PubTalk 04/2019 - California's Ecosystems

Video Transcript
Download Video
Right-click and save to download

Detailed Description

Title: The Story of California's Changing Ecosystems: As observed from space

  • How can we use images from space to help us understand changes to our coasts, rangelands, forests and wildlife habitats?
  • How can they help to predict future changes?
  • What more can we learn from advances in earth observing technologies?


Date Taken:

Length: 00:54:16

Location Taken: Menlo Park, CA, US

Video Credits

Amelia Redhill


This video is a one-hour presentation of the USGS Evening Public Lecture Series titled, The Story of California’s Changing Ecosystems as Observed from Space. The presentation is being hosted in the USGS Menlo Park facility. The host welcomes the audience and introduces the speaker, Kristin Byrd, who is a USGS physical scientist. As Kristin is giving her presentation, she is continually pointing to and referring to slides presented on the screen. The slides are a mixture of charts, graphs, and photos. At the end of the presentation, there is a question-and-answer session with members of the audience.

- Good evening. Can you hear me?

- Yes.

- Yep.

- Great. My name is Susan Benjamin. I’m one of the – more? That better? Okay. You tell me. I’m not good with these. I’m not going to be a rock singer anytime. [laughs] I’m one of the managers here with USGS on the campus. And I’m delighted to welcome you all to the public lecture tonight. And also to tell you that we have another one of these planned in a little over a month. If you come back on May 30th, Suzanne Hecker will be talking about new mapping of the Rodgers Creek Fault. It’s longer and more complex than we thought. So that sounds like a great talk. A little housekeeping. In the event of an emergency, a very loud noise will come from the ceiling. And I hope that all of you will go, carefully and quietly, out the back door and down the stairs. And we will hope that won’t happen, but you won’t want to stay in the building if it does. So it’s my pleasure tonight to introduce you to Dr. Kristin Byrd. She’s a research physical scientist at the Western Geographic Science Center here in Menlo Park with expertise in applied landscape ecology and remote sensing. She leads landscape studies of natural and working lands, focusing on wetlands, rangelands, climate, and land use change in California’s Central Valley, and in estuaries throughout the country. She received her Ph.D. in environmental science, policy, and management from UC-Berkeley, her master’s in ecology and systematics from San Francisco State University, and her B.S. in environmental science from Cornell University. Kristin uses remote sensing of wetland vegetation to quantify habitat quality for ecological forecasting and wetland carbon stocks for greenhouse gas inventories. She’s developed scenarios of land use, climate, and hydrologic change to assess potential impacts to ecosystem services in the California rangelands and to identify opportunities for rangeland conservation, increase carbon sequestration and drought resilience. She prioritizes the use of open data and open source software to help aid tool development for decision-makers. And her projects always include outreach to land managers to support conservation and restoration planning. Her talk today is titled, The Story of California’s Changing Ecosystems as Observed from Space.


- Thanks, everyone, for being here today. Yeah, today I’d like to talk about California and its ecosystems and diversity we have here in the state. As you know, California is a really diverse state. It’s actually a biodiversity hotspot, which is – which means that it has high biodiversity, but it’s also threatened by a lot of human activity. And it’s one of 25 biodiversity hotspots in the world. Part of the reason for this diversity is the topographic diversity that we have in the state. And this map here shows the various ecoregions that we have throughout the state, and you’re probably familiar with many of them, including the Coast Range, the Cascades, the Central Valley in the middle, the Basin and Range to the east, and then the Sierra Nevada Mountains. So while we have a lot of diversity, we’ve experienced quite a bit of change of the past 100 to 200 years. We’re the most populous state in the country. And back in 1900, there were only 2 million people here. And by 1950, there were 10 million people. And now we have about 39 million people in the state. And in addition to population growth, our climate has also been changing. Just based on the historical record, we’ve seen air temperatures increasing over time. The years 2014, ’15, ’16, and ’17 were actually the warmest on record. And our state is becoming more dry, and precipitation is more unpredictable. So, in the past 10 years, seven of them have been below-average rainfall compared to the historical average. And, in addition, we have – we’re experiencing increasing levels of sea level rise. Since 1900, sea levels in the San Francisco Bay area have increased by 7 inches. So, on top of that, we – you know, we see a lot of these changes. And a lot of that takes place along – in terms of our land use and how we use the land in California as well, since the 1970s, we’ve seen an increase in development of 38%. It seems like a lot, but you have to remember that the amount of developed land in our cities and our communities across the state really only represent about 4% of the whole land area of the state. But, at the same time, we’ve seen a great loss of our grassland and our shrubland ecosystems as well as our forests due to development in agriculture and things like forest harvesting. But, at the same time, we’ve actually put a lot of investment into conservation. And starting in the 1930s, the state started to really emphasize conservation of state lands where we got a lot of the state parks formed. And, if you look at these maps going from 1970 to the future, you can see an increase in conservation lands throughout the state. In the 1980s, we started to see a lot more investment in federal conservation lands. So if you look especially in the eastern areas of the state, you’ll see larger green areas, representing more conservation lands coming into the state. So, you know, it’s a very dynamic state. We have a lot of changes happening. And, luckily, we have many different kinds of satellites in the sky that we can use to observe these changes and track them over time. This map – this figure here is showing a whole series of satellites that are NASA Earth Science missions. Many of them are existing, and some of them are planned for the future. But what these satellites do – they can measure many different kinds of features, but this is – they’re tracking the reflected and emitted radiation from the land surface at different wavelengths along the electromagnetic spectrum, from visible wavelengths all the way out to radio, to be able to detect different features and the changes on the planet. So, just to give you an example of some of the things that we can see from satellite images, you probably remember the Camp Fire back in the fall. This is an image taken from the MODIS instrument on the Terra satellite showing the extent of the smoke from the Camp Fire. And, because images are collected over repeat intervals, we can see change quite easily. So this is the snow pack in California in 2011, which was a relatively wet year, and then this is 2014, that was one of our extreme drought years. You can see the dramatic change in the snow. And then we can use tools like radar to look at things like subsidence in the San Joaquin Valley – that’s the sinking of our land. This area here shows subsidence in the San Joaquin Valley where you have the bright yellow. And then here – this may seem familiar to you. This is the South Bay Salt Ponds just near to us, where we’ve had a lot of restoration take place. And looking at repeat images, we can see the increase in the tidal marshes –  here, this being pre-restoration, and then out here, we have post-restoration, new marshes being formed. So, as I – as you saw from that slide, there’s a lot of different types of satellites available to track these changes. One of the satellites that I’ve been working with most for my work is the Landsat satellite series. So this is a series of satellites that have been in operation since 1972. It’s a joint USGS-NASA initiative, and it’s the longest continuously acquired space-based moderate-resolution data archive. And what’s great about it is it collects an image at the same place on the ground every two weeks, and it covers the globe every two weeks at a resolution of about 30 meters by 30 meters for a pixel. So you can really detect a lot of different things. And it’s free. And it can be easily downloaded, and the USGS pre-processes it for you, so you can bring it right into your analysis pretty easily. So the Landsat satellite actually has two different sensors on it. One is the Operational Land Imager, and it’s tracking – it’s tracking reflected light from the visible to the near-infrared wavelengths. And then it has another sensor called the Thermal Infrared Sensor, where it’s recording emitted thermal infrared radiation, which can be used for mapping land surface temperature or soil moisture, for an example. Most of the time, I’ve been working with the first sensor, the Operational Land Imager, which is collecting information at seven different sections of the electromagnetic spectrum from the visible to the near-infrared. So it’s basically taking seven different pictures at seven different spots along the spectrum. We have pictures taken from the blue part of the spectrum all the way out to the shortwave infrared region of the spectrum. And it’s really – it’s interesting. We can take these seven different images and combine them in different ways to get a false color image to be able to see certain features that we might not be able to see with the naked eye. So here’s a picture of Crater Lake National Park. And the red is actually showing a lot of vegetation in a false color image. So ... I’ll kind of tell you a little bit more about how we work with this data. Different features can reflect light in different ways across the spectrum, and that causes them to have a unique signature. So something like plants might reflect a little bit of light in the green – because we see them as green – but they reflect a lot of light in the near-infrared part of the spectrum. And then they might – and then, as you go through the spectrum, it’s, you know, different levels of reflectance. And then something like clay, or kaolinite, will have a really high reflectance in the visible part. So this is sort of their spectral signature. But Landsat can see the parts of the spectrum within these certain bands. This is where the red peaks are. And so we’re getting information within these certain sections, and we can analyze that data in a certain way to tell us something about the features that we’re seeing. Like, for example, for plants, we can take the data and learn a lot of things about what kind of plants they are, how they’re growing, are they healthy, are they stressed, how large they are. Like, for example, one thing we’ll do is we’ll develop a vegetation index. A real common one is called the Normalized Difference Vegetation Index. It’s the ratio of reflected light in the near-infrared and red regions. And it can tell us how healthy the plant is. If it’s a high value, it’s a really healthy green – large biomass, large growth. If it’s stressed, it’ll have a lower value. So, because this remote sensing data is collected at repeat intervals, having a long historical record, we can use it for a lot of different kinds of analyses. So we can use it for, you know, map – just mapping features, creating maps, but also quantifying how much of something is on the ground. We can do historical change detection to see how much things have changed in the past. We can even use it in models to model environmental change and then eventually forecast change into the future and predict what may happen. So I’m going to be talking about some of these applications as I go through the talk. So I have actually been working a lot in wetlands in California since early 2000s. Worked in different wetlands across the state. And so my talk is going to focus on a couple of different examples of using remote sensing in different wetlands and how they’ve been changing. I’ve mostly been working in marshes. And so marshes are really characteristic of having herbaceous vegetation that grows above the water surface. In the bay – the San Francisco Bay here, we have marshes that are different based on their salinity levels. So we have tidal marshes – either saline or brackish or fresh. We can have non-tidal freshwater marshes. And then, as you go further east into the Central Valley, we have marshes as well, but a lot of them are really highly managed. They have been a really important part of California’s landscape over its history, but they’ve experienced quite a bit of change over the past 200 years. So here’s a map of the Central Valley before 1900 and kind of around present day. And you can see that, you know, before the 1900s, the Central Valley actually had extensive wetlands running all the way through it from north to south. As long as a lot of grasslands, but, as you know, the Central Valley has been very transformed, and we’ve lost about 90% of those wetlands. There’s a group called the San Francisco Estuary Institute that’s done some really amazing work on historical ecology, and they’ve pulled together many data sets and photos and maps to recreate what San Francisco Bay looked like 200 years ago. And the green here in the map represents tidal marsh. And you can see, back before a lot of European development, we had extensive marshes all around the bay – North Bay, this is Suisun Marsh, and then in the South Bay where we are now. And that’s converted quite a bit, where we have a lot of agriculture in the North Bay now, a lot of development kind of in the Central Bay, and then, where we are now, a lot of that marsh got converted to the salt ponds you’ve probably seen. San Francisco Estuary Institute did a similar project for the delta. This is going further east. This is the San Francisco-San Joaquin Delta. And this area really experienced dramatic change. After the Gold Rush, the people formed levees around marshes to drain them and convert them to agriculture. And so, as a result, we’ve lost 98% of our marshes in this area. So, despite the losses we’ve seen in wetlands, they still remain a highly valuable ecosystem for the state as well as the people that live there. And they have a lot of benefits, and those benefits are what we would call ecosystem services. So, just to name a few, they’re really beneficial as a – as a nursery for juvenile wildlife. They serve a purpose of helping with flood control. And also filtering stormwater. They can help reduce the impacts of storms, serving as a buffer. Really important for recreation and tourism. Really important sinks of – for carbon. Can be a source of employment and jobs. And also can help reduce the impacts of sea level rise. So just focusing on one of those benefits – carbon storage. Coastal wetlands especially, and coastal tidal marshes, have a really high capacity to store a lot of carbon in their soils and in their vegetation. And they can store carbon at very fast rates, called carbon sequestration. They can actually store carbon faster than forests can – two to four times faster on a per-area basis. So, because of their high capacity to store carbon, they’ve kind of gotten the name of coastal wetland blue carbon. And people have been very interested in tracking their capacity to do this as a way of – and one way of looking at it is to incentivize wetland restoration. Because this would be a benefit that we would get out of it. So, because of this coastal wetland blue carbon and how important it is, it’s now included in our National Greenhouse Gas Inventory that the EPA does every year that gets reported to the U.N. And so this inventory tracks all of our emissions that we have in the country from all of our different industrial sectors as well as our removal. So our removals are the – how we are taking carbon out of the atmosphere and sequestering it into the land base. So wetlands play a role in that. So that’s now part of the inventory. And the way that they’re able to include this – or, conduct this inventory for coastal wetlands is to looking at changes in carbon stocks from one year to the next. Because a change from one year to the next would indicate a removal of carbon from the atmosphere and into the land. And typically, they’ll track changes in carbon stocks across five different pools here, like above-ground biomass, dead wood, litter, soil organic matter. And I’ve been working on a project recently really focusing on tracking the above-ground biomass. The biomass is sort of the volume of the vegetation – how large it is, how much it weighs, how much is growing, and then the carbon that’s stored within that vegetation. So, as part of that project, I was able to use the Landsat satellites to be able to map the biomass of all the tidal marshes in San Francisco Bay and understand the conversion of the weight of the biomass to carbon content, come up with a map of tidal marsh above-ground carbon stocks. And so, as you might know, the San Francisco Bay is more saline. Closer to the Golden Gate, it becomes more fresh as you go east. And that really influences biomass and carbon. So you get higher carbon stocks as you go further east into the fresher waters. What’s pretty neat, though, is we’re able to quantify that, within this region of the bay, just within the above-ground part of the vegetation, we have over 32,000 megagrams of carbon. And if you convert that to the CO2 equivalent, you actually get the equivalent of 25,000 cars annually on the road, which is pretty neat. So you can see how the tidal marshes are offsetting those kinds of emissions. And just to give you an idea of how we actually go about making this map, a lot of – there’s a lot of work on the ground that goes into making the map. We have had a lot of people that we’ve worked with over the years that have been collecting a lot of field data in tidal marshes around the bay on biomass. And so what they do is they go out in the field, and they collect a sample of biomass. And they also take a GPS reading at the same time to be able to register it on the ground. And then they match up that biomass data with the Landsat reflectance that was taken at the same location to create a data set. And the data set can be put into a machine learning algorithm. We use something called Random Forest in order to make predictions. And then we can predict what the biomass and the carbon might be in places where we didn’t have data. And by doing that, we can – we can make a map. So that’s sort of the – in a nutshell, how we – how we go about doing that. So looking a little bit closer in the San Francisco Bay, specifically in Suisun Marsh just east of us, we did a focus study on a specific wetland called Rush Ranch. And this is part of the San Francisco Bay Estuary Research Reserve. It’s a reference marsh. It’s one of the mature marshes in Suisun Marsh that haven’t really been very impacted by people over time. And what we were really interested in is understanding how marsh habitat might change in the future with sea level rise. So there’s a lot of models available to be able to calculate this and forecast what might happen to the marshes. But what we wanted to do was see if we could run that model with remote sensing data so we can get a regional perspective of what the change might be as opposed to just in one specific location. So what’s really interesting, I think, about tidal marshes especially, is that they have different habitats within them, but their habitat changes are really tightly tied to the small changes in elevation within the marsh. So in Rush Ranch, even just an elevation change of 1-1/2 meters, you can go from a mudflat to an upland. You have a couple different marsh habitats in between that are really different according to their – to their vegetation. And marshes do have the capacity to gain their elevation as sea level rises in a gradual sort of background rate. And they’ve been doing that for millennia. So, in a way, they can keep up with sea level rise. But if the rates of sea level rise increases too much, then marshes may or may not able to keep pace. And they may end up drowning as a result. And a couple things that influence this are what the baseline elevation is of the marsh, the plant biomass – how much is growing there, and also the suspended sediment concentrations in the water. Because the suspended sediments are being deposited on the marsh, and they’re helping to build up that soil. So, for our project, we again used Landsat. We were able to map the suspended sediment concentrations in the channels. So it’s pretty neat to see sort of the gradient of the – of the sediments as you move up through the channels and into Rush Ranch. We had a neat application that we used as part of the project. It was – it was an app on the iPhone that you can – you can download called HydroColor. And you can actually measure the concentration of the sediments in the water by taking a picture of it. So we were able to use that as, like, a way to calibrate and create our maps. And, again, because Landsat takes a picture, like, every two weeks, we could get a times series across the year of how that sediment changed and get an annual average, which was basically what we needed for our model and for our purposes. So what we did is we wanted to work with a model – it’s called the Marsh Equilibrium Model. It’s a very detailed process-based model that can forecast how the marsh elevation will change gradually with sea level rise. And we provided input data from our – from our satellite. So we had input data on above-ground biomass as well as suspended sediment concentrations and used that to look at what changes might occur in Rush Ranch. Another benefit of having the remote sensing data is so we can run the model in places where we weren’t even able to get to to go collect data in the field. We can just get the data from the remote sensing. So here’s an example of some results. So basically, this is sort of present-day Rush Ranch. The marsh is a high marsh habitat. It has sort of a typical vegetation of a lot of salt grass. And there’s a couple other marshes in the Suisun Marsh that we tested as well. And, after about 100 years with a projected 1-meter sea level rise, we saw that the habitat did change. And, as there was a sort of a small sinking relative to the sea level, the vegetation changed to a low marsh habitat. So it would be a lot of – in this case, a lot of bulrushes and tules. And similar things occurred in the other marshes, too, and in some cases, even might get converted to mudflat. Another thing we can do with this kind of information is tie it with information about specific species and habitat – their habitat requirements. So a couple of endangered species in the marsh – so changes – actually California Ridgway rail and the salt marsh harvest mouse need high-tide refugia to survive. And the model can help us to track where that high-tide refugia in the marsh goes over time with sea level rise. It actually moves landward. And this is something that we’ve been – we’re learning more about is that, as sea levels rise, marshes will eventually start to migrate landward up to higher elevations. And conservationists are paying attention to this and realize that we need to set aside the land adjacent to marshes as open space so that we can allow that migration to occur and allow marshes to still persist over time. So moving back to the Central Valley – I showed this a few minutes ago, that the Central Valley has been really greatly transformed by the conversion of the wetlands and the grasslands to agriculture. This is a really great book – The Fall and Rise of Wetlands of California’s Great Central Valley. It talks about how, you know, we lost a lot of our wetlands there. Over 90% are lost by converting them to farming. But, at the same time, in this region, as well as the bay region, is a really important stopover in the Pacific Flyway for millions of shorebirds and waterfowl, including ducks and geese. They come through in the wintertime to forage and rest as part of their migration. And what’s really interesting is that, you know, back in – like, so the early 1900s, the Sacramento Valley in the north, especially, got converted mainly to rice fields, and almost all the wetlands were gone. But the ducks and the geese still came. And, because it didn’t have any wetland habitat, they ended foraging in the rice fields. And really having a large impact on the rice crop and really upsetting the farmers there. And so, because of that, they realized they had to restore the wetlands to divert the ducks and the geese from their fields back to an area where they wouldn’t be impacting them as much. So because of that, in the 1930s, there was a real push to create a network of wildlife refuges all across the Central Valley to restore wetlands and to have habitat for these – for these birds. So now, this is what the landscape looks like. It’s a – it’s mostly agricultural land. Different agricultural crops are grown here. Rice and corn are very common, especially in the north. And then we have, you know, wetlands sort of interspersed. The blue – the blue here shows the wetlands. And the wetlands are kind of a mix of public wildlife refuges and also private land. There’s a lot of duck clubs throughout the valley that provide important habitat for the migratory birds. So, again, this is another close-up of those sort of habitats that the birds use in the – in the – in the Central Valley. And they’re really highly managed. So the croplands – the croplands and the wetlands are actually flooded up in the wintertime to provide aquatic habitat for the ducks and the geese when they arrive in the migratory path. And we’ve been working with a group called Point Blue Conservation Science and have been tracking the flooding up of these wetlands and croplands using satellite imagery – Landsat again – and tracking how that surface water habitat is changing over time. I might mention that the water that gets used to flood these wetlands and croplands are all through, you know, management of our water resources and allocations of water to the refuges and to the farms for the different state and federal water projects that we have. So I just want to share – this is a pretty cool animation showing the change in our surface water across the Central Valley. This was done by my colleague, Matt Reiter, at Point Blue Conservation Science. So you can see – this is all managed. This is all, you know, done by people. And that, in the summertime, there’s not a lot of water on the landscape, but as we move into the fall time, they start to flood up the wetlands and the croplands, and you see a lot more area flooded. That’s the blue. And this is an interesting area to look at. This is called the Grasslands Ecological Area. It’s down by Los Banos, and so I want to – I’ll talk a little bit more about that in a minute. So taking this information about flooded wetlands and croplands and how important they are for wildlife, we’ve been working with Point Blue on a large project to take this remote sensing data and use it to forecast habitats for migratory birds and – the birds themselves, and use that to manage water use across the valley to support both the wildlife but also other beneficial uses of the water, like groundwater recharge. And so we’re doing both short-term and long-term projections of these habitats and the species distributions, and we’re hoping to use it in a way to support coordinated conservation for wildlife across the valley. So to do our projections and our forecasts of habitats, we’re using remote sensing data. And this evening, I just want to talk really briefly about a couple of different remote sensing data sets that we have been developing as part of – as part of this project. And that would be moist soil seed plant distributions – their coverage and their yield – their seed yield or production. So you might wonder what moist soil seed plants are. So, again, these managed wetlands are flooded in the wintertime, and they’re drained in the spring, intentionally to grow a certain kind of plant – these moist soil seed plants that are high in nutritional quality. And they serve an important food resource for the migratory birds that are coming through. They’re actually not native plants, but because of their high nutritional value, they’re really prized as, you know, a food resource for the waterfowl. But, across the whole valley, there really isn’t a lot of information about how much is really grown, but people do need to know that information to understand how many – what populations of waterfowl they can support across the entire valley. So what we did is mapped all the managed wetlands in the Central Valley using, again, Landsat satellite data. And we focused on mapping two main types of moist soil seed plants – the swamp timothy and the watergrass. And we created maps across a whole time series of 10 years. And, again, I just want to emphasize, it’s really interesting that the growth of these plants are – it’s almost like a – it’s like a source – kind of a farming, where the managers, you know, drain the land and then they plant seeds and they manage it and really try to boost up their yields as much as possible so they can provide the best quality habitat for the wildlife. So, in order to make our maps, we had – like I mentioned with the – with the blue carbon stuff, we had to go out in the field and collect a lot of data to build our models. And, in 2017, I had a field team – mostly Austen Lorenz and James Anderson, who spent the entire summer crossing the valley, from northern Sacramento Valley down to the bottom of San Joaquin Valley, collecting field data on the plant locations and their biomass and their distributions. And we were able to use this to build our models. And just had to give them – because working out in the middle of July and August in the Central Valley, really high temperatures – 113 sometimes, so you really had to be careful and manage your – manage your days well. So just to give you a little bit of information about how we were able to do our mapping, we used something called phenology metrics in order to tell these different species apart in our maps. And so phenology is sort of the study of the pattern of growth over the course of the year of different things. And our plants had very different phenology patterns, which we could use to tell them apart. So the watergrass – this is a map – this is a figure showing days of the year from spring to fall, and then this is sort of their growth – their growth pattern or their level of greenness that we see from the satellite. And so, in watergrass, it starts off kind of slow, and they get a peak greenness kind of around mid-July. And then turn brown by about late August. Where swamp timothy becomes really green in, like, say early May, and then it becomes brown really quickly – like, around June or July. So this is the kind of information that we can use to create our maps. And, in doing so, we ended up with a pretty highly accurate map with accuracies of over 88%, which, in the remote sensing world, is actually pretty good. And then, specifically for swamp timothy, we were able to create a model of its productivity and its seed yield. And we did that just using some linear regression models and taking those vegetation indices that we can pull from Landsat in order to create a model of the productivity. And then, with those predictions, then predict productivity in different parts of the region and then create a map. So, just to talk a little bit about what we found, I want to focus in on that Grassland Ecological Area again in Los Banos in the San Joaquin Valley. It grows a lot of swamp timothy. It’s actually a really important wetland area in the Central Valley. It encompasses about 180,000 acres of wetlands, and that makes it the largest block of contiguous wetland habitat remaining in California. Even more so than San Francisco Bay. So here’s an example vegetation map that we were able to produce. And the bright green area shows all the swamp timothy. The purple is emergent vegetation, like a lot of bulrushes, cattails, and that sort of thing. Not a lot of watergrass here. And then this map shows how productive that swamp timothy is for 2017. The lighter areas are showing more highly productive areas. And there’s a close-up of a – of a wetland where there’s higher production kind of on the edges of the wetland pond on the inside. So, because we were able to do a time series analysis and get maps going back to 2007, which encompass that really critical drought period of 2013 to 2016, we could understand how water availability affected our capacity to grow this type of vegetation that’s used for supporting our waterfowl. And what we found was, on those critical drought years, which is the vertical lines here, like 2014, 2015, we had a lot more swamp timothy growing in the valley as opposed to watergrass. And the reason for that is swamp timothy does not require summer irrigation. And watergrass does, and the managers at these wildlife refuges just were not receiving the water allocations from the water projects that they typically were, and so they had to change their management practices and influence what they were able to grow. Then, as you can imagine, with less water, they had less productivity, so they weren’t able to produce as high of a seed yield as in the wetter years. So we really want to look into this a little bit more to understand how it affects the capacity for the valley to support waterfowl populations during these periods of drought. So going back and working with Point Blue, what they were able to do was take these maps we made of surface water and of this vegetation, and then can put it into a model to predict the species distributions of a number of different birds species. Here’s showing examples of four different duck species and their sort of patterns of where they’re most likely to be found across the valley. So we have pintail, green-winged teal, shoveler, and mallard, as an example. And what you might see here is the mallard is kind of the generalist. It can be found all over the place – kind of all distributed across the valley. Where the pintail and the green-winged teal are actually much more dependent on those flooded habitats, so have much smaller ranges and distributions. And if you look kind of a little bit further into the model, you can understand why that is. Certain variables are more important in describing where some species are found versus others. And, again, things like the pintail and the green-winged teal were very dependent upon the availability of flooded wetlands, while the mallards were not. They could be found in many other different kinds of habitats. And maybe you’ve observed that because mallards are actually pretty commonly found throughout our communities. So looking into the future, what kind of satellite might we expect to see in the 10 years or so? Right now, NASA is working on a new type of satellite called a hyperspectral satellite called the Surface Biology and Geology Designated Observable. And, instead of having seven bands or nine bands, it might have, like, 100 or more. So taking snapshots at many other wavelengths along that electromagnetic spectrum. And so you have that much more detailed information coming in. You can get a lot more information about the features on the Earth. You can get information about biodiversity, species composition, biochemistry, the nutrient content of vegetation. Much more information about plant stress and canopy water content. Even mapping minerals. So we’re working with this kind of data in a new project that’s just getting started. And this is taking place really close to here in our neighborhood in the South Bay salt ponds region. So you might know that we have a very large restoration project ongoing in the South Bay area here – 15,000 acres, which is the largest tidal wetland restoration project in the western U.S. And you hear a lot about the tidal marshes being restored, but you might not know about the mudflats and the importance of those habitats. The mudflats that grow and are present just on the outskirts of the marshes are really important, again, for foraging, for shorebird habitat. And they have something on them called biofilm, which is a – sort of a photosynthetic green substance, and it’s a kind of a mix of microbial bacteria and diatoms and protozoa. But they’re really important in the – in the diet of shorebirds. And so we’re going to be using hyperspectral data to be able to quantify this biofilm and understand its nutritional quality to see how important the mudflats are for the – for the shorebirds. So that’s a new project. And hopefully we’ll have something to report on that in a – in a year or so. So just to end, I really want to acknowledge all of my co-authors and collaborators. A lot of this work was done in partnership with many people here at the USGS and in other organizations in the bay. And also wanting to acknowledge my funding, both through USGS and NASA. But I’ll take some questions.


- If you have questions, it would help if you used the microphone so that everyone can hear.

- Yeah?

- Recently, I’ve been reading and hearing that there’s a effort to roll back some of the regulations governing wetlands, particularly the temporary or vernal ones here.

- Oh, right.

- And how does that affect this? Or does it? And is it a major concern if they do roll it back?

- Sorry. Are you referring for the San Francisco Bay wetlands?

- Pardon?

- Are you talking about the San Francisco Bay restoration or – in particular or …

- No, just generally. You know, wetlands and that.

- Yeah. I’m not as familiar with some of those regulations on sort of the seasonal sort of, you know, ephemeral wetlands. But around San Francisco Bay, because they’re connected to the navigable bodies of water, I think they’re still protected under the Clean Water Act. So I feel – I feel that there’s just a lot of – and you’ve probably seen, you know, many bonds being passed in San Francisco Bay to support wetland restoration. I feel like there’s a lot of momentum.

- So you mentioned, with rising seawater, how we need to preserve land beyond existing marshes. Are there any maps showing what that estimated land distance is to – that needs to be preserved so that marshes can keep recreating?

- That’s a good question. Yeah. It tells – it determines a little bit on the topography of that upland land area. So if it’s a really, really sort of flat, open space, that would provide an opportunity for marshes to migrate further in because just very small changes in elevation that allow them to. Because if it’s a really steep upland area, then they would kind of get stuck.

- Trapped.

- They wouldn’t – trapped. They wouldn’t be able to do much. But I haven’t actually seen – there may be one – you know, a Bay Area-wide map of those areas. I think it’s – I’m sure people are working on it, but I haven’t seen a final product.

- There are several wastewater treatment plants around the bay, and they discharge. How does that affect – or are they trying to work that into the map here? I don’t know.

- Oh, yeah. I haven’t – I haven’t really worked with that much here at all. But I do know that, like in Sacramento, they’re upgrading their wastewater treatment plant. It’s going to be changing in the kinds of – or, reducing the amount of nutrient inputs into the delta, at least. And so there’s a lot of studies to see how that might change the ecosystem. I don’t know what the results are going to be, though.

- Just curious about whether there are also other dimensions you can look at in the – in the light. Perhaps the polarization or things like that?

- Oh, yeah. Yeah. So, with radar, polarization is actually a really important tool. So different – there’s different kinds of ways of – to get different kind of returns of – polarized returns. And that can be used to detect different kinds of features. So, like a vertical-horizontal polarization would be useful for looking at tracking wetlands, for example.

- Mm-hmm. And then specular versus, perhaps, more scattered, diffused light? Is that something they can do also, or …

- Yeah. In terms of the different surfaces and the type of light?

- Yeah, for example, surface of water is more of a specular …

- Oh, yeah. Actually, in terms of that, that becomes actually more of a problem than a – they try to take images of water at certain times of day to remove that glint and that – and to be able to actually detect the features in the water they really want to get at, like, say, chlorophyll or suspended sediment.

- Okay.

- So it looked like you had, in the future, coming up with Landsat 9, I think?

- Yep. Mm-hmm.

- And that’ll give you, what, higher resolution?

- I think they’re working on it.

- Not so much higher resolution, but the continued ability to keep imaging the Earth every two weeks. And hopefully more often than that. It’s the – to have that continuous coverage of everything from 1970 on into the future.

- Oh, are you implying that Landsat 8 is already dead?

- No. It’s still going. Yeah.

- No. Eight is good, but 1 through 5 and 7 are not doing so well.

- Oh, I see. Thank you. And then, one of your slides, it looked like – I work at NASA.

- Oh, okay.

- And one of the slides looked like you had one of our ER-2s. Are you doing imaging from …

- Oh. Yeah. Back in the Rush Ranch project, we did use some – well, we actually used [inaudible] data to try it out for biomass mapping. And we had some PRISM data. I don’t know if that was flown with ER-2 or not. But, yeah, we used that – we had another model of the suspended sediment based on the PRISM data too.

- Okay, yeah. Because we have, like, ER-2s down in Armstrong. I thought maybe you might use them for higher resolution imaging as well.

- Yeah. Yeah. Mm-hmm.

- Good.

- Yeah.

- Thank you.

- This is obviously an enormous amount of data. And we don’t collect data just for the sake of data. So I would ask, what, in the last 10 years or whatever, have been real-world policy changes as a result of this? What are the biggest wins in successful change of policy that you’ve seen as a result of the data collection effort?

- That’s a good question. I think that’s something that, at USGS, we are constantly striving towards is to make research relevant. And I think a couple of examples that I have here is with the coastal wetland blue carbon maps, we’re able to improve our ability to do greenhouse gas accounting of coastal wetlands on the national level. So that gets recorded directly into out national greenhouse gas accounting. And it gets reported to the U.N. You know, it’s a small amount. It’s a small increase. But I think that increasing our ability to quantify these things at any level is an improvement.

- Yeah. Well, that’s better accounting, but where is the policy change?

- Yeah. Yeah, and then …

- That what – what are they doing differently as a result of better accounting?

- Yeah. And so …

- What I’m not seeing here is how this culminates in difference in policy.

- Yeah. Yeah. So, in terms of, I think, on-the-ground policy, coastal wetland blue carbon accounting is being used to incentivize restoration because there are some new voluntary carbon markets that are variable so that people can actually get paid to do wetland restoration as a result of improved accounting. And so, if they can get paid to do the restoration, it’s more incentive. And so hopefully more restoration occurs. It becomes more feasible. And then we have all the improved benefits that we get out of wetlands for people and wildlife. And then the other example is the work we’ve been doing in the Central Valley. And so it’s a new project, and we’ve really been working with a lot of managers across the valley to talk about applications. There’s a lot of interest. But it looks like – a couple different things. On a valley-wide scale, we can use it to run bioenergetics models and understand what kind of populations of waterfowl can be supported across the valley and how that might change in a drought year, which might incentivize managers to hopefully, you know, find ways to increase yield so they can support wildlife. And then also, on sort of a smaller refuge basis, on a wetland-scale basis, understanding the variability and productivity across time and seeing which wetlands may be not performing as well as others. Because they – you know, a lot of our refuges don’t have large amounts of stock, and it’s very difficult to monitor and track things on a very detailed level. And so the remote sensing data can help them to do that, and then to be able to focus their efforts on, like, do I need to, you know, restore this particular wetland here to increase my yield? So kind of that targeted assistance with management is kind of what we’re aiming for. Anything else? Oh, one more, I think.

- First of all, I want to – I want to commend you for an excellent presentation. Thank you very much.

- Oh, thank you.

- This is a little to the side, but I’m curious. I was in the satellite and missile business in my career.

- Oh, cool.

- And I’m curious as to what contractor is making this new satellite for you? And is it the same one that made the previous ones?

- Oh. I don’t think that that has been worked out yet. They’re just trying to – they’re in the very beginning stages to talking about the new hyperspectral. They’re really working on figuring out the requirements. And there’s a lot working – trying how to figure out, you know, what do we need? How much, you know, repeat intervals or resolution – that kind of thing.

- I see. Any particular contractor working with you on developing the criteria?

- I don’t know. I wouldn’t get involved with that, so … [laughs]

- Thank you.

- Anything else? Okay. Well, thanks very much.


- Thank you all for coming.