morning. So I won't expand too much on that. The next slide gets into the weather technology. So these 30 new weather stations, we really started to target Valley Center. Because as I showed you in that first graph, Valley Center is seeing an upward trend in winds faster than what appears to be the rest of our service territory, and we're studying that.
We're working with universities. We're looking very closely at ways that we can continue to serve that community through new weather stations, new sectionalizing devices, and find ways that we can immediately continue to decrease the impact that PSPS can have on that community. And we do a lot of it through working with their fire agencies, with the fire chief, with their fire safe councils, and find out what can we do to minimize the increasing impact in that region. The thirty second weather data right now, and you'll see it's really giving us the insight and the confidence to keep circuits energized in some cases. Because we may see a quick spike up to 60 miles an hour potentially.
But then when we put on the thirty second reads, we could see that it's dropped off very quickly, and it gives us really good confidence, that we are tracking well, with the data in a particular area. And then we've implemented all of this new intel into public safety power shutoff databases, which are going directly to our utility incident commander. So our ability to process the data and make the best, most informed decisions in real time, taking into account all those factors that feed into public safety power shutoff and making sure that we do have all the best available data is something that was a big improvement in 2020, and we're still expanding upon that in 2021. The last point that I'm I'll touch on before handing it over to Jonathan is the transferring of load. Now, we're continuing to study the impact at every circuit in our service territory.
And in many cases, critical resources in the community may be fed by multiple directions. So we will do analysis coming into an event and in some cases switch over and transfer the load to come in from a lower risk area. And even in some cases, we'll say, well, it looks like both areas could be impacted by PSPS, but we could get this one restored more quickly because it doesn't require a helicopter to patrol, and just as an example. So we can start making switching plans that will, you know, in some cases even decrease the length of a PSPS. And we're finding that this is having a positive impact on the community.
So, again, I appreciate the opportunity to talk about how 2020, a very eventful year, played out and how we were able to decrease the impact of public safety power shutoff. Look forward to the question and the discussion afterwards. With that, I'll hand it off to Jonathan.
Speaker 6
Thank you, Brian. If we go to the next slide, I'd like to start off by including our approach into the 2020 plan. This past year, we really assigned two task force teams that would look at the specific solutions that we could come up with in our internal teams for PSPS impacts. And we leveraged the engineering type team. That was one of the teams that was looking at engineering solutions.
The second team was looking at customer service type solutions that we could provide our customers during PSPS events. And I'll focus on the engineering solutions for this particular slide, but I also wanted to highlight just briefly some of the customer service oriented solutions that were created in 2020 that benefited our customers during TSPS events. Brian mentioned the community resource centers, of the additional activities that were done, but also the additional community resource centers that were added in 2020. Installing some changeable and movable road signs that display PSPS related messages in highly traversed HFTV intersections helped customers stay informed, especially customers that were hard to reach and maybe didn't have access to online tools. The other thing that we were able to do was take advantage of enhanced AM radio spots during PSPS events and pre and prior PSPS events as well to inform customers.
And then how to videos that we initiated and shared with customers so that they could get crucial information on navigating PSPS events. And also a PSPS app that we provided customers that they could install on their mobile phones so that they could receive alerts, right, even if they didn't have an SDG account when a PSTS event was evident and any additional information as well. So all of this to say that our approach was twofold. Customer service related solutions, as well as engineering solutions, more technical in nature, that I'm going to share next here on the slide. Brian already shared the generator programs that we have with the microgrids we highlighted, specifically in Shelter Valley.
This is one example. Sorry, go back to the slide. Yeah, so microgrid solution we already shared. But on the weather stations, besides the thirty second solution, as Brian highlighted, the additional weather stations helped inform the surgical process that we can take for how we de energize circuits. So instead of de energizing larger sections, we were able to minimize the impacts by going to smaller sections.
Additional sectionalizing devices allowed us to break up the circuit even more so that customer impacts were reduced. And then transferring loads of circuits, parts of circuits, to another circuit that would not be impacted by PSD incidents also helped. So in this next slide, we will discuss some of the comprehensive solutions, right. So we take the single incidence that I talked about between all the solutions, the engineering solutions, and that is highlighted on the right there, the undergrounding whole home generator, microgrids, low transfer, sectionalizing, and weather stations. And for the December 2 through the fourth event, where we had over 73,000 customers that were impacted, we were able to reduce the impacts by 7,300 customers.
So the totals for each, for example, by undergrounding, strategically undergrounding areas of PSPS, we were able to save 5,773 customers from being impacted. And so on, right? So with the microgrids, we were able to save three forty eight customers from being impacted. On the right, it's a picture of how we were able to save over 1,700 customers from customer, sorry, PSPS impacts in the December 7 through the ninth event. Now each PSPS event is different, right?
The scope and scale and frequency is different for each event. And so the solution on the right is different from the solution on the left. The mix of solutions is still the same. But how much reduction we get from each solution is different. So now as we go to our next slide, we'll look at a summary of the three events in December.
The chart is showing the customers that were impacted, right? The blue bar chart shows how many customers were impacted for each event. The middle bar shows how many customers were able to be saved from our situational awareness reduction, so our situational awareness tools. Brian shared with you the thirty second tool that allows us to understand how high the wind speeds go, right, on a thirty second interval, but also how long those wind gusts lasted, right? So if they're going to be lasting a lot longer and we anticipate and see that the wind gusts are remaining and sustained, then we're able to de energize the segment of circuit.
Whereas if the thirty second hit of high winds is brief, then we're able to ride through and avoid PSPS impacts. And so with that, you see in the December 2 to the fourth, 5,199 customers were saved from PSPS impacts. The bar chart on the right for the December, the 7,300 customers or so that were saved, I already shared with you with the engineering solutions that were able to save them from the impacts. So overall, 15% reduction for that December event. On the middle area, the slide for the impacts for December are shared.
Over 15,000 customers experienced the PSPS event, but over 22,000 customers were able to avoid impacts to them by using the thirty second technology. Now on the December event, it wasn't the thirty second wind speed capability that allowed us to save the impacts for the 19,000 customers, but it was the fire potential index that Brian talked about, meaning the fact that there was a red flag warning was one condition. But we also had additional tools to understand the situation within our service territory and the impacted areas to understand that the fire potential index wasn't as elevated or extreme, and therefore avoid impacts to our customers. So the numbers would have been that much higher without the fire potential index tool, the thirty second weather tool, as well as the engineering solution that we discussed. So in summary, when we look at the approach that we took in 2020, we focused very specifically on solutions for our areas of impact and the customers that were impacted in previous years to provide solutions.
We combined engineering and customer care solutions, as well as we used our granular data that we're able to get from our situational awareness tools to leverage the capability and use it towards inducing the SPS impacts. So appreciate your attention today. And this last slide is a picture of this next slide is a picture of the 2021 plan and leveraging some of the same engineering solutions as well as customer care solutions that we're going to have towards reducing even more impacts from about 3,000 to 7,500 customers in 2021 based on the fact that we had similar events as 2020. Each event is different, so there may be more or less customers saved from impacts, but this is the overall estimate for 2021. Thank you for your attention and look forward to the questions.
Speaker 16
Thank you. And since we're, I'll take moderator's prerogative here. We're a little bit early, and then I'll let everyone go a little early. But with the FBI, can you just describe for everyone, since we've talked about the weather, what what influences the FBI?
Speaker 18
Yeah. Absolutely. So the the FBI has a a few different factors. One is, of course, the weather. Right?
So we take, wind speeds, and then we take, dew point depression, which is the difference. It's pretty much a measure of humidity in the atmosphere. Then we take live fuel moisture, which we generate with AI models. We take dead fuel moisture, which we develop through partnerships with UCLA and the United States Forest Service. And then, we also have the greenness of the grass, which we incorporate from satellites.
And we continue we combine all of those, relate it back to historical fire occurrence, and we can calculate the probability of large fires developing on any given day, given the the fire weather conditions. This is something we started running operationally in 2012. We started sharing it with stakeholders across San Diego, and it's now being built by utilities across the state, and even up into the Pacific Northwest as well. And a big part of what we will find is we continue to enhance it, right? Like any science based tool, there's always a next step.
So this year, we're working with the San Diego Supercomputing Center and one of their capstone teams, and we're looking closely at additional inputs such as fuel temperature, solar radiation, and then duration of those to understand how those can impact fire potential to just continue to enhance, continue to improve our ability to assess the potential for catastrophic wildfire on the landscape.
Speaker 16
Thank you. I think that helps complete the picture you were painting there. And with that, we'll take a break and have everyone come back at, give you two minutes back that we took earlier today, come back at 02:30. I want to welcome everyone back. Showing 02:30 on my computer.
And I want to thank all the panelists for the presentations provided, which have provided some further insight into some of the questions that we had prepared from WSD, the wildfire safety division, which have now been fully or partially addressed to some degree or a large degree, And I would like to build upon those as well as provide additional opportunity for questions and discussion. So, as we have been doing with the workshops, I will start with a question and then I will turn to the stakeholders that we have identified previously on the line from TURN, from Muncie Gray Road Alliance, Public Advocates, Will Abrams, and Green Power Institute to each take a turn. And then also we have Nathan Bluhn on the line who's moderating the chat with questions from the audience. And as before, we'll ask any of the stakeholders to raise their hand and take a turn at a question and we'll go around. So with that, I had a question that I think has to a large degree been answered, but I wanted to give one more opportunity for each of the utility representatives to talk a little bit about the WMP initiatives that you have described that account for direct impacts on reducing PSPS scope frequency duration.
And if I realize that you have described there are a mix of opportunities there, but if you could underscore which initiatives really you see either now, if not now, but in the future having a large impact or underscoring what that initiative is and how you're measuring and potentially how large that impact would be, maybe starting with PG and E.
Speaker 17
Thanks, Kevin. So this is Aaron Johnson with PG and E. So as we outlined, I think it's harder to provide a sense of what that will be going forward. The two single biggest contributors, which contributed to threefour of the reduction for us this year, were the improved granularity of our weather modeling that accounted for about a little over 50% and about 25% came from tighter scoping of transmission events, which was largely about increased granularity of understanding of the robustness of that system and where there was potential risk. So again, that drove threefour of the reduction that we saw in the size of events.
There are similar initiatives. And then again, the overall outcome was 55 smaller than it would have been under the 2019 state of infrastructure and modeling capability. So I think a couple of thoughts that I just offer on that. Future improvements are highly unlikely to be quite that dramatic, right? So I think it's fairly tempered in our expectations going forward, given the improvements that we saw this year.
But those are the two areas where, again, we would continue to see the largest impact. For us, I think one of the challenges, one of the things, an example I would share and again, that's very hard to predict what that will be until the continued refinement work is completed this year, where there will be refinement in both of those areas. So we will continue to refine the meteorological models. In transmission, while that scoping has largely matured, what we now have the capability with a clearer set of criteria is to go after targeted work on that transmission system to really reduce, take particular segments of transmission lines out of scope based on those clearly established detailed criteria, and to really target work, if you will, at the low hanging fruit where perhaps the removal of a dozen trees can take a line out of scope as opposed to some other lines where maybe 1,000 trees would need to be removed and that might not be realistic. So very targeted work there, and we've identified at least six lines that we think could come out of future scope with those established criteria.
So again, very hard to say what it will look like. One more example I'll give is The sectionalizing devices that we installed last year really provided a ring around Tier one and Tier two to really isolate that Tier one area where we don't expect fire risk, where we couldn't always electrically isolate it. What we saw with the increased granularity of PSPS is that not exactly, but many of the events really were just quite nicely aligned with the Tier three areas. And so our sectionalizing effort this area this year is to install more sectionalizing devices around the Tier two and Tier three boundaries in the electrical system. So again, those ones at the Tier one, Tier two boundaries will be helpful in the event of a very large weather event.
But as we get more and more narrow and targeted, an initiative like that will be we need to respond with our sectionalizing around the areas where we're starting to isolate some of these PSPS events. So that's a little bit of color. Obviously, the increased inclusion of distribution tree overstrike potential makes it impossible at this moment to predict exactly what to offer a projection on what that will be. But as that process works its way through, we intend to be as transparent as we can about what those outcomes look like and what a prediction would look like once that's all settled in terms of the impact on potential scopes of events going forward.
Speaker 16
Thanks. Appreciate that insight into what you're seeing and what you're thinking. Maybe turn to SE.
Speaker 14
Yeah, thanks. So far as far as WMP, there's a table and I had that in my presentation citing the reductions that we expect in scope, frequency and duration. The primary drivers behind that, I covered some of those, that's the exception process, grid hardening. We're also looking at evaluating potential locations where we have underground systems that potentially could be supplied by backup generators, a lot of feasibility analysis that's needed there because there's some siding challenges and other things that we need to get over in order to make those both cost effective as well as feasible. And I would say the second piece of that is in addition to reducing the sort of the physical de energizations and the impacts, it's really about how do we assist our customers.
We also talk a little bit about how to improve or expand our critical care battery backup systems to all medical baseline customers to extend our rebate programs, to help customers deal with PSPS. So I think between the initiatives around the grid, as well as our customer programs is really our focus strategy to assist them PSPS and reducing the impact this year.
Speaker 6
Yes. This is Jonathan Oldmeyer. It is similar in our approach to where we put a table showing our 2021 reductions, just like I shared on the slide in the WMP filing, looking at what mitigations we plan on putting in place for PSPS impact reductions. And those are a combination of the engineering solutions and the customer solutions that we shared. So we as we continue to develop and implement those solutions, we expect to see reduction, efficiency as well as impacts to customers and administrations.
However, that's also balanced with how each event is different and each PSPS event may have different characteristics. What we do is we usually compare it to a previous PSBOs event when we say we're able to reduce impacts. For example, in 2020, we compared our reduction projections to 2019 and said we will be able to reduce by 25% to 30% as compared to the 2019 event. And so we're doing similar for the 2021 comparing it to 2020 by saying that we might be able to reduce 3,000 to 7,500 customers in the past. Going forward, long term, we do share our plan to, as I did earlier, hardening, with more undergrounding capabilities with power production to be able to have more reductions.
Speaker 16
Appreciate it. Thank you. With that, I'm going to turn to the stakeholders, and I'll go in reverse order from previously and start with Mr. Abrams.
Speaker 13
Thank you very much. My questions are around the communication associated with the public safety power shutoffs and sort of definitional terms that are used across the different utilities. So in the wildfire mitigation plans from 2018, there was really a lot of discussion and definitions around reclosers. That evolved over time to a broader definition of sectionalization devices. And the communications around those sectionalization devices as they land in the hands of public agencies, often doesn't provide them with the tools they need to understand the gates and the constraints associated with controlling the power shut offs and where it is affected.
And so I was hoping that you could all sort of talk to A) how we can be more transparent associated with that information, so that the public understands that. And B) whether you feel like it's your responsibility to sort of take the lead on defining that, mapping that. So if you asked an emergency manager within the municipality, they would be able to identify here, here and here are these types of sectionalization devices. And therefore can help you to limit the harm caused by the public safety power shut offs, and obviously maximize the protection that it provides. And similarly around those communications, I want to understand whether each utility identifies that it's their responsibility to communicate around public safety power shut offs in terms of where resiliency centers are going to be ahead of time.
So I know before going into wildfire season, where those are going to be, whether you feel like it's your responsibility to communicate effectively around the use of generators and other things to keep us safe, or is it just your responsibility to provide materials and then sort of left up to the individual whether they actually take that information in going forward. So that's my general question there. Thank you.
Speaker 17
Yeah. Thank you for the questions, Mr. Abrams. So there's a lot of questions inherent in that. So let me speak to a couple of them, and maybe some of my sister utilities will snag other parts of it or can speak to their elements.
One thing to go back in time is the recloser program. So a fire risk mitigation that is fairly mature for us at this point is we added communication devices, which we did not necessarily have in some of our remote locations to SCADA system, to all of our reclosing devices. And so those have all been outfitted now with reclosers. And based on different weather models, a weather operational process that comes out at six hundred every morning, those devices are turned on or off for the day. We used to have to turn those devices off for the season.
We did that for a couple of years while we were going through the automation. So that's largely been complete. It has been beneficial to the narrowing of PSPS to have that automation on those reclosing devices because that has given us, again, greater granularity to isolate PSPS events. So that program ended up spilling over, if you will, into helping us narrow PSPS. There are certainly other devices that help sectionalize the grid that are not reclosers, switches, even some fuses can serve in that any sort of breakpoint in the electrical grid.
So reclosers are a subset of sectionalizing devices, if you will. And some of the sectionalizing we install is more reclosers. Some are switches, some are types of fuses. And it really just depends on the topography of the grid as to which is the right solution in any given location, as determined by our distribution engineers. Hopefully that helps with some of that.
In terms of communicating those locations, we have tried to provide a series of maps to affected communities that are more about what is most likely affected in terms of as we add this sectionalizing, how does that narrow down? And through a series of workshops that we held with, we used local emergency management at largely the county level. We invite tribal government into that. We had a series of meetings with them where we would go into our GIS system and try and show where some of those improvements are so they could help understand which neighborhoods were coming out of scope. We have a series of maps we call the most likely to be affected areas that we post both on our website and for emergency management partners in our PSPS agency portal.
And they show you a sort of, not necessarily the location of all those devices, but the outcome on PSPS in terms of geographic limitation from those improvements. So that's what we do. In terms of communications with stakeholders, we have a series of I guess from a backup generation standpoint, we keep a fleet of generators to support customers. Our messaging very much is that customers need to develop their own emergency management plans. This last year, we were very supportive of hospitals and vote tabulation centers.
And so we did go and provide generation there to ensure that the election was not disrupted and that hospitals were not disrupted within the high fire threat area. And then again, we keep a fleet of backup generators that we will support various entities when those emergency needs come up in the events. We encourage customers to not make that their base plan. They should have their own base plan for resiliency. But we're there to support them.
And what we often find is the odd agency that hasn't planned well, and we support them. Or in many instances, where backup generation fails for entities, and we're able to step in and support them in that. And so that is a program that we have and that we continue to support. And then in of community resource centers and things like that, a big learning for 2019, we built a plan on our own, not closely connected. That was one of those definitely in hindsight moments.
It was not the right way forward. So now going forward, we build those plans for where those locations are with local government to determine where is the best place to situate those. And then in actual events, we activate based on their feedback collaboratively which one of those we're going to activate. And sometimes we'll even add new locations based on feedback. A fairgrounds is not available because there's an event that weekend, so we will find a new event.
But we now do that. We try our best to do that in lockstep with local emergency management so that we're synced up there and there's strong agreement across the utility and local leaders about how and where those should be. So hopefully that covered a number of your questions there, and I'll
Speaker 13
to you Kevin. Just a clarifying question on that. I guess part of what I'm trying to understand, know that's a lot of the activities, but part of what I'm trying to understand is who's accountable so that we have a strategic plan that we can rely on. So as an example, if we were to ask who's responsible for if a mayor doesn't know where there are these sectionalization devices, is that your responsibility or is that their responsibility to go and get that information? Similarly, if customers don't know how to use a generator, or don't know where their closest resiliency center is, or don't know about how to manage during a power shutoff, is that your responsibility that the residents don't know, or is that somebody else's responsibility?
Because I just, I think having accountability in the plan of what you own and what you don't own, and I know that sometimes it's split fiftyfifty, but at least we understand who owns it. If you could just talk to, you know, sort of what PG and E owns in terms of the communications around the PSPS events, please?
Speaker 17
Yeah, so I think we have a very strong responsibility to communicate those locations for those. So we do not consider it. We like to partner with emergency management to help get that information out and to determine those locations. But we feel we have the responsibility to publish that information on our websites and to raise that information, raise that awareness. We certainly will not be successful without the partnership of local government in communicating that same information through their channels.
So while we are accountable for delivering those services as part of our plan, it cannot be done successfully alone by the utility. So I think there is some joint accountability there in terms of communicating that. A lot of those resources and information, we are accountable to put forward for the public's consumption. But again, success will depend on that strong local partnership.
Speaker 14
So this is Eric from Edison. I guess I can go next. So first of all, I do want acknowledge and there's a really great question around sexualization devices because there's a lot of different things that we use. We have reclosers as you mentioned, we have remote control switches that are operated remotely. We have overhead, we have underground, we have devices that could be operated under load.
We have devices that have to be de energized in order for us to actually operate them. So there is some level of confusion, think, or maybe misuse of terms when it comes to being sort of accurate about the capability and where we can actually sectionalize and how. I'll just give one example. We had a situation where we have the ability to sectionalize, but it requires a crew to go out there and physically operate a device that is pad mounted and when we had a larger public safety concern, those crews had to pull off, but that's different than if we send a signal to a device and we could just simply operate it remotely. So I do think it is our responsibility to provide a clarity in how our system operates.
We are looking at increasing the level of transparency in our decision making both with state local governments, as well as our emergency managers. So one of the things we're also embarking on is particularly for highly impacted communities is to share circuit by circuit, what are some of the mitigations that we have planned and what are some of the limitations? So some of that will include, you know, just how we operate the system and what we can do remotely and what actually requires the manual intervention. Building upon what Aaron said, I mean, I think as far as information that we provide, we do own having the need to communicate in event as best information we can, particularly around not just the status, but where our CRCs are located, where our CCBs are located, community crew vehicles, our resource center locations, and we do so twenty four hours in advance. But we also have to realize there are going to be challenges too with we try to locate those where the weather events are happening.
So there are always going to be some level of subject to change kind of things, but it's upon us to do the best that we can. The one thing I think you also mentioned is about generation. So I think where we have places where we would be supplying generation for the underground load block situation that I mentioned, where we might have some locations where we could pre plan ahead of time. Those are probably the examples of where we can show some of that information. I think when it comes to what we would do in an event though, I think working with our County emergency managers on the process associated with what we can and can't do, I think is going to be extremely important.
And lastly, I'll close with our critical infrastructure customers. We need to figure out a way to work really closely where we have significant issues at risk. One of the things that we did leading into the last event of January, recognizing we have both hospitals and COVID facilities that are out there that are critical. We reached out to every single one of them. We asked what their resiliency needs were, if they had additional ones that weren't being met, and if there's a way that we could do some additional sexualization or isolation, we would take all those things into consideration to try to mitigate as many issues as we can.
So that's just an example of where we have a really significant issue, how we could figure out different ways to communicate. So hopefully I answered most of your questions.
Speaker 13
Yeah, that's great, thanks. I do, yeah, I agree with you that there's this sort of conflation of terms and sort of how that's evolved sort of from public consumption is that, you know, we used to call them reclosers and now they're sectionalization devices as if all those are the same thing. And, you know, we've gotten from 50 in our county to 80. And so as if we know what that does. So I do think it would be extremely helpful, as you said, to sort of understand what each of those are, the limitations of that, what it can and cannot do so that emergency managers can contribute and help with that sort of cost benefit analysis of where those gates are put.
So thank you. I appreciate that.
Speaker 6
Just from SDG and E's point of view, for the sectionalizing devices or the communication or general information that our community safety partners and the public need, we do have a very good robust process from our point of view to collaborate with the emergency services or safety partners to understand what is needed for, and then after each event, going back and looking at what can we improve. And we've heard that from Virginia Medicine as well. So with that, though, the most helpful information when it comes to sexualizing or big closing, we think, is to share a map and show what are the impacted areas, potential impact areas, or even during events sharing the updated information as to what communities are going to be impacted rather than pointing out the sexualizing device. That provides a little bit more information. It may require a lot more on us to share how we characterize those communities, but it's been helpful in our case.
As well as back to the resiliency center's question, yes, we do have a responsibility for the community resource centers and identifying and sharing with the community as to where these will be. And we do that not just before investing, but also, in our planning. So as we collaborate with the different safety partners and our wildfire safety fairs and our commission share with the community every year, we try and educate and do outreach programs so that people understand ahead of time, ahead of these events, where these community resource centers are. And what is the assistance that they can have, it be generators, whether it be other assistance when it comes to PSPS events and even some of the tools that I highlight earlier through the information that we have on our websites or through our PSPS app.
Speaker 13
Great. Thanks. Do you think those maps would be included in the WMPs or attached so that we understood, here's where all those devices are and here's where those resource centers are and things of that nature? Would those maps be included in those submittals?
Speaker 6
We have a GIS submittal that we had to have in our WMP so that we laid out all the plans based on what we submitted, right, including the hardening, including the community resource centers and other things that we got, all those features in GIS, Thank you.
Speaker 16
Thank you for that Q and A on that topic. I'd like to turn to Green Power Institute at this point.
Speaker 11
Hi, thanks. So my question is for PG and E, but also for the other utilities. So PG and E provided a plot of PSPS mitigation as a function of mitigation activity. And I was wondering if there's an understanding of why the distribution grid hardening efforts really seem to have minimal ability to reduce customer PSPS impacts. So if that's related to essentially intrinsic to the selected grid hardening mitigations or where they were implemented, Or does it have to do with the amount of those mitigations implemented?
So is there not enough, say, covered conductors to currently make an impact on PSPS mitigations?
Speaker 17
So this is Aaron. I'll jump in on that. I think the answer is yes to all of the things that you highlighted. So let me sort of tick through hardening as how it relates to PSPS. So first of all, hardening until 2021 for PG and E has been primarily driven by larger fire risk prioritization.
General climatological wildfire risk is different than the acute catastrophic large wildfire potential that we're trying to head off with PSPS. And there is some overlap in where those risk models point you, but they are measuring different risks and they are not fully aligned. And the general climatological risk was the source of prioritization for previous hardening efforts. So what you see as an outcome of that is that a circuit segment that was hardened over the last couple of years might not be electrically isolatable. It might be a stretch of circuit, a segment of a circuit in the middle of a larger circuit.
And it can't be traced back to a source of electricity to keep it energized during an event. So that's been one challenge for us. We are beginning, as we prioritize hardening work this year, to incorporate PSPS, that acute wildfire risk, along with the climatological risk. We are doing that and balancing that portfolio in a subject matter informed way this year. We are looking to mature to see if there is a single alignment around risk model that will help us going forward.
The reality is that there are other mitigations that may come into play around PSPS. Can we do a microgrid? Are there other forms of support that we might be able to offer? So we looking to mature our capability to understand on a circuit by circuit basis what the different mitigations are that are possible. So that sort of explains why the hardened circuits that we've done so far don't necessarily show up as a reduction in PSPS.
We do have some projects this year that construction schedules willing may be able to help reduce PSPS impact this year. And those are largely a handful of projects that were started last year because they generally involve some undergrounding. And those are not single year projects in order to get all the permits and work aligned locally. So the other thing that's critical to the removal of hardening is coming up with a standard that helps us understand how we would raise the threshold for a hardened circuit. And let me differentiate when I talk about a hardened circuit from when you harden a circuit, there's sort of three ways you can harden it that we look at.
First, can we remove it? Second, can we underground it? Third, if either of those two are not possible, then we would look at overhead hardening, which is covered conductor, stronger poles, different spacing on the cross arms, etcetera. And what we have not yet established at PG and E, but we will have established for this fire season, is what that higher threshold would be. And we have been trying to understand the risk model of that better and have some data on the performance of hardened circuits through fire season and through PSPS events in order to make a field validated decision around how to do that, right?
We can build all the models in the world and make assumptions there, but we wanted to see how those assets actually performed. And we now have some of that data to understand how some of those assets performed. And so the intent is to have that threshold established this year. Now there will be a very limited number of areas where that threshold, that higher threshold for a hardened asset, it doesn't take it entirely out of PSPS scope, but it would significantly raise that threshold. And so we will have that for this year, and then that will go into play.
But again, there will only be really a handful of circuits that are electrically isolatable that we could actually put that into play throughout this fire season. So you will begin to see that much more, I think, over time in the coming years in our wildfire mitigation plan as we continue to harden those assets and to be able to trace some of those back to sources so that they can be removed from future PSPS consideration or that standard can be raised. Happy to have follow-up if there's additional questions.
Speaker 14
to what Aaron is saying. I think he hit on a number of important aspects here associated with scope as well as process and standards. I think one of the things and one of the reasons why we have moved into the direction of hardening and cover conductor, it partly is due to our operating history and what's occurred over the last couple of years, as well as defining the scope of hardening. And so one of the things that I think it's important for us here at Edison to understand is how does the scope of cover conductor overlay on top of the way that we have de energized our system over the last couple of years? So we just had the discussion about different sectionalizing devices.
There's a couple of lessons learned here that we can apply. One is the amount of covered conductor that is needed to be installed and what Aaron referred to as isolatable segments. So as we get more and more into a segment by segment process, as opposed to circuit by circuit process, it sort of brings to light the areas that are more impacted than others. But in order to be able to raise a threshold, you have to have sufficient coverage across the geography of the risk areas that you're trying to mitigate. So I think we learned a lot and that's part of the acceleration of scope that we're applying for this year in our highly impacted communities.
Because I do think we've reached a level of awareness in operating knowledge to one be able to cover that risk, but also to look at other alternatives to sort of maximize the flexibility of our system and minimize the amount of cover conductor that we really need in order to raise the thresholds. So I think the last part of it too, is there's still some, I think work needing to be done on really understanding just how far we could go with thresholds. So we've taken some interim steps here using the high wind warning versus the wind advisory level, feeling very confident that that is a good interim measure for now to implement. And then from that, we can measure both the reduction in PSPS events that we expect to see and also provide some clarity in terms of future operating experience. What additional things are we going to have to do?
So I do think, you know, both in the short term, as well as the long term that grid hardening and particularly covered conductor in addition to the other options here is going to be really important for us. More granular we get in understanding the distribution system.
Speaker 11
Great, thanks. That was really helpful. Just a follow-up question then.
Speaker 11
I guess what is the progress towards determining what is sufficient coverage for a covered conductor to actually start making impacts?
Speaker 14
Yeah, so part of it is defining, looking at consequence models where we have the highest risk. The second thing, and this goes to the point of sexualization is if we have our sexualizing device that's located in a place where we're, you know, only de energizing the customers that are in high fire risk areas, but de energizing ones before they get to the high fire risk areas. So that provides us an opportunity to limit the scope and relocate or install a new sexualizing device just to the area that we're most concerned about. The other thing I'll mention is that if we have inadequate coverage, I guess the thing we don't want to avoid is, you know, covering, I'll just make it up, you know, 20 spans of cover conductor and you have another 12 that are also equally risky because our models say one thing, but actual field conditions might actually say something else. So fully covered means ensuring that we have an understanding of the environmental conditions, the physical layout of our distribution circuits, how the isolation and sectionalization works and where we can fully deploy a covered conductor in order to then say that section can now move up in terms of threshold and become more resilient to the contact report and object concerns.
Speaker 16
Thank you. Just want to make sure that, San Diego, you didn't have any thoughts you wanted to add to that topic.
Speaker 6
Yeah, I just, I think we will continue to all refine our risk modeling tools, right, to take the input from the field experience and be able to have a refined model that can help us be more predictive about the PSPS impact reduction. Hopefully as we develop those models, we'll be able to have more capability to share how much reduction we can expect. But other than that, it's similar to my response would be similar to SM-two zero three.
Speaker 16
Appreciate it. Thank you. I would like to next turn to public advocates, Justin.
Speaker 19
Hello. Justin Heiber, Public Advocates Office. Question is more directed at SoCal Edison, but I'd like to hear from the other utilities as well. Do your PSPS wind speed thresholds on individual distribution circuits depend on the local vegetation type in that area? If so, how granular are you looking at them and how frequently updated is this vegetation information?
Speaker 14
So our wind speed thresholds are based on a number of things. We look at the historical wind speeds for the area. We look at the National Weather Service on advisory level. We look at the reacts consequence levels And then we also look at vegetation in terms of whether or not there's a mismatch between what our risk models say and what actual operating in our field knowledge tells us otherwise. So I think we have a ways to go when it comes to more granular sort of structure by structure information on vegetation, but because we started our journey looking at it on a circuit by circuit level, as we get into segment by segment review, that's where I think more of these opportunities will arise.
So there is some emergent work that is underway seeing if we could leverage things like a war model for looking at more dynamically based thresholds that leverage that type of data. But where we sit today is utilizing those various historical wind speeds, as well as other indices to inform our wind thresholds.
Speaker 17
This is Aaron from PG and E. I'd just add that the way the wind speed data locally is incorporated is looking at the outage history. And so that's the way that wind sort of shows up, and looking at how those outages correlate with wind speeds. The example that we would give is that that sets wind thresholds differently depending on the type of wind that is experienced locally. So the example that we'll often give folks or use is in the Sierra Foothills or in the Sierras where higher wind speeds are quite common, we see fewer outages at those low wind speeds than we do at higher.
And in some ways it's because branches from trees and the like have often already been dislodged because those winds are quite common. Unlike say down south in our service territory more, say in the Fresno or Bakersfield area, we would certainly expect more outages at lower wind speeds because we don't see the same level of wind on a consistent basis. And so that ends up incorporating itself in our models through the outage history in the various areas.
Speaker 18
And from the perspective of SDG and E, we established something that we call our alert speed, and we developed that through the ten years of historical data, which we've collected from our weather network. So it's the top 1% of winds, generally, and just like Aaron said, that's very different based off where you go. But in addition to that alert speed, we do a couple things. One, we update it every year. So 2020, as we know, had lots of red flags, lots of high wind events.
So our alert speeds are actually trending a little bit higher, as we recalculate them every year, and we look at what is normal for a given area. And then the other thing that we look at is we calculate a vegetation risk index. We look at the exact number of trees, past a sectionalizing device. So on an area that could be impacted by PSPS, we look at species, we look at any historical vegetation related outages in that area, and then that vegetation risk index is also incorporated into the alert speed, where if we have a place, let's say you're in a eucalyptus forest, we may look at the ninety fifth percentile, that top 5% of wind, just because we have the history of impact.
Speaker 19
To follow-up, do you foresee further expansion on taking into account the vegetation classes on the segments? I think Eric answered the question on that one, is that it is continuing in development. But for PG and E and San Diego, is that kind of an ongoing project?
Speaker 18
Yeah, in San Diego it is. That's one that we're working specifically. Again, we have another project going on right now with academia, where we are combining all of that vegetation data, but then now we're also coupling it with all of our historical weather data running back thirty years. So we're trying to what we're actually studying now is failure rates of each individual tree. And that the goal here is that we can get some operational model and then an AI based model that will give us additional information coming into an event.
Speaker 17
Yeah. And I would just add for PG and E, that's that's what we're actively looking at right now, and we're trying to understand and look at some of the correlations and understand. We've looked at tree density. Tree overstrike potential is the one that we think has the most promise currently, right? So that's the amount of tree that if it falls at just the right angle is longer than the line.
So we're trying to isolate what is the actual variable that really presents the most risk as we look at different ways you could think about trees coming in contact with our assets. So that is something we're definitely looking to do. We were down to a segment by segment level on the transmission system. Again, much more manageable at 50,000 structures than at 700,000 structures, but we're attempting to put that in place for this year as well. It won't have the same specificity of the transmission system, but it will be very much headed in that direction.
And over the longer term, that is certainly we would like to have a similar level of granularity that we have with our transmission system with the distribution system.
Speaker 16
Thank you all. Next, I would like to turn to Doctor. Mitchell from Massy Grade, and then I'll go to the chat.
Speaker 10
Hi, Joseph Mitchell. There have been a number of incidents over the last few years where PSPS was not activated in the right place in the right time and fires started. Also, there's a lot of anecdotal evidence about PSPS being initiated in places without really strong weather conditions. So the question is, and this is mostly for SCE and PG and E, because SDG and E has been fortunate in this respect, is this an issue with the meteorology models? How are they being validated?
And if there are issues, how are they being addressed? Or is it a weather station issue? Or is this just a consequence of system topology? And just to give SDG a hook in here as well, have you just been lucky in not having this happen? Or do you think you're doing something that's a little bit more sophisticated than the other two utilities that have enabled you to better right size the PSPS?
Speaker 14
This is Eric. I guess I can go first since Edison was mentioned first. So to answer your question, I mean, I think one of the things that we do to ensure that we're using the best information possible is we have moved towards using as best we can actual conditions versus forecast. Because we recognize that weather forecasts, you know, as far as the accuracy, as I mentioned in the presentation, we are working on developing better accurate models than what we currently have today. So we've currently forecasted a two kilometer granularity.
We're moving that over to a one kilometer granularity, but even with that, we think that, you know, trying to forecast wind speed at such a local level is going to have its inherent challenges. And so that's why we are deploying more weather stations, even using mobile weather stations where we can to sort of help us understand real time situational awareness during the event. I will say that, you know, like other utilities and PG and E and SCG and E, we have all accumulated information and verification of damage that has occurred after each one of these events. And so I think if anything, the PSPS is designed to be a mitigation. It's a mitigation that we use and it's intended to ensure public safety and the thresholds and all the techniques and things that we do are designed to one do it consistently, do it in a way that we believe is addressing the most significant risks and then using the best information possible.
And in our case, it's using real time information.
Speaker 17
Thanks, Eric. This is Aaron from PG and E. I'll build off of what Eric said. I just want to acknowledge that there are only a handful of utilities in the world that are attempting to do something like PSPS. So we are continuing to learn.
And I think a lot of credit goes to San Diego Gas and Electric, which really pioneered this process and opened their books and their process to all of the other utilities for us to come down there. And especially in the early days of this program, we spent a lot of time closely connected with them, really building our program off of the foundation that they had laid. So I want to give them credit where credit is due. Our meteorology teams continue to stay very closely aligned so that we are looking at similar characteristics. We all have different risks, and so we need to approach those slightly differently.
But there are many similarities in the program, and we continue to kind of work together to make sure that where there are best practices that we can steal liberally from each other and benefit from those. And that's very much a common mindset in the utility business. One of the benefits of not competing with each other is there's a lot of open sharing of information. And so that makes it easy to adopt each other's best practices. I will say that from the standpoint of the perception of the utilities, it's a process that we have to have a learning mentality, a continuous improvement mentality.
I think I provided a slide that showed some of the takeaways each year. And to the previous questions, I believe, from Justin with public advocates, we are looking very much our learning from last year is, are we incorporating vegetation risk appropriately at the distribution level? Can we get more surgical and granular there? And I think each year we've had, unfortunately, some incident that points us towards one area of focus along with just general process improvement and a lot of the customer support and systems that go around it. There will always be customers who will say it's a running conversation, and the feedback we generally get is some version of it wasn't windy at my house.
And when you run a program based off forecasts, that will always be instances where it doesn't materialize. What we're trying really hard to do with the sexualizing devices and the increased deployment of weather stations and the granularity of our modeling is to try and avoid it's one thing if the forecast doesn't materialize. It's another thing if there's a neighborhood that we couldn't keep in lights, even though it wasn't at risk just because of the configuration of the electrical grid and where the fire risk sat over that grid. And so that's the part that we're in particular working on so that there's far less collateral impact, which is why we installed 600 secondtionalizing devices last year at PG and E to really isolate those tier one areas that might potentially be impacted, even though tier two and tier three are the areas where we're really more concerned about wildfire risk. So hopefully that provides a little bit of insight.
Speaker 18
Yeah, and then just to build on that, I mean, first and foremost, the way that archaeologists now work together at the utilities across the state with the shared mission of advancing fire science right now. That is gaining momentum and will continue into the future. Additionally, you know, when we first started building our weather network in 02/2010, you know, I went to the textbooks, studied Santa Ana, and they said the windiest spots are the passes and canyons. And, you know, after studying and putting this, we found out that in San Diego that wasn't actually the case, right? So the science continues to evolve over time.
We continue to learn more about how Santa Ana winds behave in our service territory. And then we've since then been able to target the windiest areas, and then work with the universities to fine tune our models. So I think the main point is that the science has been an evolution. It wasn't something that happened very quickly. It was something that through a kind of continuous research and learning, we've been able to just continuously better understand how Santa Ana winds impact our particular service territory.
And we're still making improvements based off events that we had last year with the high number of events, we find even more opportunities where we do still see surprises. We continue to react and we continue to integrate, the lessons learned. And then, you know, we do integrate this into our engineering practices as well. I don't know, Jonathan, are you on at all?
Speaker 6
Yeah, thank you, Brian. I just wanted to highlight the fact that, you know, earlier on after the 2,007 wildfires in San Diego Gas and Electric Service territory, we did start with our transmission hardening. And that was, you know, started but also leveraged and balanced by the weather information as well as the fire science information that we were starting to build from our 2003 experience by bringing on board, you know, fire science experts, but also going forward as we learn more from our weather data, leveraging what would it take to harden our transmission system. And that has helped to reduce the impacts, right, by not taking up larger areas of the customer base through PSPS events, really being able to limit it to our distribution circuits that are smaller, in fact, to our customers.
Speaker 16
Thank you all. In our last few minutes, I'd like to turn to the chat. And I believe Nathan has been monitoring. There are several questions in the chat. Do you have any other questions, Nathan?
Speaker 1
Yeah, there's a couple of questions from the chat. The first one is from Jacqueline Ayers from Acton Town Council for SCE and open to input from PG and E and SDG and E as well. SCE's twenty twenty one WMP states that ignition consequence is factored into the decision to trigger a PSPS event. However, ESRB eight only allows utilities to de energize when there is an imminent and significant risk that their infrastructure will ignite a fire and does not specify that utilities can initiate PSPS based on how big a fire may get. So can you please explain how SCE believes they can lower the PSPS triggers based on ignition consequence when doing so is not authorized by ESRB eight?
Speaker 14
Thank you for the question. So yes, ignition consequence is factored in our decision making. In addition to FPI and wind speed, we also factor in as we talked about reacts consequence, but hence we de energize when we believe that there is a significant risk associated with those conditions. I'll say that one of the challenges with using a full segmentation model is that rather than tracking say 200 circuits in a relatively large event, we may actively track as many as 700 segments at any given time. So one of the things in our operations procedures, as we look to de energize circuits and understand which devices will operate and really go through the process of confirming, verifying those devices, it actually takes an average of about five to ten minutes per de energization action.
So you can envision that when we start getting into situations where things are escalating quickly or we have a relatively large event, we need to figure out a way where we have high confidence that our thresholds are going to be breached. How do we do so in a way that we prioritize it based on risk? So we look at things like additional risk factors that may cause us to take action sooner with the belief that the weather stations are reading in such a way that these conditions are very likely to occur, but being able to do so where we're not waiting, you know, and then subsequently things really escalate out of control and we miss the opportunity to de energize to ensure public safety. So hopefully that answers the question. It's not that we're lowering our thresholds.
It's more that we're prioritizing how we do de energization with a high confidence that our thresholds will be breached and using our best information possible to assess the risk.
Speaker 1
Great. If PG and E and SDG and E don't have anything else to add, I can go to the next question, which is also for SCE, which is regarding the TSPS mitigations for access and functional needs customers. This is from Megan Sumaji from the Joint Local Governments Coalition. SCE's WMP and PSPS Corrective Action Plan focus on partnering with community based organizations and other organizations to share information with AFN populations so that they can increase their personal resiliency plans, as well as SCE's existing backup battery programs. But the need for resources and assistance goes beyond information and personal resiliency plans.
PG and E's partnerships in 2020 with food banks, Meals on Wheels, etcetera, were greatly appreciated and widely used by impacted customers. The lack of similar resources from SCE is noticeable. For example, the current independent living center receives resources from PG and E, but they can only be used for PG and E customers, which leaves independent living centers, ILC members served by SCE out of luck. What plans does have to provide resources beyond information and backup batteries to AFN customers during PSPS events?
Speaker 14
Yeah, thank you for the question. And I'll kind of leverage a little bit of the spirit in which Aaron had talked about other utilities and learning from each other. So as part of our action plan, one of the things that we recognize is, I'll just take the critical care battery backup program. Last year we had every intention of reaching more customers. At the timing of which we went for procurement and solicitation for the batteries, COVID impacted in part some of our ability to do that, but we learned a lot and a great deal in terms of how to expand our programs.
So a little bit of what we've seen in other utilities, we're now expanding those critical care backups to all medical baseline, knowing also that we have overcome some of the supply chain issues that we've had. But there's a greater opportunity here for AFN that we recognize and one is in addition to partnerships, it's really about understanding the needs of the AFN population, you know, beyond just the current programs and services that we offer. So we're going to be conducting an AFN research study to understand what the needs of those populations might be so we could better augment our programs and services to meet those needs. We're going to undergo quite a bit of marketing efforts to get more information and education out there about the things that we provide. We're going to be looking at alternative ways to support the community, including looking at what we can do with food banks and other organizations to provide a broader support.
So those are the things that are outlined in our action plan and we certainly look forward to expanding our role in that space.
Speaker 1
Great. And I also have a question from John Hooker from Wildfire Safety Division. There was some, I think, some answers in the chat, but for the benefit of those who are just participating remotely or cannot see the chat, the question is for Brian from SDG and E. We can see a later end to fire season recently. How about the beginning?
Is it moving earlier, staying put or moving later as well?
Speaker 18
Yeah. Generally, what we're finding is that the the start of fire season is very dependent upon the rainfall patterns the winter before. Right? So if you look at where we are right now, we're you know 40 to 45% of normal rainfall in the Southern half of California. That will tend to lead to an earlier onset of fire season, right, unfortunately.
When it comes to actual wind driven fires, which are the cause of most catastrophic impacts in Southern California, those generally are tied in to, the colder air starting to come in in the fall. So those have actually been trending a little later. And the risk of those has been, moving further into the winter because we're not seeing that onset of winter rainfall. The climate science right now, doesn't really indicate that we'll be seeing those, earlier in the season. But that being said, 2020 was a bit of an exception where we did have three red flags in the month of September, which was, earlier than usual.
But the short answer is that, you know, where those wind driven fires, we don't expect those to consistently start earlier, but we do expect them to last longer in the winter due to lack of rainfall.
Speaker 14
And this is Eric from SCE. Maybe relatedly, we are still tracking the precipitation that we are receiving in our service territory. So, you know, there is some evidence that we're again looking at a drier than normal situation. I think one of the concerns we have, you know, it's different than the wind driven fires that we see as Brian was explaining. But last year, one of the things that we had to contend with are what we call field driven fires.
These are things that just kind of add to each other when fuel conditions just don't have an opportunity to change as a result of expected precipitation we would want to see in the wintertime. So actively tracking it, I agree that the winter season might last longer, but I wouldn't necessarily rule out fire season actually beginning sooner as well.
Speaker 16
Thank you. I think that really addressed a question that we had on our list earlier about climate change into the future and how that may play in. But if there's anything else to say on that matter, feel free to chime in. But I wanted to turn to a question on response during an implementation, and it's related to the conversation earlier about communication with the county emergency and state emergency management agencies, and the question really is, what is your incident management team use? Do you have a dedicated incident management team for public safety power shutdowns?
And then the second part, there was some discussion about provision of circuit sectionalization information and other information. Is that included in the portal information that is provided to in your communication with emergency management government agencies? And that's for all utilities.
Speaker 17
This is Aaron with PG and E. I'm happy to start. Thanks. So I'll take them in reverse order. For the provision of circuit information, we did offer and provide coaching for anyone that wanted it, circuit specific information for any emergency management office that wanted that information.
Many of them that are more sophisticated with their GIS capabilities were interested in having that. Others were less interested. There's a range of resources and approaches out there. So we have made that available for those that want it. Our mindset is to make the information available.
We also I like to joke in some of those meetings with the emergency managers that I'm not trying to turn them all into electrical engineers. We need to help them understand the outcomes for their communities. And so that's why those outage maps that we provide both in various GIS formats as well as PDFs, because many local first responder types are much more interested in a PDF map as well. So we provide both. And so it's more about, let me help you understand where events are possible and where they're likely, as opposed to, let me give you circuit data and you have to sort of figure out the electrical system, right?
So that's kind of been the approach. And again, it's jurisdiction by jurisdiction. There's different levels of interest, and we want sort of rise or lower how we engage there based on the level of engagement that they want to have and the level of understanding that they want to have. So that's been our approach there. In terms of incident management teams, we do not have dedicated incident management teams within PG and E.
We used a four team rotation for PSPS events last year. It's really two teams because there's a day and a night shift inherent in those twelve hour shifts. And so we have also had, in non PSPS times, we attend we have been building out an eight team capability to address some of the employee fatigue issues and the disruption of core business when the same employees are pulled in to help manage those events. It's particularly challenging for some of our non operating units where they don't sort of work incidents and emergencies, incidents and events. And so it can be disruptive to their more office based nine to five Monday through Friday schedules and the work that they perform in planning and functions like that.
So that is a clearly identified issue that we are working on this year for us. I would highlight that one of the areas that we're looking actively right now and we haven't settled on developing a permanent team, is for our plans function in the EOC around which is the brains, right? Meteorology does a lot of that, but the plans function is the one that turns the weather risk into, through a series of tech tools, into operational plans of where we're going to de energize, which is then used for customer notifications, agency collaboration, all of those sorts of things, public maps. And that function is so critical in the events that we are looking at while we do have some dedicated members that build those tools over the off seasons over the last couple of years and tend to work the events, we think that we would benefit from a slightly larger team of folks that are dedicated to that function given its criticality to PSPS and how the PSPS season seems to be expanding. And so that's something we're actively looking at is having a dedicated team to really be the brain trust of executing the PSPS events that all the other teams in the EOC build off of for those actual events.
So it's a little insight as to where our current thinking lies there. Thank you.
Speaker 14
And maybe I can build a little bit off of what Erin was talking about. In terms of the sexualization plans, that is going to be part of a broader effort to engage our County emergency managers around the decision making process in addition to our grid hardening plans and specifically the types of issues we can get into in operations. As far as the incident management team is concerned, so we have about 172 employees who are trained in ICS, NIMS and SIMS. We had been in this mode for a number of years. One of the things that we did a little bit differently last year is we established a dedicated and permanent team.
This right now we have about 32 people who are dedicated to PSPS. These would be the primary folks that would be stood up in terms of an activation. And also once we're done with an activation, these same folks go in and evaluate our performance and implement corrective actions that are needed to get to the next event. Right now, the way the team had started, it was more or less operations focused, but we have built some additional functions and essentially now have folks that are able to handle twenty four hour events in a dedicated way. And then when we get into very complex events, we have what we call surge resources that are leveraging our ICS structure to bring in additional personnel to handle not just the complex PS events, but when we get into other situations with wind storms that happen concurrently, if we happen to have other hazards that exist, if we get into situations with rotating outages, we have a process of escalating our teams to deal with increasing and increasing complexity.
So hopefully that answers it from SEE side.
Speaker 18
From the SDG and E perspective, when we really look at the communications with, you know, the county and the response, I mean, a lot of that starts with the comprehensive training and exercise program, which, you know, is will be implemented throughout the summer as we get ready for the twenty twenty one wildfire season. Also to build on what Aaron mentioned, right, where we take all this weather information, and we generate that detailed circuit list. Then our team of fire scientists and meteorologists, we then input all of this information into a GIS portal. And this GIS portal becomes, you know, again, it's very important to the execution of PSPS, but part of it is that that establishes a REST API to send all of the detailed PSPS information out to our community partners, including Cal OES, including the San Diego County Office of Emergency Services. So it gives them all of this information really as soon as we know it, and as soon as we get it into our system, and do our QAQC from a fire science standpoint, we hit send, and it becomes immediately available to all of those teams.
You know, within our emergency operations centre, we are all trained, all responders are trained in ICS, multiple levels, and that is what we implement. Just to build on what Eric said, really deal with the complexities like what we were dealing with in September with the heat waves and everything that came with that. The PSPS, we had the Valley Fire, 17,000 acres burning. So having that complexity to be able to scale and deal with the multiple issues simultaneously are the only that's the only other thing that I'd like to emphasize there.
Speaker 16
Appreciate it. We have I know we're a little past our our time, but I wanted to take one more question from the stakeholders, if that's okay with our partners. From Henry Burton, I believe you have your hand up.
Speaker 12
Thank you. I'll actually defer to Joe Mitchell if he has another question as well.
Speaker 10
Okay. I'd like to go back to the question of hardening. So, a couple years back, I raised the belt and suspenders question. So, if PSPS is going to be executed under many circumstances anyway, is hardening duplicative. I know that PG and E had raised the question or made the statement that catastrophic risk is different than climatological risk.
And if that's the case, what is the ratio? Because it seems to me that most of the losses that California has experienced have been on the catastrophic side rather than the climatological side. And related to that, if we wanted to get rid of PSPS, just from a visionary standpoint, what hardening strategies or other technologies would need to be put into place in order to eliminate the need for turning off the power?
Speaker 17
This is Aaron Johnson with PG and E. I can start that conversation, but I definitely look for the feedback from my sister utilities on this one. So
Speaker 17
things we look at when we train our PSPS models is we look at how they've performed against various fires. There's a sort of a two by two matrix that we often put together where we look at sort of, you look at wind and weather risk. And what you see is that there are a large number of fires that have started, where they are what our meteorology team would call hot, dry summer fires. And they tend not to be as wind driven. And sometimes that's when the fire starts.
And then maybe a couple of days later, the wind will pick up and that fire will turn into a much larger fire. But it doesn't initially start as one of these fires that's moving at these record speeds that we've seen the last couple of years. So PSPS is very much about the wind patterns, which are there's a small window of Diablo winds in Northern And Central California that tend to occur in early summer. And then we don't see that same wind phenomenon picking up again until the fall, really in September, and then it tends to persist. And it's even present in the winter, but oftentimes accompanied by moisture, and so the fire risk isn't there.
PSPS is very much intended to identify that risk. Our larger hardening effort is addressed at the wildfire risk that occurs on any given day during fire season, maybe not on one of these larger wind days. So I would say the programs are not duplicative. They are complementary. When I think about PSPS, as you pointed out, Mr.
Mitchell, PSPS is a mitigation for that risk. Increasingly, as we've seen this wildfire risk really escalate in the last several years, it is an immediate term reaction. And our intent is to look at that and say, that is not necessarily a sustainable mitigation. And so how can we begin to mitigate that? Part of it is getting better about where we do it and understanding where that risk really exists.
But even as you narrow where it is, we will reach limits on that, as Brian pointed out earlier, in the canyons and peaks, which is where we often see that highest fire danger. And so it's more consistent with the literature in PD service territory. And so we will have to harden assets or come up with other mitigations there. And we are looking in our wildfire mitigation plan. We have the REFKEL pilot, different pilots for looking at are there other technologies that we can apply to the electric grid.
Again, this is state of the art type of stuff where we would find other ways other than rebuilding that infrastructure underground or hardening it that we might be able to apply technological solutions which would allow us to mitigate that risk at a lower cost. So we're certainly looking to pilot some of those technologies. And I think that will help us with sort of the vision of how you move away from PSPS over time. But again, I'd really welcome an answer on that vision question, because we have been moving from year to year. And I would say in terms of self criticism that we have not developed yet a long term plan for what the end state is of the PSPS program.
And that is something we are turning our attention to now, now that we feel like we have a better handle. We do have much to do on the PSPS program to continue to improve. But because we have a better handle on that program based on the progress we saw last year, we think we have some capacity to begin really understanding what that vision would look like and what we might be able to do there. But again, really looking to my sister utilities to collaborate on that effort and understand, and especially would welcome Brian's comments because they are further along on this journey certainly than we are. Not to put you on the spot, Brian.
Speaker 18
Well, it's okay. I'll just put Jonathan on the spot. No, you know, and I know as we continue to look at the hardening, you know, I will invite Jonathan to join as well. But I think to your point, one of the biggest impacts that we've seen is building the weather network. What we have found is local known conditions, where we've identified several areas where the winds are measured, you know, 85, 95, even over 100 miles an hour in places where GO95 didn't have us built at that point.
So now the integration of all of this weather information really is to now build to the local known conditions. So I think that is a big benefit of integrating all of this science, and it is something that, to Aaron's point, you know, creates this separation between, you know, the hardening of the system to what we now understand the weather to be and the PSPS.
Speaker 6
Yeah, thanks, Brian. I would also add to what Aaron and Brian had said, just that, you know, as we continue to harden our systems with the cover conductor and undergrounding technologies, right, we may not be able to apply the underground in certain areas because of terrain, because of, you know, not being able to get easements or other things, right. So we might continue to depend on the covered conductor. But even in that case, lying debris and, you know, additional things that we may not have, you know, anticipated may come to be an impact, right, more so than we're thinking now with additional climate issues. So if we find that we're more successful with experience and current technologies, then I think we will be able to leverage that and see great results towards reducing PSPS impacts, as well as wildfire risk impacts.
And as I said earlier, improve our risk models accordingly. But then we may also have to continue to leverage after PSPS impacts what are some of the mitigations that we can do, just like we shared earlier. We have fixed generator solutions that we're putting out in front of customers, whereas customers are installing, you know, permanent generators. And while if you do, right, enact or put in place PSBS action, then the customers can still at least have power for the whole home maybe even just at least partial load. That may be solutions.
Microgrids are another solution may need to continue to leverage as we continue to have PSPS impacts.
Speaker 14
And from this perspective, I agree with all of the comments made. Maybe the only thing that I'll add is, you know, PSPS grid hardening, they're all efforts to reduce wildfire risk. And, know, if we are able to deploy more cover conductor that reduces the frequency of PSPS, that also has a significant buy down or risk to prevent wildfires. So I don't think that these are duplicative. I also don't think that we would expect that if we do harden the system with covered conductor that we'll never do a PSPS event again.
I think there'll always be extreme weather events, but it does raise the threshold at which we would be activating a PSPS protocol. So just to add that additional, but I agree with what everybody else had said.
Speaker 16
I really want to thank you all for the discussion. And definitely appreciate the questions from all sources, really great questions and discussions. So thank you all for a very valuable session. And at this point, I would like to turn it over for closing to Coco.
Speaker 15
Hey, Kevin. Thanks for that. Just real quick, I do see Will Abrams has his hand raised. I probably don't need as much time for closing remarks. So Will, I don't know if you had a question, but I'm happy to offer you an opportunity to ask one last question if you want.
Speaker 13
Thanks, Coco. Yeah, no, I appreciate it. It's sort of picking up where that last question left off. And really what I'm trying to understand is sort of the intersection of when catastrophic wildfires occur at times when we are doing a public safety power shutoff, and how to analyze that situation to better inform how we manage shutoffs in the future. So I think, in the past, sort of the reaction have been when fires occur, when there's shutoffs is that, Oh, well that's because we should have gone broader with the power shut off.
And that, you know, and some sort of self praise that aren't we so glad that we shut off the power because these fires would have been a lot worse hadn't we shut off the power. And of course, in some cases that is the case. And I just wanted to understand how each of the utilities were analyzing those cases where we have fires, catastrophic fires at the same time as shutting off the power to sort of understand the contributing factors and really do an analysis of, you know, did we, you know, was it the right prescriptive measure for the power shutoff? Where did we perhaps not apply the power shutoff in the right way? And to what extent you believe that those are sort of unique opportunities to really analyze a situation where those two things collide.
Thank you.
Speaker 17
This is Aaron Johnson. I'm happy to answer from a PG and E perspective. And I do apologize, I have another commitment, so I be jumping off here in a few minutes, but I can stay a few extra minutes. Exactly what you described, Mr. Abrams, is what we try to do.
So what the meteorology team is going through right now is analyzing all of the two forty plus incidents of damage that we saw during events, as well as looking at the fire season itself and the number of fires that were there and reanalyzing all of that data and trying to understand if there are elements that we can put in to our system. I'll give you an example. From last off season, we had the Kincaid fire, which again, we haven't seen CAL FIRE report on it, but we understand at a high level that our equipment was responsible for that fire. And so we looked at what are the criteria there. That was a very healthy piece of equipment that had been inspected multiple times.
There had been high definition cameras taken. Nothing was identified. However, the fire conditions on the ground were quite extreme during that time. And so we adopted something in our transmission protocols we call a black swan criteria. It's basically saying, what would happen if an ignition did start?
Are the conditions so extreme on the ground that we have to consider de energizing, even though there's nothing on that transmission system that points us to the normal criteria that we would look at. And so we adopted that criteria after 2019 into 2020 and put that in our protocol to continue to learn from those different events that we see out there. So it doesn't necessarily have to be a fire that was started by utility equipment. We want to learn from every incidence of damage. We want to learn from the outages we have in our system.
And we want to just learn from the fire behavior of that fire season and incorporate all of those things and continue to just retrain the models and continue to get a little bit better every year at our understanding of what really drives this and accommodate the changing conditions, right, as we see, as San Diego pointed out, changing weather patterns over time as well. So that is very much the mindset we have. It is quite humbling because of the opponent that we're taking on here is pretty challenging and in terms of wildfire, but we would like to think that we are incorporating all of that information. We're certainly endeavoring to do so to try and get a little bit better at our understanding each year based on all of that information and data.
Speaker 14
I think you covered that really well. I don't know if I have a whole lot to add other than we learn from every ignition event. So regardless of whether it's PSPS or non PSPS, I will tell you that if something like that occurs, you know, in situations that are PSPS, you know, have conditions that would potentially, or even warrant a PSPS, we'll start really trying to understand what went right or what went wrong with a decision either to activate or not to activate. We had an interesting situation one year where we had what's kind of sort of considered a monsoonal phenomenon where we just had a very quick rapid wind event that you wouldn't be able to predict and it's very short lived, but the conditions on the ground are off the charts in terms of what would cause typically a PSPS. So we would look at that and consider not just the PSPS protocols, but our protection schemes, if we would need to have, you know, recloser blocking restrictions in place under conditions that are not necessarily classic PSPS conditions, there are other types of things that we could do, but all of that in terms of learning from ignitions, learning from what happens before, during and after events is all important for us to digest and consider.
Speaker 18
Thanks. So a little bit to add from an SDG and E perspective, you know, from having this program in place for over a decade, what we fortunately have not been in a position to this point that we have had a catastrophic wildfire on the ground during a public safety power shutoff. But what did come to mind was last year, day before Halloween twenty nineteen, we were in a very extreme event. We did have a public safety power shutoff in place, and there was a the Saaday fire broke out in our service territory, not related to the utility. But the resources, there was not a big pull on the firefighting resources.
So they were able to keep that contained at 100 acres, where a lot of our experts in our modeling was immediately showing that had that not been contained and all resource were available, that could have turned, into a catastrophic event. And then after they got that contained, just a couple hours later, the Miller fire broke out in the community of Valley Center also had the potential, and a lot of our consequence modelling showed that that could have been a catastrophic event. But again, due to the lack of strain on the firefighting resources, they were able to get that one as well. And I'll leave that, leave that there.
Speaker 15
With with those answers, I think we will go ahead and conclude this panel as well. I want to thank all of our panelists, not just this afternoon on the PSPS panel, but all of our panels throughout the two days of these workshops, as well as a big thank you to all of our moderators and the IT folks behind the scene making all this work. Really appreciate all the effort and work that went into this. I think there was plenty of good discussion that came out of our various panels these past couple of days. And one thing that I do wanna flag for any stakeholder or or folks in the chat who were not able to get any of their questions answered, just wanna remind folks that we're still in the open discovery phase of the WMP proceeding, And so we would remind folks that you can still send any data request directly to the utilities through that process and get your questions answered that way in the event that, questions remained unanswered.
With that, I just wanna cover, and I'll be pretty brief here, a few last steps and and just the closing remarks. One is just flag some upcoming things. We have opening comments on the WMP due March 17 and reply comments due March 24. I did see just this afternoon that we did get a, request for extension, which we will be considering and responding to, shortly. Beyond that, we do have the small and multijurisdictional utilities as well as the, independent transmission operator, WMP, is coming in on March 5.
And, additionally, on top of that, we do have the supplemental responses from the large IOUs for the quarterly report evaluations that'll be coming in on this Friday, February 26. So I just wanted to flag those few things. And then finally, just want to remind folks that these web workshops were recorded, and the Webex recordings as well as the workshop materials that were projected will be posted online on the WMP web page. Bear with us. It does take several days for us to render and get the videos to upload online, and so we will be getting that on as soon as possible.
But be on the lookout for that. Other than that, I want to thank everyone for their participation. I know two full day workshops is especially in this virtual environment, is a bit, taxing on folks. So I'm hopeful that, people found it useful, and, thank you all for your continued engagement and participation. With that, we'll go ahead and conclude our workshops for the large IOU twenty twenty one WMPs, and look forward to the comments and reply comments coming in, in the next, several weeks.
Thanks, everyone.
Ask anything about this video — I analyse its transcript to answer.