[00:00:00] Bridget Scanlon: Welcome to the Water Resources Podcast. I am Bridget Scanlon. In this podcast, we discuss water challenges with leading experts, including topics on extreme climate events, over exploitation, and potential solutions towards more sustainable management. Today I would like to welcome Matt Rodell to the Water Resources Podcast.
Matt is the Deputy Director of Earth Science for the Hydrosphere, Biosphere, and Geophysics at NASA Goddard Space Flight Center. And prior to this position, Matt served as the Chief of the Hydrological Sciences Lab. for many years. He has received numerous awards, the most recent being elected as an AG, American Geophysical Union Fellow in 2022.
And I've known Matt since he was a PhD student here at UT in the late 90s. Thank you so much, Matt, for joining me today.
[00:00:57] Matt Rodell: Thanks for having me, Bridget. It's a pleasure to talk with you again.
[00:01:01] Bridget Scanlon: Right. So today, I think we're going to focus the discussion on Matt's work related to the GRACE satellite data.
GRACE stands for Gravity Recovering Climate Experiment, and Matt will explain how it works and for monitoring terrestrial water storage globally since 2002. And then Matt also links it to global modeling. And Matt was one of the first people to describe how the satellites and his PhD work looked as what we would expect from the satellites before they were launched. So today we will talk about monitoring terrestrial water storage changes, how total water storage anomalies are used in hydrology to monitor extremes, droughts and floods operationally with the U. S. Drought Monitor, and how we can use these data for water resources management.
And finally, we'd like to discuss a little bit about what he sees the future of the next GRACE mission. So I know that's quite a bit to cover, Matt, but appreciate your filling us in on some of these issues. So maybe first talk about how GRACE has really revolutionized our understanding of the global water cycle.
It's very easy for us to think that we always had global data and we always had these global coverages, but we didn't. And it really was a game changer in hydrology. And maybe you can explain what it was like originally, and a lot of hydrologists were very skeptical of GRACE data. So maybe you can describe what GRACE does, how it works, and then some things about the initial, the inception of GRACE.
[00:02:36] Matt Rodell: Yeah, so people throw the term unique around quite a bit, but the GRACE method of observing terrestrial water storage really is unique. Most satellite observations we have are looking down at the earth and making measurements of either reflected or emitted wavelengths in the EM spectrum and different wavelengths are valuable for different types of observations.
But you're basically limited in that if you're studying hydrology, you can't measure anything below the first few centimeters of the ground surface when you're using that sort of traditional remote sensing measurement. So the idea behind GRACE, and it was really, it was conceived as a geodesy mission for measuring the Earth's gravity field, which is important for many other things like predicting satellite orbits and that sort of thing.
So the Earth's gravity field is not uniform. Actually, if you go to the top of Mount Everest, you're going to weigh a little bit more than you did standing near the Dead Sea or something like that, enough that you'd feel it. But these variations in the gravity field around the Earth are enough to cause predictable perturbations to the orbits of satellites.
And so knowing this, geodesists figured out that the best way that they could measure and basically map Earth's gravity field would be to fly satellites and have two satellites, one basically observing the other. So that's what GRACE is that it's unlike most missions, one satellite with all the instruments on it, it's actually two identical satellites orbiting the earth in tandem.
They're about 200 kilometers apart on average. And at the initial altitude at about 500 kilometers above the earth, and they have a near polar orbit and over the course of about a month, they provide enough coverage to produce a map of the Earth's gravity field. The key measurement is actually the distance between the two satellites.
So the distance between the two satellites is going to vary for these large variations of the gravity field, things like where there are mountain ranges or ocean trenches, that kind of thing. So many. meters, tens of meters, perhaps, of change in the distance between the satellites. But for month-to-month variations in the gravity field, and this is where it gets interesting for hydrology, we're talking about sub millimeter scale variations in the distance between the satellites. And they're able to measure those tiny distances because the original GRACE had a microwave ranging system that every five seconds would make a measurement of that 200 kilometers, and you're measuring that down to the size of about this red blood cell GRACE Follow On, which is the newer mission, which launched in 2018 also has a laser ranging system, which provides even finer or more precise measurements of that distance. So every 5 seconds, you're measuring that distance, which tells you something about perturbations, the orbits of the satellites relates to the gravity field.
You also know the exact locations of the satellites based on GPS information. There's sort of over the course of a month that you put all these data into a supercomputer, which then generates a map of Earth's gravity field. And then, the measurements are so precise that from month to month, we see changes in that gravity field that are then related to changes in mass at the land surface.
And so those changes in mass are things like ocean circulation and tides, atmospheric circulation, and then over the land, changes in terrestrial water storage, which is the sum of all the ground water, soil moisture, snow, and surface waters and ice.
So if you can imagine if you had a big snowstorm and you had a foot of snow in the ground over a large area, that's a huge amount of mass. And it's enough to actually change the orbits of satellites. And that's the concept behind GRACE. And so we have, from GRACE, we have these monthly maps of water storage anomalies. So by anomaly, I mean a departure from the long term mean. Can't tell us anything about... the total amount of water stored in an aquifer.
People often ask me, what's going on with my aquifer? How much water is left in the High Plains aquifer? That kind of thing. And, and GRACE can't answer that. It only tells us the relative changes. So it would be like, if your bath tub were half full and you had some water and you saw the water level go up, maybe you don't know the total amount of water in there, it went up by an inch, that's sort of what you get from GRACE and GRACE Follow On.
So it's completely different from the other types of observations we have because again it can, it senses mass changes at the surface and at depth. So deep aquifers you can measure from space now. You couldn't do that with any other type of satellite observation. And so we've never had this type of data before, but you mentioned that there was some skepticism among hydrologists. That's absolutely true. It took a long time for a hydrologist to really embrace GRACE and for the beginning Jay Famiglietti and some only hydrologists who cared about and ones were dismissive. Oh, well, it's spatial resolutions way too low. And I don't understand what this means. People wanted to have things like soil moisture in the top five centimeters every day or every hour or whatever.
And GRACE can't provide that type of precise measurement, but what was again different was there's no other observation for telling us what's happening in aquifers around the world, for example.
So once people began to understand what it was giving us, yeah, I mean, yeah, the spatial resolution is coarse. You're talking about maybe 150,000 square kilometers is about the smallest area. That's for the size of the state of Illinois. Whereas most remote sensing systems are giving us something on the order of kilometers or tens of kilometers in terms of spatial resolution. So low spatial resolution, low temporal resolution, it's monthly and there's data latency associated with it.
So you might not get today's data until three months from now or something like that. But we found ways to use it and now it's really one of the most popular remote sensing systems for hydrology and was recommended in the most recent National Academy of Sciences Decadal Survey for another mission, which I guess we'll talk about later.
[00:08:16] Bridget Scanlon: Right. So that was a great description of what the satellites do, and I think there was a comic Tom and Jerry to describe these two satellites following each other and measuring the distance between them then to get variations in Earth's gravity, and those being controlled, monthly variations in Earth's gravity being controlled primarily by variations in water storage because water is so heavy. And so you could track then droughts and flood periods and other extremes and groundwater depletion and other things globally. And even in regions where we have very little data or we have very little literature like Russia and many other regions are where they don't share the data.
So it's really nice. And hydrologists initially were used to doing field scale studies and we're used to borehole data and things like that. And so it was really a stretch to look at the large scale. But now I think over time begin to realize the value of the large scale. I mean, when you want to look at the impacts of continental water storage changes on sea level rise or other things, I mean, that's really the forte of GRACE.
And so this continuous measurements over a global scale were amazing. And so terrestrial water storage. We weren't really accustomed to that. And as you say, we want to know soil moisture, we know ground water, whatever. But I think the value of terrestrial water storage as a parameter in itself is becoming increasingly recognized.
And it was an essential climate variable by the Global Climate Observing System last year, one of 54 different variables. And so I think it's becoming important in its own right, the total water storage, terrestrial total water storage. And maybe you can, my feeling is that we should look at those data first before we start jumping the gun to say, is it occurring in reservoirs or is it soil moisture or is it groundwater?
But to really just look at the trends in terrestrial water storage. Any comments on that? It seemed like your Nature 2018 paper, where you looked at trends, that's what you emphasized was trends in terrestrial water storage anomalies.
[00:10:26] Matt Rodell: Yeah, I think that's right. I think, especially as a climate indicator, it's really valuable as sort of a holistic indicator of what's happening with the water cycle.
So like you said, it, because it integrates all the water storages. It tells us a lot about sort of these long-term changes, whereas something like the surface waters or soil moisture is going to respond pretty quickly to atmospheric phenomena. The total terrestrial water storage, especially because it includes groundwater, is really going to integrate over a longer period and tell us more about inter annual variations in the water cycle and longer, so it could be longer term trends.
And so in the 2018 paper that you mentioned, we basically looked at the trends in terrestrial water storage from the start of GRACE in 2002 up through 2016. And so each location on Earth, we basically remove the seasonal cycle and then fit a linear trend and then try to explain why this trend was happening.
And obviously, we call them emerging trends because we don't know which ones are going to continue. So some people would say that's not really a trend. It's just sort of inter annual variation. But clearly, if we look at them, and take into account auxiliary information that we have, we can sort of separate them into three bins that that would be (1) natural variability, whatever trend we're seeing is just related to natural inter annual variability, things like El Nino Southern Oscillation, how that changes over time.
So maybe there's a region that happened to be in a drought in the beginning of the time period and happen to be wet at the end. And so we end up with this increasing terrestrial water storage, but it doesn't really mean that it's going to increase forever. So there's that natural variability.
(2) Then there's the climate change impacts. Those are really the hardest ones to identify. Some of them are really easy. So like the ice sheets and glaciers, the Greenland ice sheet, the Antarctic ice sheet, those are some of the biggest initial results from GRACE. We're being able to finally estimate the rate of ice loss from these huge ice sheets.
And if you think about it, that's one of the key variables we have to know impacts of climate change. What's global warming going to do to the planet? That and the sea level rise, which is also measured by GRACE. But then there are other regions where maybe climate change is causing precipitation to increase or causing it to decrease. And over time, that's going to cause increases or decreases in terrestrial water storage. And we may be seeing the beginnings of that in some places, but again, really hard to say for sure. The best way we can sort of try to understand that is to look at the predictions of climate models and see where they align with what's actually happening with the terrestrial water storage.
(3) And then the third bin is direct human impacts. So that would be things like groundwaterpumping in northern India where they pump a huge amount of water out of the ground and then a lot of it's used for rice paddies and wheat and most of the water evaporates faster than it can recharge and so you have a long-term trend of aquifer depletion in that area.
Other places we've seen actual increases associated with direct human impacts and that would be things like filling of reservoirs in parts of China. The Three Gorges Dam is huge. And when they filled that reservoir and ones around it, we could actually see an increase in terrestrial water storage in that part of China, which is pretty amazing.
[00:13:35] Bridget Scanlon: So I really liked that paper because I felt like it was comprehensive and you tried to attribute to the causes of these changes in different regions globally.
And sometimes it's fairly clear if it's irrigated agriculture is probably the elephant in the room in terms of water resources and its impacts. So if there's very little irrigation, like in most, much of Africa, then the most likely cause is natural climate variability. And you saw big increases in storage in the Okavango Delta in that region, increasing trends there.
Other regions where we know it's human, like Central Valley in California or Northwest India or North China Plain. Then human intervention. And then if the region has not been subjected to drought or flood, it's not a drought or flood because you know that from the precipitation data. So we're trying to get a kind of the best guess at what's happening in these different regions. I thought you did a very nice job of covering all of the potential causes.
And as you say, it's early days to look at long term trends. And so it'd be some of these may be emerging trends.
I also really liked your recent work on climate extremes, identifying wet and dry areas or areas that have been subjected to wet and dry area periods, and then trying to quantify the intensity of those and how that has changed over time because my feeling is that a lot of times we focus on droughts and water scarcity, but we have floods or wet periods. And so I think the challenge in the future of world water resources is going to be trying to manage these extremes. So maybe you could describe a little bit, Matt, about how you did the analysis to quantify the intensity of those extremes and what some of the main findings were.
[00:15:28] Matt Rodell: Yeah, the initial idea. that I had was, can we identify the biggest wet events, biggest dry events in a GRACE data record? And so Bailing and I discussed how we would do that. And we came up with the idea of a clustering algorithm, which basically automated way it looks at the GRACE data and it sorts out places where there is a deviation from the climatology.
So if you have a normal seasonal cycle, you're going to have ups and downs in terrestrial water storage. But if you remove that and just focus on variations from that climatology, we've decided we'd look at regions where they're either one standard deviation above or one standard deviation below their normal for that particular location and time of year.
And then we the clustering algorithm basically looked for adjacent grid cells that were experiencing the same thing. They're either above one standard deviation or below in the same direction. So it clustered the cells both spatially, where they were adjacent to each other, and in time. And so in that way, it was able to sort of build up these clusters of very wet or very dry cells that we would call the wet ones, wet event or pluvial and the dry ones, obviously, we call that a drought.
And if it was long enough and big enough, then we had a threshold where we'd say, okay, this is an event that we're interested in. So we did this with the GRACE data from 2002 through 2021, through the end of 2021. So it includes GRACE follow on data. And it identified about 500 wet events and 500 dry events in the GRACE and GRACE follow on data record.
So the first thing we did was rank them. We were just looking at what were the biggest ones. And it turned out the biggest one by far was a wet event in Central Africa, which was actually ongoing, still ongoing at the end of 2021. So the, the units we're using something called the intensity. So the intensity is a metric that combines the spatial scale, sort of the depth of the drought or the wet event and the time period.
So the units end up being cubic kilometers of water times months. So, cubic kilometers of water either gained or lost from the region times the duration of this event. So, a lot of the biggest events were on the order between 5 cubic kilometer months. And the, but this big one in Africa was over 30, 000 cubic kilometer months.
It was like way bigger than any of the other ones. So that was kind of surprised and pretty amazing to see, basically cover all of sub-Saharan Africa in this huge region, some other big ones. We saw there was a wet events in the Eastern half of the U S started around 2018, went through maybe 2020 or 2021, which I remember I live in Maryland and, I can remember it was a very, it was a wet period and when I say a wet event or pluvial doesn't necessarily mean there was flooding. It wasn't like I was looking around and there was my backyard was flooding or something like that. It was just a long period of more rain than normal. So that's sort of what we mean by a wet event. And so the aquifers are full and reservoirs are probably full. You're, you might be wet in your yard. You don't have to worry about watering the lawn or anything like that, but there can be floods associated with it. If you're experiencing a pluvial in your region, then you get a big rainstorm. That's when you could really have flooding because you already, all these stocks of water are full.
There's no where to put the excess water. So yeah, so that was a big one, the eastern United States, and then there are various well-known droughts around the world. One that was notable is that there's been a drought ongoing in much of Europe. They've really been suffering from drought there, and that was ongoing at the end of the time period. So a lot of interesting events going on. So ranking them and looking at the biggest ones was one aspect of this study.
But then we started thinking, well, what's driving this? And the initial thing you think of is like El Nino and the other climate indices. Oceanic oscillations often drive these types of events.
So we looked at seven different climate oscillations, including El Nino. And so we had those time series and then we took, we call it global total intensity. So take all the wet and dry events happening at a given time in a given month, and you take the absolute value of them. So basically you're adding up, you're able to add up the wet and the dry events together, get this global total intensity at any given time.
And so we correlated that with all these climate indicators and also the global mean temperature, the global warming could be related to this as well. Well, it turns out that the global mean temperature was the big winner. I mean, it was significantly correlated with the global total intensity of the wet and dry events around the world. Much better correlation than it was El Nino or any of the other indicators and the past eight years or so we've had the eight warmest years on record and it turns out that's when the global total intensity has been the highest as well. It's really astounding when you look at the time series and see how well they're correlated and how this is sort of a harbinger for what's to come, right?
I mean, we, this is significant correlation. It's not a fluke and makes sense intuitively. If the air is warmer, that means that during a drought, the air can hold more water. It's the atmosphere is thirstier. It can pull more water out of the land surface. And so the drought can be deeper and longer and maybe larger spatial extent.
And then on the flip side, when you have a, when you have wet events, pluvial. You can have the atmosphere is able to hold and transport more humidity. So you pull more water, more humidity out of the ocean, transport it onto the land. And then if you have a wet event, then potentially that could, it could be bigger and longer.
And that seems to be what we're seeing. And I'm sure there's more nuance to it than that. I mean, global warming is going to affect things like atmospheric circulation patterns, that sort of thing. But I think, this net effect that we see, more droughts and more pluvials as the earth. Worms.
I mean, there's a good chance that that's what's coming for us. And people have been curious about this for a long time. Talk about acceleration of the water cycle. And will we have more droughts and pluvials? But it's hard to say for sure because a lot of its model data, which people don't trust. Well, here's some hard data that we have from the past 20 years.
It shows it's already happening. It's already here.
[00:21:40] Bridget Scanlon: Right. I think that is a really nice analysis. And I often think we overemphasize droughts because maybe they extend for longer time periods. But then I think for water resources management, we really need to take advantage of the flood periods and try to manage those then to get through the droughts.
So you mentioned cubic kilometer months. And so for American listeners, some of the hydrologists, they're in, they're used to a million acre feet. So 1.2 cubic kilometers is a million acre feet. So they're fairly similar. And it's really nice to do this large scale analysis. And you also mentioned the drought in Europe.
And I recently spoke with Megan Hart at Aon Insurance. And one of the biggest causes of natural disasters in 2022 was heat waves and droughts with the mortality attributed mortality about 20,000 people, which was about a third of the global related to natural hazards. So I think Europe is maybe experiencing more droughts and we generally don't think of that, but that of course it can affect energy production and transport along rivers and all.
But it's nice then, and then you mentioned Central Africa, the huge wet period that was by far larger than many of the others, 30, 000 cubic kilometer months, and other biggest one was about 10 and many were 5 to 10, 000. So very interesting analysis. And I think emphasizing that we are often subjected to wet and dry periods. And so the challenge then for water resources would be to manage those extremes. So you develop a lot of products and oftentimes when somebody says to me, Oh, well, Texas is in drought or whatever, I go to the U S drought monitor and I look at the time series and I see where we are relative to the longer term context.
And I think you have a developed a drought monitor product also that the U S drought monitor incorporates. into their program. Maybe you can describe that a little bit. And trying to provide an operational product, that's pretty challenging, but it's very important when we hear about these things and where are we relative to the long term and how bad is this, that sort of thing.
So maybe you can describe the drought monitor product a little bit, Matt.
[00:23:59] Matt Rodell: Yeah, so as I described, but we can use the, I'll call it raw GRACE terrestrial water storage data to understand droughts and what events, but it's really not very useful for operational drought monitoring because of the latency. You only have data three or four months after. After the present, then that's, that's all data. The drought or the flood could be done long gone by then and also the spatial resolution is an issue. So, so what we decided to do, and this is 15 years ago now, is we basically took the GRACE data and we used it in a land surface model as a constraint on the land surface model.
So let me explain this a little bit. A land surface model is a numerical or a computer model of the physics of the water energy cycle on the land surface. So you can think of it like the land component of a weather or climate forecasting system. And it's pretty sophisticated. Again, people get scared by the word model, but really it's a model is something that in this sort of model is something where we have a good understanding of the physics that describe what happens.
For example, the precipitation after hits the land surface, some of it's going to saturate the soil and some of it's going to percolate further down and recharge the aquifer, some of it's going to run off, some of it's going to evaporate right away, some of it's going to be used by plants and transpire, and then you have various other different types of physical processes happening with the energy cycle as well.
So we can model this, we have a pretty good understanding of it, so we can convert what we know into equations and so we have these land surface models and they take as input things like meteorological conditions, so the amount of precipitation, the solar radiation. The wind speed, humidity, surface pressure.
These are all the inputs that drive the model forward in time. And those are what's nice about that is most of those are observed where they come from these atmospheric analysis systems that basically integrate a bunch of observational data. So it's more or less observational data that are driving the models forward in time.
The models also have static parameters that are based on things like the vegetation type and the soil and the topography, so they're regionally specific. So we have these land surface models that can run on their own, and that's been happening for decades. But what we decided to do is take the GRACE data and use that as a constraint on the model.
So the model can simulate the things like the groundwater and the soil moisture. And snow and that the simulation is is fairly good because we have observational inputs, but we can make it better by taking our GRACE observations and then constraining it. So where the model starts to veer off from reality, the GRACE data help to to bring it back in line and then the flip side of that is the GRACE state or this coarse resolution, they're not available in real time, but by incorporating them to the model, it basically disaggregates this coarse resolution GRACE data. So it brings it down from 150, 000 square kilometers to the resolution of the model, which for us is either 12 kilometers or 25 kilometer grid cells.
We run the model on a 15 minute time step and we produce output on a three hourly time step. That's obviously much more useful than a monthly time step for a lot of applications. And we can run it because these meteorological inputs are available in near real time. We can run it and have output within 24 to 48 hours.
So we've been doing that since 2011. We've been running the GRACE data simulation in these land surface models and then taking the output on a weekly basis. And basically producing drought and wetness indicators and what these look like is if you have at long term climatology. So we run the model starting back in 1948 run up to present.
So we have this long term record of the variability of groundwater and the soil moisture and the snow, how they vary on a seasonal basis, and then the larger range, wet periods and dry periods. And so we can use that to basically construct a cumulative density function, which is basically your climatology of what we expect for a given season, or a given time of year, in a given location, what's normal, what's high, what's low.
And so we can take current conditions that come out of the GRACE to the simulation model and then compare them to the long term climatology. And so we get these drought indicators that are in the form of wetness percentiles. So if you're in the fifth wetness percentile, for example, what that means is that for that particular location in the same time of year, basically the same week of the year, going back to 1948, it's only been drier than that 5 percent of the time if you're in the fifth wetness percentile. So it gives us the whole range. It's not just a drought indicator, it's a wetness indicator. And so we produce maps of the current wetness conditions in the, the shallow aquifer, the, the root zone soil moisture and the surface soil moisture.
We produce these maps on a weekly basis and deliver them to the National Drought Mitigation Center. Which is where they're then used as one of the inputs to the U. S. Drought Monitor, which is our premier drought monitoring tool for the U. S. They have many different inputs, many different indicators they look at. Some of them are model products, some are observations, and then they also talk to state climatologists and, and other stakeholders to get an idea of the conditions. It's a useful input for them. They don't have any other groundwater type inputs, so, so that was a first for them to have some sort of a groundwater indicator that they can incorporate into their drought maps.
And so it's used in that way, our maps are also publicly available through the National Drought Mitigation Center website. And we have, we have a lot of users, although we don't always know who is using the data, because we don't collect information, we don't make people sign up or register or whatever.
But it's funny, every so often, when there's a glitch and the data aren't available, these users will come out of the woodwork, send me an email sometimes and say, Hey, what happened? I need this data. Oh, sorry. Thanks for letting me know. And we'll get that. Going again and then say, by the way, how do you use these data?
And a lot of the users are, do things that are related to agriculture, some related to insurance industry or energy production. And the strangest one I got was a woman who worked for an agency in Nevada who is forecasting the migrations of, of wild horses and use the data for that. So I thought that was really cool.
Again, wouldn't even know like the whole range of users. I think there are a lot of them out there, but I don't even know how many of them are.
[00:30:19] Bridget Scanlon: Well, I mean, it's great because you can provide continuous coverage, both spatial and temporal. And so that's really nice. And then the model providing the long term from the late 40s to present gives you, well, it's dry now, but how dry is it relative to the long-term record? And so That's always a fundamental question you have and people, it's hard for them to gauge themselves because we don't really remember how dry or we don't have independent data. So I think that long term context is very important. And then also because you're looking at changes in water storage, you can also use those data then.
To quantify what the cumulative water deficit is during a long-term drought and at the end of the drought, what you need to overcome that deficit then to recover from the drought. And I think J. T. Rieger did some really nice work looking at flood potential index. So as you said earlier, during a provial period, it might be wet, but it may not be flooding or whatever.
But if you know that you're close to the long-term maximum in water storage, maybe you're much more vulnerable to flooding, then so can also be used for a flood potential index. But oftentimes I think it may be more suitable with the monthly data to the drought index because the flood potential index they might want to know maybe more at the shorter timescales with less latency.
But I think the other aspect is if you have a big flood like we had Harvey in Texas, how much water accumulated over that region, you can talk that up with GRACE. And then how long is it going to take for that water to dissipate? So I think these things are advantages of the GRACE data and how they can be used to evaluate these extremes.
So another aspect of your work, Matt, is looking at, I mean, GRACE, I think, has been very valuable to scientists and hydrologists and everything. But I think it has also been extremely important for communicating the status of water resources to policy makers in it. And your work in Northwest India in the late 2000s demonstrated that.
And then also the work in the central Valley, maybe you can describe those a little bit.
[00:32:38] Matt Rodell: Right. So one of the first big hydrological phenomena we observed with GRACE was this massive depletion of groundwater in Northern India. We're able to pretty easily. put the puzzle pieces together and see it's one of the most heavily irrigated parts of the world.
They irrigate mostly using groundwater and at Northern India's semi-arid region. So it makes sense that if you use a lot of water, a lot of water out of the aquifer, it's not going to recharge fast enough to, to keep up with that. So I think the Indian government, and this is back in 2000, I think they probably had some grasp of what was happening, but it wasn't really making the news. It wasn't really well known. And their data, they had data from wells, and it was kept on some Indian government website that some people had access to, but not necessarily people outside of India. And so when we published that paper showing this massive rate of depletion in Northern India, it really opened some people's eyes and both in India and outside. And I think it started to open the discussion of the policy. I mean, they had a policy that the farmers could basically have free electricity to pump as much water as they wanted to because they wanted to support this agriculture to feed a huge population. But the effect of that was that there was no incentive to save water.
So you use as much water as you can and you figure out, well, if I don't use it, then my neighbor's going to pump it out. So I might as well use all the water I want to. And so that was sort of exacerbating the situation, I think. And so following that paper, like I said, I think there was a lot more discussion on policy and how to conserve the water.
Aquifer levels are still declining in India, but I think at least it's, people are aware and probably making some hard decisions on ways to, to conserve the water. It's not just India, it's, we're doing the same thing in the U. S. We've known for how many years, 70 plus, that the high plains aquifer has been declining at a slow but steady rate.
And it's really only people in the Southern part on the fringes that really that maybe feel it so much, the wells go dry and they have to either drill a deeper well or move on to something else. And likewise, in the Central Valley in California, which produces, what is it, half of our nation's produce, quarter or half, it's a huge amount of produce we consume, comes from the Central Valley, and that's another place where they take water out of the aquifer, the farmers pump water from the aquifer, they use it to irrigate crops, and they're using water.
The water's evaporating in the atmosphere faster than it can recharge, so there's a long term decline, and it's interesting there because it's not sort of this straight shot that we see in India, it's really up and down, like I've heard it described as a tennis ball bouncing down the stairs. So you have these wet periods, and the water level comes back up, but not quite as far up as it was the last time there was a wet period, and then you have a dry period and a drought, and it's a combination of the drought and farmers needing to pump more water, because Because you're not getting as much rainfall and so you have this rapid decline in, in the groundwater.
And so we, we saw that in the Central Valley. I mean, it was well known because Claudia Faunt and others had used well data to describe the same thing going back many decades. But it was another thing where, you know, their case where the fact that we could see it from space and provide this sort of holistic view of what was going on and put a number on it in terms of cubic kilometers of water being lost.
I think that got the attention of some policy makers, and it wasn't long after that that there were discussions of the groundwater policy that ended up... Resulting in the act that passed and so over the course of the next couple of decades, they're going to be implemented, implementing this policy to try to conserve water in the Central Valley, which is usually important for not just the farmers there, but for the whole nation, because if they, the aquifer, the water becomes either too saline or inaccessible because it's so deep or just the wells are just dry.
You can't go any deeper. That's gonna be a big problem for us in terms of producing food for our nation. So getting a hold on it and being able to use the water in a sustainable way is critically important.
[00:36:39] Bridget Scanlon: Right. I think the work in India was fantastic. And I think you had more recent papers where you were showing water storage increasing in peninsular India.
And maybe some of that might have been linked to they're trying to do managed up for recharging, storing more water so that for longer periods so that it can infiltrate. So all of these things, I mean, we can track the changes in water storage. It takes us some time sometimes to figure out what's causing those changes, but at least we have the data to see the changes.
And then we can try to figure out with other sources of data then what might be causing them. And I think the case in California, Jay Famiglietti worked. closely with the Water Boards there and the Sustainable Groundwater Management Act. It's a great example. And I think the 60 minutes program with Mike Watkinson and Claudia Fund, I thought that was really nice because it gave equal emphasis to satellite data and ground-based monitoring and that we need all of the above.
And I think that's a very important message because I hear. Some people say, Oh, well, we can do it all with satellites. So I think we need to keep the ground based monitoring going also. So I think with the operational aspects of GRACE and the drought monitor, and with the people becoming more accustomed to using GRACE data, I think that puts pressure then on continuing the mission.
Maybe you can talk a little bit about, we had GRACE, then we had a year gap, and then we had GRACE follow on, and what the future of the GRACE missions are. Or great like missions,
[00:38:10] Matt Rodell: right? So yeah, GRACE lasted for 15 years, 20, I mean 2002 to 2017. And then NASA prioritized. NASA and Germany prioritized launching a successor GRACE follow on which there's 11 month gap.
But we had GRACE follow on came online in 2018 and then when we had the United States had a decadal survey in Earth sciences, which is basically a committee from the National Academy. National Academy of Sciences, which recommends priorities for Earth observing missions that NASA should launch over the next decade.
And one of the five, the most important missions that it recommended was a mass change mission, which is basically a successor to GRACE file on. And so that's being developed now. I'm on a science team for that and hope is to launch that around the 2028. timeframe. So that would just provide continuity.
It would actually be called GRACE C for GRACE continuity. But in terms of the future, you've always been thinking about, well, what can we do that's better for satellite gravimetry, or maybe would give us higher spatial resolution, better accuracy. lower latency, that sort of thing. And there are two real directions that are being discussed.
One is to have multiple pairs of satellites in orbit at once. And the reason that's important is because we're somewhat limited in terms of the spatial resolution accuracy by the temporal undersampling of the Earth, meaning that there are especially changes, rapid changes in the mass of the atmosphere that are happening that we can't sample well enough with a single pair of satellites.
And so if you have multiple pairs, you basically can reduce the errors associated with that. And so together with the European Space Agency, NASA is looking into that. Europe is considering launching what they call their Next Generation Gravity Mission. Which would be another pair of satellites similar to GRACE except to be flying at a lower altitude instead of 500 kilometers, 400 kilometers, basically the closer you are to the earth, the higher spatial resolution you can get.
But the flip side of that is there's more atmospheric drag, which is measured. Currently using onboard accelerometers on the satellites, but as you get lower and lower, the errors associated with that become larger. So this next generation gravity mission would then use a satellite that has the Xenon thrusters on board, which would sort of counteract the atmospheric drag.
The thing that's in the center of the satellite called the proof mass is sort of free floating with a bigger satellite bus around it. It'll be protected from the atmospheric drag, so that's probably more technical than you wanted, but the idea would be that you'd have the second pair of satellites at a lower altitude.
It would also be at a lower inclination, so it only sampled the area between about 65 north and 65 south, whereas GRACE goes all the way up to the poles. Nearly, and the sort of mid to low latitude region is the part that's really under sampled by the GRACE missions. And so providing more observations down there will be hugely helpful.
And the idea with the constellation of the two of them, which we're calling MAGIC, would be you'd have these two pairs. You can find the data from both of them and come up with better space resolution, better accuracy. And that we'd be looking at maybe around 2032 timeframe.
And then the other thing that is in the very early days is quantum gravity, which I won't get into too much because it's pretty complicated and a lot of it goes over my head. But you're basically using a satellite with a cloud of atoms floating in it. You can control using lasers and then you look at how the cloud of atoms reacts to the force of gravity. And potentially that could give us much higher resolution or similar resolution with a single satellite instrument. If you had more satellites, you could get some pretty big increases in the spatial resolution.
So that's exciting. But then we're talking about decades from now and we'll see if I'm still working that or if I'm watching in an interested way from retirement.
[00:42:05] Bridget Scanlon: Well, I mean, it's amazing where we have come from and what GRACE has been able to accomplish. And I think hydrologists are always saying, oh, they would like 50 kilometer resolution rather than they would like the highest resolution possible.
But to achieve that, you'd have to have a sapphire in a very low elevation and the lifespan would be zipped. So there are all these tradeoffs. And so the next NASA mission is very similar to what the original mission was, which, and then it optimizes, tries to optimize, I think, lifespan and resolution and things like that.
So it's a lot of things to consider, but it's very easy for us to get accustomed to having all of these data, but it really opened our eyes early on. And you have been there since the beginning, when you did your dissertation on what you would expect to see with GRACE. And then what we actually saw, so it just blows my mind, the confidence that the people had originally in GRACE and putting forward the mission and doing all of that when we knew so little.
But it has made a huge contribution to understanding the water cycle and the climate impacts and the human impacts, and that should help us better manage. the resource, I think, in the future.
[00:43:20] Matt Rodell: Yeah, I mean, I, I feel really fortunate that I was there at UT when the PI Byron Tapley was at UT as well and allowed me to get Insider Scoop and Jay Finlay and I were on board, like we said, some of the first hydrologists who were interested in GRACE.
So it was a bit of a risk for me as a PhD student to have my dissertation on a pre-launch study of a mission that hadn't launched that no one was completely sure it was going to work. It ended up being a good reward in the end. So I'm really. Happy that I did that. But looking back, I was like, wow, that was pretty risky.
And especially with so many doubters out there that it would ever provide anything useful for hydrology.
[00:43:58] Bridget Scanlon: Right. But I mean, I think that's where science should be, but we oftentimes just want to go the safe route and things that are tried and true and stuff like that. But they should really give return on investment and we all are benefiting from it.
So our guest today is Matt Rodell, and Matt is the Deputy Director of Earth Science at NASA Goddard Space Flight Center. And I really appreciate you taking the time to explain a lot of these concepts and applications. And thank you so
[00:44:27] Matt Rodell: much, Matt. Yeah, that was fun. Thanks, Bridget.
