 Green Think Tech Hawaii, welcome to Honolulu on another brilliantly sunny day. We as everybody in this vast audience knows we are headed for 100% clean energy by the year 2045 and one of the important avenues to achieving that is data. What in the world does data have to do with 100% clean energy? We're just about to find that out. Our honored guest today is Troy Wooten who is our drumroll data science specialist. I've been on two meetings this morning, one local one on the mainland and I mentioned the field of data science specialist and I got these very curious looks. Wow, how are you guys are way, way, way ahead of us. What does he do? I said, well, tune into the program and you will find out. So our data science specialist who's been with the Hawaii State Energy Office for, I believe, well over a year now, Troy is Troy Wooten and he has put together the data portal and we're going to he's going to walk us through that and tell us about the potential that this has for achieving 100% clean energy. So Troy, welcome, welcome to the program and please begin by telling us, oh, well, before he tells us what a data science specialist does, how in the world do you get a degree in data science specialist? Or is this an entirely new field that does not yet have a degree attached to it? Well, thanks, Howard. Yeah, good question. Data science, it is a relatively new field. And it's sort of an amalgamation of things, you might say. So it's really unifying the fields of computer science with applied data analytics and statistics and now, you know, in this day and age, machine learning and advanced techniques, such as deep learning, are coming more and more into the picture. But the degree that I got was actually a computer science degree with a focus in data analytics and data science. And so basically, it's a lot of techniques and algorithmic developments and developing software to manipulate data and visualize it and disseminate it in a way that is readily approachable and digestible by a late person. Yeah, that's something we should emphasize. Is this is intended for public use? Is it going to be what do you call it an open source? That's correct. Yeah, so if we could, if we could transition to the first slide in my presentation, I prepared a slide deck. And maybe maybe onto slide two. And we can introduce, yes, our new energy data portal. And I'll explain what this is for a moment. So it's basically an open access page on our website, energy.hoai.gov, that provides access to a set or suite of curated high quality data sets that our data science team in house has done the work of preparing and aggregating together from myriad sources and synthesizing these curated packages of high quality information and sort of snippets of meaningful details about the energy sector in Hawaii. And first question, Troy, what is curated? Good question. So by curated, what I mean to say is, you know, we take a lot of data from a variety of sources and there's a lot of high quality data out there on the internet coming from federal and state and local entities, as well as the utilities and DOE and national labs. They're all producing tons of data all the time and they're sharing it through various means and media. And what we do is we pull in all of that information and we kind of add our own touch of spice to it, let's say. We take data from multiple sources and we join it together and we augment the data that the information that was already there and we package it in such a way and in many cases in a visual way so that key insights and trends sort of just jump out at you when you approach that data, right? So a lot of times with raw data, there's information that's embedded in that data, but you have to kind of have a specific set of skills to know how to reach into it and extract meaningful insights. And so we've kind of done a lot of that lead work for you and produce these approachable packages that are much more readily digestible. Yeah, I'm so glad you're doing that because, as you can imagine, I read a heck of a lot of technical data or technical papers. And as I'm scanning through, I'm saying, OK, interesting, interesting, interesting, now, where is the graph that illustrates all this? Oh, there it is. OK, now I understand. So I'm very glad you've structured it that way, Troy. Oh, thank you. Yeah. And that's the hope. The hope is that this is going to be a tool for the community and for energy stakeholders to leverage and kind of more quickly access the information that they want to know about with less friction and fewer levels of indirection involved. And so a lot of people doing this research with you are people whose goal it is to provide their little contribution towards more and more clean energy and less and less carbon in the atmosphere. That's correct. Yes. So we are, you know, working in collaboration with other state entities, as well as the utilities and the DOE national labs on a lot of our data analytics and data acquisition efforts. And yeah, trying to, you know, find that set of common objectives and common goals, which are, as you mentioned, net zero carbon by 2045 and 100 percent renewable energy in the electric sector as well, as well as decarbonization, decarbonization of transportation and electrification of that sector. And so really all of our analytics work is trying to kind of fuel and provide the necessary insights to guide our progress and our navigation towards those goals. Yeah. So I guess just to give a high level overview of the data portal itself, I wanted to start by kind of giving a kind of a little bit of a deep dive into the functionality of the interface, so if we could move to the next slide. Functionality of the interface, I like that. So basically, if you go to energy.y.gov slash energy dash data, you're presented with a page that looks just like what you see here in the screenshot. And essentially what you have is a portal page that has a collection of data that has been grouped into subsets or categories such as renewable energy, non renewables, transportation, energy efficiency, et cetera. And what that's intended to do is to kind of group the data in a way that makes it more readily navigable by a user who's just visiting the portal for the first time. The other key set of functionality is we also have a search engine. And so at the very top of the page, if you type in search terms into the search bar, you'll see that the search results immediately are updated based on the keywords that you type in. Beautiful. Yeah, exciting stuff, Detroit. And so if we move to the next slide, we can cover the rest of the functionality real quickly. So as I mentioned, if you type in search terms such as the word fuel into that search bar at the top, you're given an updated set of filtered results below. So the only data sets that appear are going to be those that contain the term fuel in either the title or the summary description or a preset list of annotated keywords. That have been applied to each of these data sets to make them more readily searchable and accessible. And then you give you begin to give examples down there in the graphics. Exactly. Yes. So what you see underneath the search bar and the navigation menu is an example of three data sets that we have on the portal page. One of which is visualizing ethanol and biodiesel consumption in Hawaii. Another of which is showing the breakdown of vehicles in the states. And another that's showing the distribution of the fuel economy of vehicles geospatially. And so that's just the snippet sort of a sample of the larger suite of data packages that we have available for every data set that we have available on the sites. We have a dynamic interactive charts that's tied to that data that presents it in a visual way that's interactive and sort of more approachable than just looking at a table or a wall of numbers. It's intended to kind of like I mentioned those key insights and trends to just pop out at you immediately upon looking at that data. And that's that's what we're trying to do there with the visuals. So now if we move on to the next slide, another thing that I wanted to focus on a key piece of the functionality is for each data set that's represented by a card within this user interface. If you hover over the title of the data, you can click on that title actually and it brings up a pop up window that shows more information and allows you to download the full data set. If we move to the next slide, we can see an example of what that looks like. So in this case, this is a data set that's showing the average residential electricity bill per household and month on the various islands over time. And so this pop up window, it appears it's kind of superimposed over the main contents of the page and it presents a larger version of the interactive chart as well as containing a link to download the full data set and summary information and descriptive metadata that explains what that data is and what it contains, as well as sort source attributions that explain where we got the data from. And I noticed that this is broken out by islands, what the individual islands consumption is. And I see a Molokai is way down on the lowest end of consumption. And I had the good fortune to spend a good deal of time working on energy measures on the island of Molokai. And I was doing an energy audit of Molokai High School and the principal was taking me around and I asked, well, what's your hot water consumption? They said, oh, we just washed the pots and the pans in hot water. And I said, what about the gyms? And he laughed and said, you know, our hot water systems for the gym broke down quite a while ago and we couldn't afford our utility bill anyway, so we just left the hot water system broken. And so the kids take showers in cold water. And I said, isn't there a hue and a cry? And he said, no, no, no, these are country kids are used to it. So that was a good example of a very high energy prices. And the entire island of Molokai is considered a low income area. So that's good little snippet about the relationship between behavior and income levels and so forth. But that was a little digression there. Oh, no problem. Yeah, thank you for sharing that insight. And yeah, going back to the slide, if you look at that chart, you'll notice that, you know, despite Molokai having some of the highest electricity prices in the state, their monthly bills are actually the lowest in the state because they're consuming such a low amount of electricity relative to the other islands. Just as Howard mentioned. Yeah, so those are those are the types. Oh, sorry. Go ahead. Yeah, we've when I was there, this is many, many years ago, we were pushing solar, solar, solar, water heating. Back then, and we got a lot of takers. And since then, there's been other initiatives on Molokai to help residents reduce their utility bills. So this is it's showing those low numbers is a success story to me. Yeah, I agree. All right. So let's move on to the next slide if you don't mind, Howard. So in addition to highlighting the functionality of the data portal, the other thing that I wanted to make sure to do in this talk is to highlight the benefits of this open data portal and sort of the larger data strategy and the effort that the office has been undertaking over the past couple of years. And so one of the things that we've been trying to do first and foremost is just unifying all of the disparate data that's out there. And by disparate, I just mean to say that there is a lot of high quality data out there coming from federal entities such as, you know, the National Highway Transportation Safety Administration, the U.S. Energy Information Administration, the EPA through their fueleconomy.gov website tracks, you know, the fuel economy for all light duty vehicles that are released on the markets. We also get a lot of data from other state agencies such as the Department of Business, Economic Development and Tourism's Research and Economic Analysis Division, that chart that we had on the previous slide showing the electricity bills by island that comes from their energy data warehouse. So we pull data from both their monthly energy trends as well as their larger economic data warehouse, the subset of that pertaining to energy. And we pull that into our infrastructure and produce our own sort of insights from that. And I see what you're doing is grouping like with light, be it from our sources, mainland sources, national lab sources and putting them all together under one specific umbrella. Exactly. Yes. So what we've done is we've taken all of that disparate data and we've unified it in a unified data platform that runs on the cloud. And in our case, AWS, which is Amazon Web Services. And then we also make use of a service level platform called Amorphic. And what Amorphic does is it kind of simplifies our ability to leverage the AWS platform to build this unified information and data repository on the cloud. And then using that kind of core data platform, we're able to build public facing applications and data products such as the energy data portal that we just released, as well as a variety of other products and visualizations that are going to be coming down the line. And then so as an example on this slide on the right, what we have is a visualization, a core of left map of fuel economy by zip code on the island of Oahu. And so regions that are shaded or sorry, this is plug-in electric vehicle percentage by zip code. Sorry, get them mixed up sometimes. We have a few core of left maps on the portal that look quite similar. But sorry, this is plug-in electric vehicle percentage by zip code and regions that are shaded darker have a higher percentage of plug-in electric vehicles in their light duty vehicle population than those that are shaded in a lighter color. And so as you can see, the kind of key geospatial trend just jumps right out at you, right? You can kind of see where a lot of the plug-in electric vehicles are residing on the island. Yeah, and the darkest segment looks like a zip code 9, 6, 8, 2, 1, which is the Wailaiiki area, which also happens to be the highest income zip code in the state so that we see so far a correlation between high income and EV ownership, which makes sense. But something that the energy office is attempting to do is ease the way for middle income and even lower income people also to acquire electric vehicles. That's one of our big, big initiatives. So this gives us an excellent place to start. Here's the good news about 9, 6, 8, 2, 1. Now, what can we do about the other zip codes? That's correct. Yeah, we need to have an intimate understanding of where we are currently in order to better know how far we have to go, right, in order to fully electrify ground transportation. It's going to take a lot of work and a lot of variables are going to factor into that equation, for sure. What's the phrase if you can't measure it, you can't master it? And this is an excellent example of measuring now. That's an excellent phrase. I like that a lot. Yeah, that very much gels with what our objectives are. And so now if we could proceed to the next slide, another benefit of this data portal is that the data science team within HSEO has done a lot of work basically annotating every package data set that's available within the portal with rich metadata. And so first, I'm going to define the term metadata. So what is metadata? It's basically data that describes other data. And that's not a super useful and for not a super useful definition, but basically what it's saying is if we have a data set, we need to know what that data set actually is and what it's describing, what real world entities is it referring to? And so metadata are things like, you know, starts with a descriptive title of the data set, like what is this data set and then a summary description of what it contains. Also source attribution. So where did we get the data from? And when was it created and how, as well as for every field or variable within that data set, such as in this example of field economy by zip code, right? We have the zip code tabulation area or ZCTA, which is basically the geographic representation of a zip code that allows us to tie a specific data point to a location on it. And then we have the average or median fuel economy by zip code rendered as a shaded region. And so in this particular example, regions that are shaded with a lighter color have a higher fuel economy than regions that are shaded in a lower color. And you'll notice a similar trend that kind of correlates with the distribution of plug-in electric vehicles as well. We have areas with higher efficiency vehicles that also have a higher percentage of plug-in electric vehicles. And something else I get out of this graph is the fact that looks like the Y and I coast and parts of the North Shore have the highest fuel consumption. And those are also the areas with the lowest per capita income. So you have this inverse relationship. And that gives us the baseline for challenges. How do we reduce the fuel use of the people who need mostly to reduce it due to their limited incomes? So that's, again, we're measuring something that we've defined the problem right in one little graphic here. Definitely, yeah, that's very well stated, Howard. And yeah, one thing that folks who visit the portal will notice and see right off the bat is that a lot of the data packages are kind of highlighting that heavy dependence on petroleum that we have for our energy purposes, both within the transportation sector but also in the electricity generation sector. And so that's a good segue into the next slide, actually. So the final benefit that I wanted to highlight of this data portal is insight generation through the bringing together of disparate information and it's juxtaposition in a visual media. And so what do I mean by that? Well, if we look at the chart on the right, we see plotted the percent change in residential electricity price over time for each of the islands over the January 2021 baseline that we set in this particular case. And basically what this chart highlights is that electricity prices have shot up rapidly over the past year and a half to close to two years now on all of the islands except one outlier and that outlier is the island of Guay. Now, why do you suppose electricity prices would have remained relatively stable on Guay when on the other islands they've shot up dramatically? So the reason for that is that we have such a high dependence on petroleum for our electricity generation. And in the recent months, the price of petroleum crude oil has shot up astronomically. However, on the island of Guay, KIUC boasts an almost 70% annualized RPS as shown in the center chart versus Hawaiian Electric only has a just under 40% RPS when you look at the 2021 annualized RPS for both utilities. And KIUC being the utility that provides electricity to Guay having such a high RPS and such a high penetration of renewable electricity sources, you see that the electricity prices for their customers have not succumbed to the recent price hikes in the same way as customers on the other islands have experienced. This is just so useful, not just for us in Hawaii but for the entire nation and indeed for the entire world. To me, Kauai is the portent of the future. It's much easier there because there's only 74,000 people on the whole island so they can control things and it is a locally owned utility. So it's much more flexible and much more accommodating to the needs of the residents and they are able to shut down their power plants on most sunny days, just shut them down because solar takes care of everything, solar and other sources. So this gives us a peek into what could be the future including lower energy prices, electricity prices if we would just follow the example of Kauai and this takes it from the abstract to the very real. Those are real life people living on Kauai and they're paying real life utility bills. So this is a great look into the future, Troy. Oh, thank you, Howard. Yeah, so these are the types of insights that we're hoping to foster and facilitate through our larger data strategy and data cleaning and preparation efforts but also through this public energy data portal as well provide these sorts of key insights and trends for folks to access readily and to make use of in their own decision-making be they a stakeholder or just an interested citizen. Yeah, I just got a couple of minutes left and you just gave one example of one data set. I believe you have developed so far 30 such data sets. That's correct, yeah. So what are some examples? Sure, so a lot of the data that we pull from is coming from the Energy Information Administration and they track kind of high-level trends in Hawaii's energy sector. So where is the energy coming from and which sector is that energy being used within and in what amounts? So one of the key data products from the EIA that we make heavy use of is their state's energy data system or SEDS. And that basically gives high-level breakdowns of what are our primary sources of energy across things like petroleum, coal, solar, wind, biofuels, et cetera, and then which end-use sectors is that energy being used in and in which amounts? And so we have data packages that show detailed breakdowns of our petroleum use within the transportation sector, for example, where we've even teased apart the transportation sector into marine aviation and ground transportation. And just to pose that with our consumption and electricity and residential, commercial and industrial so that you can see kind of proportionally where the majority of our petroleum is being used. Another example is we also have data sets in the energy efficiency category that are highlighting the electricity consumption of state agencies and tracking how that is changing over time so that we can see how we as a state are leading by example. Beautiful. And we're virtually out of time. Could we see the last slide, please? Which gives Troy's contact information. Would you enjoy receiving questions and information from the public, Troy? Yes, yes. So anybody who again wants to visit the Energy Data Portal can go to energy.hy.gov slash energy-data. And if folks have questions or suggestions or new data sets that they would like to see published or anything else, feel free to email me at bill.tida.buten at hawaii.gov. Beautiful. And as I said, I was on two seminars earlier today and nobody had even heard of your position. So I think maybe Hawaii is leading the nation in this regard. So, hearty congratulations, Troy. And we must say fond adieu. Howard Wiig, cold green. See you next time. And again, thanks so much, Troy. Thank you, Howard. Thank you so much for watching Think Tech Hawaii. If you like what we do, please like us and click the subscribe button on YouTube and the follow button on Vimeo. You can also follow us on Facebook, Instagram, and LinkedIn, and donate to us at thinktechhawaii.com. Mahalo.