 I'm Jordan Geller with the Clean Water Initiative Program and I'm joined by Leslie Matthews from the Lakes and Pond program and she's going to be presenting on the Lakes scorecard. If anyone has any questions you can ask during if it's content related otherwise hold questions until the end. For everyone online I'll field questions and help with any technical difficulties which hopefully there are none anymore. And then if everyone can complete the survey monkey I'll send out a link after. Your feedback will help us improve the Clean Water Electricity in the future. And I think that's it. Are you ready to take two? Here we go. Thank you again everyone for coming. I'm Leslie Matthews in the Lakes and Ponds program of the Watershed Management Division of Vermont D.C. And I'm here to talk about the Lakes scorecard. So a Lakes scorecard was originally the brainchild of Amy Peacot about 12 years ago. She wanted to create an easy to use tool for our lay monitors and the public to understand the conditions and trends in their lakes. And over the years everyone in the Vermont D.C. Lakes and Ponds program has contributed to building and enhancing the scorecard. The scorecard is based on data we already collect here at Vermont D.C. or through our lay monitoring partners. It's a public education tool based on science and we do add a little bit of PB&J professional best judgment also to back up our scores. To get to the Vermont Lakes score the easiest thing to do is to Google Vermont Lakes scorecard. And a link should come up at the top of your page that will take you to the webpage, the Vermont D.C. webpage for the Lakes scorecard. From there you can open the latest version of the Lakes scorecard in Google Earth. You do need to have Google Earth, the free Google Earth software, installed on your computer for this to work. And that will bring up the Lakes scorecard as shown here in Google Earth. And you can zoom into a particular lake and click on the scorecard icon. And you will get a pop-up with links to more information and the SecchiDisk icon which is the scorecard icon in the color there at the top of the pop-up. So the scorecard provides a color coded evaluation of four lake characteristics displayed in the icon that resembles a SecchiDisk. Which for those who might not be familiar is a widely used tool for assessing water quality. The four characteristics are nutrient trend, shoreland and lake habitat, mercury pollution and invasive species. And the scores are all color coded as blue equals good conditions, yellow equals fair conditions, red equals poor conditions, and gray or white means insufficient data. You can also bypass the Google Earth interface and go directly to the scorecard data layers from the Vermont DEC Lakes scorecard webpage as shown here. And that will open up the main Lakes scorecard page from a link on the website. And from here you can also see the scorecard icon in the top left of the page there. And you can browse to any lake from the drop-down list at the top to see different scorecards for different lakes. There is one quirk of the software that you have to click View Report after you select the lake that you're interested in from the drop-down list. And in addition to the main Lakes scorecard which I'm going to talk about in a lot more detail, you can also pull up a list of aquatic plants including both native and invasive aquatic plant species. And you can also pull up a list of fish species in the particular lake of interest. So from the main scorecard data page you get to see the underlying data that informs the color scores for the second-disk quadrants as well as some general Lakes statistics on the left side of your screen there. The first of the second-disk quadrants, the top left quadrant of the second-disk, is the nutrient trend score. This score is based on trends in spring and summer phosphorus concentration, chlorophyll A concentration, which is a measure of algae abundance, and second transparency which is a measure of water clarity. And we evaluate those trends using a Kendall-Tow rank correlation statistical test on the annual means for each parameter. To calculate the trend score, we use a point system to score each parameter from zero to two with two being good, one fair, and zero poor. The points are assigned based on the statistical significance of an increasing or decreasing trend or no trend at all. And I'm happy to answer questions about the details of these calculations later. We rescale the three summer parameters to come up with a single overall point score for summer data. And then we add together the spring and phosphorus score and the summer score, the spring phosphorus score and the summer score to derive the final score based on all four parameters. For lakes that don't have lay monitors, we generally don't have summer data, in which case the final score is based only on the trend in the spring phosphorus concentration. So this point system is objectively based on the underlying data, but we did insert a dose of professional best judgment and have scaled the final values so that the scoring system made sense to us based on what we know about our lakes. So this is an example of a lake with a fair trend score. This is Lake Caspian. Caspian Lake scores a zero for summer phosphorus because it's highly significantly increasing and a one for spring phosphorus because it's moderately significantly increasing. And then it scores twos for Secchi and Chlorophyllé because they're stable, which leads to a final overall score of two which translates to a fair trend score. One more example. This is an example of a poor trend score. Lake Willoughby, unfortunately, is an example of a poor trend score because it scores zeros for both spring and summer phosphorus because both of those are highly significantly increasing for Lake Willoughby. Leslie, can you just define generally what it needs to be significantly increasing so that's statistical terms? Right. So significantly increasing means that it has a p-value less than 0.05 and in statistics a p-value represents the probability of rejecting the null hypothesis. You're sorry. In a sense, it's a measure of how confident we are in that trend. So when we distinguish between highly significantly increasing and significantly increasing for the Lake score car, we're saying if it's highly significantly increasing we have a lot of confidence in that trend because it has a very low p-value. And if we say it's not highly but just moderately significantly increasing, that means we have good confidence, confident enough to say that it has a trend but we're just not as confident. And that confidence depends on different factors. It depends on the consistency of the trend over time and also the amount of data we have. So we might be seeing a positive trend, but if we don't have enough data points, we still might not have enough confidence. That's helpful. Hope that helps. So the next quadrant, so that's the trend score, the upper left quadrant of the little Sekigisk icon. The next quadrant of the Sekigisk icon is the Lakeshore and Lake Habitat score. So for this score we have adapted a Lakeshore Disturbance Index that was developed by EPA for the National Lake Assessment Program. For this index we visit 10 random sites around the lake and make observations about the presence of various human disturbances like buildings, lawn, roads, et cetera, near the shoreline. Extensive research by the Lakes and Ponds Program and also others in the literature has shown that these kinds of development disturbances in the near shore zone lead to degraded shallow water habitat for fish and other creatures that live in the water. So we use this Lakeshore Disturbance Index as an indication of what we expect the quality of the shallow water habitat to be. And this index is a measure of both the intensity and also the extent of development on the immediate shoreline around the lake. So the final Lakeshore Disturbance Index is the average of the intensity and extent of the development around the lake. And I can go into that more detail later if people want. So then our shoreline and lake habitat score is based on a scale that was derived by the National Lake Assessment by EPA for this Lakeshore Disturbance Index. The third quadrant on the bottom right is the Aquatic Invasive Species Score. And this score is quite simple, no fancy calculations. A lake score is good if we have no record of any of the invasive species you see there listed. Eurasian water, mouthfoil, varied belief, water, mouthfoil, zebra mussels and so on. And a lake score is poor if at least one invasive species has been confirmed in the lake. And Lesley, just to confirm, it's only those species that, I mean, I guess those are the ones that... Those are the ones that I guess one that's missing there is curly leaf pondweed. And I think we would also score a lake poor for curly leaf pondweed, but I think I forgot to add it to this. Good catch. And the final bottom left score is for mercury in fish. So, oh dear, that didn't come out. Apologize this slide is missing a graph, but it's showing up on my screen, but not your screen. Do we have to just do one more click? Oh, maybe, yeah. There we go. Voila, that's weird. Because I didn't animate this, but anyway. So our mercury scores are essentially based on a 2004 study that was conducted by Vermont DEC from which we know that there's likely widespread mercury contamination in our lakes. And in some lakes we know specifically that mercury levels in fish exceed safe guidelines for fish consumption. Only two lakes in the state actually score good for mercury, and that's because those lakes are highly nutrient rich lakes in which the mercury contamination is less available for uptake by fish. So that's the second disc scores for the lake scorecard. In recent years, we've added a couple of additional scoring features to help people understand the status of their lake and its surrounding watershed. So the water quality standard status score is based on the water quality assessments that the Lakes and Pond program conducts and reports on to the EPA. On Google Earth and on the scorecard trends and status page, lakes are color coded according to their status score. So you see here that Curtis Pond is the yellow polygon in the bottom right imagery of the screen because it has a score of fair for its water quality standard status. Say that three times faster. The status score is directly related to our assessment of the current status of the lake as reported to EPA. Lakes that are impaired usually because of phosphorus or nutrient pollution score poor for this category. For lakes that are altered due to flow manipulation, we use the term altered rather than impaired, but those are also listed as poor for this score. Flow alteration for lakes typically means the lake undergoes substantial water level manipulation because of the operation of a dam. We also report lakes that are stressed, which is the yellow score here, which means we have concerns about nutrient levels or aquatic invasive species populations or some other problem that we think needs to be addressed, but we don't have enough data or it doesn't yet rise to the level of an actual impairment listing. And the causes for any stress or poor scores for this scorecard category are shown at the bottom of the page. So here for Curtis Pond you can see that it's stressed for nutrients and phosphorus. Leslie, do you have a question? Here's the yellow outline. I'll get into that. Coming up next, or soon. But I do want to point out that it's possible to have a trend score that's good because the trend score parameters are stable or even improving, but at the same time, for example, if Curtis Pond, the status may be fair or poor because the lakes and Pond's program has determined that the nutrient levels stable are nevertheless too high to be in compliance with our water quality standards for that lake. So trend and status mean two different things and it's just good to keep that clear. Leslie, if you wouldn't mind just backing up. So for this, the color shade of the chart you have green on the top, sort of a turquoise in the middle and then the light blue. Right. Those colors correspond to the trophic condition thresholds that we define for our lakes in Vermont. And for those who don't need a refresher on trophic condition, the darker green at the top is lakes that fall into the trophic category and those are highly nutrient enriched productive lakes. The middle teal colored bar is mesotrophic lakes that are intermediate and the blue bar is lakes that are oligotrophic, which means that they have low nutrients and clearer water. So this is just another example of a lake that has in this case a poor trend score but it has a good water quality standards score, Lake Willoughby. So Lake Willoughby has beautiful crystal clear water. It's currently in quite good condition but it has a poor trend score because we're seeing some concerning increases in phosphorus levels that could spell problems down the line in the future. So it has a poor trend score that it highlights the fact that we might need to address issues for that lake even though it's water quality standard status is still very good. So now we get to the outline that you were asking about Jordan. This is the other scoring category that we just added in the last couple of years and this is the watershed disturbance score. We wanted to give users of the lake scorecard some information about watershed condition for their lake because that's a place where oftentimes it may be possible to improve water quality conditions by implementing better stormwater management, manure management and so forth to reduce inputs into the lake. So for this score we adapted a landscape development intensity index published by Brown and Vivas from the University of Florida. So we use GIS imagery to categorize and quantify land uses in the watershed and then each land use category is weighted according to the degree to which it can negatively impact water quality and the final index is a weighted average of the land use areas and intensities. And the outline shown on the map in the lower right corner is the outline of the watershed and the color corresponds to the watershed score. And then we derived our own Vermont specific thresholds for scoring the watershed index to our good, fair, poor categories by identifying breakpoints in the scored data set. So that's the lake scorecard. How do we use the scorecard? In the lakes and ponds program it's been useful for program planning purposes. It's helped us identify where we need more data, where we need more education about shoreline or watershed management practices and so on. It's also been a valuable tool for our watershed planners in the watershed management division's basin planning program. They've looked to the scorecard for information about where to focus projects in the watershed that might help improve lake water quality invasive species or shoreline management or any other elements that they include in their planning process. And we hope that the scorecard has been valuable in helping our partners including our lane monitors and others to think about how they can work to improve their lake score over the long haul. To that end, we include a handy checklist for lake associations or watershed groups to come up with ideas for projects or community education efforts. And we also have a host of other information on our website to help watershed groups and lake associations find resources to address issues. So how are we doing? These graphs summarize the numbers of lakes in different categories of the scorecard for each score. So if you look at these graphs sort of in the aggregate you can see that it appears that for water quality trend we have a lot more lakes that have good scores than have fair or poor scores. And the same is true for aquatic invasive species. Despite the challenges that some lakes face with extensive invasive species populations we still have many lakes that are uninfested. The shoreline and habitat scores on the top right reflect a finding actually of the 2007 National Lake Assessment that we participated in and it showed that shoreline and lake habitat is a very important, the most important stressor really in Vermont. And this is in some ways good news because it's something that we can address through better shoreline management practices. And then mercury, again we know that we have widespread contamination of mercury in our lakes and much of this comes from power plant deposition from outside of the state that we don't have as much control over. So despite the generally overall good news from the lake scorecard we did notice a couple of years ago that quite a few of our lowest nutrient lakes with exceptionally good water quality like Willoughby here were nevertheless showing some alarmingly increasing phosphorus trends. So here's Shadow Lake and Glover, Maidstone and so on. So we decided to take a deeper dive into our long term spring phosphorus data set to begin to look at what might be going on with this. So we have 153 lakes greater than 20 acres that were sampled at least once during the 1980s, once since 2000 and then at least three times overall with a median of 11 sampling events per lake. So this is about half of the lakes in the state over 20 acres that we have this very long spring phosphorus data set for. So taking this data set we broke the lakes down into trophic condition. I explained these trophic condition categories earlier. And then so we classified all of the lakes in this data set according to their trophic condition based on their average total phosphorus concentration in the 1980s. And then to look at overall long term phosphorus trends and avoid this statistical problem of multiple comparisons I use what's called a linear mixed effects model, the results of which are shown here. So each vertical line in this graph represents a lake. The black dots represent the mean of the estimated slope for total phosphorus so the rate at which phosphorus is increasing in that lake. And the vertical lines represent a 95% confidence interval around that slope. The length of the line is an indication of the degree of confidence. So when the entire vertical line is above the horizontal red line as in the ones that are circled in red there that say increased TP when the entire vertical line is above that horizontal red line that represents lakes that are significantly increasing in total phosphorus. And below the line represents lakes that have significantly decreased. And when the vertical line crosses the horizontal red line that means that that slope is not significant that we consider that lake to be stable. And the lines are color coordinated according to trophic category. Blue is a ligatrophic, green is a trophic and red you trophic. And the point of this graph what you can see is that almost all of the ligatrophic lakes so all of those blue vertical lines are increasing are not crossing the red horizontal line, they're above the red horizontal line. So all of those ligatrophic lakes are increasing in total phosphorus over the last 38 years as well as many of the mesotrophic lakes. On the other hand almost all of the utrophic lakes are stable are actually decreasing as you can see on the left side of the graph. So this is the 30 years worth of data? 40, yeah. So based on those confidence intervals in that graph I just showed I estimated that 96% of the ligatrophic lakes, the low nutrient lakes in Vermont have increased in phosphorus over the last 40 years. 38% of the mesotrophic lakes have increased but in contrast almost none of the utrophic lakes have actually increased in phosphorus and 22% have actually decreased in total phosphorus. So in a sense that's good news because it suggests that the efforts that we've put into addressing issues for our most nutrient pollutive lakes may be paying off and we're not seeing deteriorating trends over time for those lakes but it's very concerning that almost all of our ligatrophic lakes are increasing in phosphorus even though by and large they still have very good water quality. It's a disturbing trend. I also did the same analysis using annual means summer phosphorus data collected by our lay monitors and that's shown here. It basically shows the same overall results. Oligotrophic lakes in the lay monitoring program are all increasing in phosphorus. So why are so many of Vermont's lowest nutrient lakes seeing increasing phosphorus trends? So that's the million dollar question. We really don't know for sure yet but we tossed around a bunch of different hypotheses. Recovery from acid rain we don't think is probably a valid hypothesis because two-thirds of these oligotrophic lakes are moderate to high alkalinity lakes that wouldn't have been impacted so much by acid rain. We definitely are considering something related to climate change, longer duration of stratification, might lead to more net internal loading of phosphorus from sediments. Unfortunately we don't have long-term oxygen profile data on these lakes to be able to really answer that question. Another possibility is that we could be seeing more runoff due to increases in the intensity and precipitation related to climate change and these hypotheses are not mutually exclusive and finally we could be seeing the results of land use practices in the watersheds. So one thing to note in terms of the climate change hypotheses is that I also ran the same analysis on summer data from Maine's lakes. They have a similar long-term data set for phosphorus that goes back 40 years and Jeremy Deeds from the Maine DEP provided me with his data set and it turns out that virtually none of Maine's oligotrophic lakes are increasing in phosphorus. And it's interesting to note that Maine has been requiring setbacks and buffers around its lakes and streams for the past 40 years whereas Vermont only recently passed the shoreline protection law for lakes and still has no stream buffer protection at all in its regulations. And currently we are applying for a grant with the University of Vermont Spatial Analysis Lab to obtain one meter resolution land use data for the Maine oligotrophic lakes to test the hypothesis that our lakes have seen more shoreline and streamline development than Maine's lakes and that might be contributing to the increases in phosphorus in our lakes versus Maine's lakes. So even though the watersheds overall for most of these oligotrophic lakes are largely forested we might be seeing the effects of concentrated development particularly around the streams and lakes in the immediate riparian zones. So that's why we want to use high resolution GIS data to test that hypothesis. Also in the meantime our colleague here at Vermont DEC, Sean Revolato, has used recently available one meter resolution GIS data to look at land use land cover in the 100 foot buffer around 160 Vermont lakes with long term phosphorus data. So he used a data set with 41 lakes that have increasing phosphorus trends and 119 lakes that are stable for phosphorus and he compared the percent cover of buildings in the immediate shoreline of those 160 lakes. And this is the rather complicated slide that shows the results of the statistical analysis he did but the bottom line here is that lakes with increasing phosphorus trends have greater percent cover of buildings within 100 feet of the shore statistically than do lakes that are not increasing phosphorus. So this is another piece of preliminary evidence that makes us think that land use might at least be a contributing factor to the increasing trends that we're seeing in some of our lakes. So luckily we have many best management practices available to address most of the threats to water quality from these shorelands including managing stormwater runoff restoring living shorelands with practices like vegetative swales, rain gardens or simply by creating no-mo zones along our shores. These kinds of shoreline practices range in complexity from the story of living shorelands as here to just or doing something more complicated. This is from a project that was done on like bone-mazine using encapsulated soil lifts to restore shoreland and prevent erosion. Our website provides lots of resources for people interested in what they can do to improve water quality and habitat in their lakes and Amy Peacot is a good person to contact for more information about that. Her email address is there. And so just in summary the Vermont Lake Scorecard summarizes and interprets multiple data sets to help us understand each lake's trends and status. Long-term monitoring data is critical for identifying both our successes and challenges over time. And long-term trends for most nutrient polluted lakes suggest that improvement efforts may be paying off as indicated by those stable or declining trends in eutrophic lakes. But we've also been able to focus in on our very precious oligotrophic lakes that we need to pay more attention to those lakes in the shoreland and watershed management practices for those lakes to reverse the disturbing trends our long-term data has revealed. And I'm ready for questions. Thank you. Anyone online have any questions? I have a question. I don't know if you already said this, but how often does the Lake Scorecard data update it? Every year we initially didn't update it as frequently, but over time I've been able to automate more and more of it and now we're at the point that we are able to update it every year. So the data that I showed today is through 2018 and over the winter I'll update it with the data that we collect this summer for 2019. Steve, is it? Yes, could you speak more to the lead issue and you mentioned for power plants, mercury or whatever. Yes, sorry, mercury. Could you talk a little more about that? So probably most people have heard of acid rain and the deposition of sulfates and nitrates that has caused acidification of lakes that came from the burning of coal in power plants in the Midwest. Well, another thing that happened with that, was those air masses floating over Vermont from coal burning in the Midwest was mercury deposition. There's also mercury contamination that was deposited all over the landscape in Vermont and mercury bio accumulates in fish. So that means as the fish feed and as the bigger fish feed on the smaller fish, mercury concentrations can become high enough in the fish that they're not considered safe for human consumption. And the good news is that the 2004 study didn't find widespread mercury concentrations that would be necessarily dangerous for consumption, although these tests are very expensive and the sample size was very small. But we know that all of our lakes have been impacted by the mercury deposition and that's why we score virtually all the lakes in the state as in fair condition for this, because especially certain fish species that we know accumulate mercury the most could be compromised and people should be cognizant of that. Perry, do you want to say anything more about that? No, I have a different question. Did that answer your question? Yes, thank you. I have a quick related question. Do you think that mercury contamination has greatly increased because it's been a little while since they've... Yeah, so we do think that it would be good to revisit that study because it was done in 2004 and it's been a while. I think that... And I'm really speculating here and I am not a mercury expert at all. I relied on my colleagues for that part of the scoring. But I think it's not likely that it's increased substantially because the deposition from coal plants has declined in general from the implementation of the Clean Air Act regulations that reduced that impact. So whether... How much additional mercury is falling in the landscape? I don't know, but we might still be seeing increases in lakes if continuing inputs are coming off the landscape into lakes. So I guess I don't really know the answer to that question and it's something that we might want to address sometime. My colleague Kelly Merrill and I have talked about the need to repeat that study or something like that study to track how this is going. I was just wondering about the availability of LiDAR data and how that's improving our understanding of the watershed stresses or will it be improving our understanding of the watershed stresses? I think the combination of LiDAR and also the higher resolution land satellite data combined are likely to give the best ability to categorize accurately land use data in the watersheds and all of those types of data sets are becoming more available. In the past we've been somewhat hindered but in our ability to evaluate that near shore zone that we think could be very important for its impact on lake water quality because the land use data available is 30 meter resolution and that means that it's to course a resolution to actually be able to categorize the narrow band of immediate shoreline area that we think is important. So that's the significance of the one meter resolution imagery. Now we can look at that near shore area and get a more accurate quantification of the land use activities in that zone. Does that help? Yes. Are there any other questions? Anyone online? Okay. Well, thank you so much Leslie. If anyone has any questions later feel free to email Leslie. I'll be sending out a link to a survey monkey if I could fill that out and the recording in the presentation they're going to be posted online shortly. Thanks everyone again for attending the final clean water lecture of the season. If you have any suggestions for clean water lectures when we start back up again you can email nr.cleanwaterbt at vermont.gov or just write it in the survey. Thanks.