 Good afternoon everyone and welcome back to the fall meeting of the committee on solid earth geophysics. I'm Torsten Becker, chair of the committee. We're pleased to present it to day two of a 2D program on solid earth science and sea level change. Sea level rise is one of the most critical problems facing society. The solid earth plays an important role in unraveling the evolution of sea level of a range of spatial temporal scales. This meeting reviews the state of sea level science discusses some of the interactions with the solid earth and explore some of the important questions that we still need to answer. Yesterday, we're at overview talks on sea level glacial isostatic adjustment and transient solid earth deformation. We started out with a fantastic overview by Ben Hamlington on mapping sea level change in space and time. What we often hear about is the global mean sea level change. And we know that over the historical and surmental time sea level has risen on a global scale by about one millimeter per year between 1900 and 1990 and about three and a half millimeters per year more recently between 95 and 2020. And this increase in sea level we now understand has breaks down into different components and about a third of that global increase in sea level is related to thermal expansion. And the other two thirds are related to mass increase that mass increase. It's mainly due to the melting of the ice sheets such as in Greenland and Antarctica. And as Dr. Hamlington reviewed, we now have a much better handle on how this works and how the overall budgets. And what is emerging is a very consistent and very robust picture of global sea level change. And the projections of that sea level change into the future, of course a very much contingent on what will happens in the way that humans modify the environment in terms of our CO2 emission pathways. But when we consider the actual impact of sea level for communities on the regional effects become important, and we soon realize that sea level change is a very complex problem that is expressed on different spatial temporal scales. It very much matters which process is considered for what time scale. And so as Dr. Pikic then explained to us in his second presentation on sea level on the solid earth insights into ocean circulation and climate. Then when we ask locally as a policymaker what are the largest uncertainties on the change in sea level, then often this breaks down to understanding the ocean and atmosphere system and natural fluctuations in the climate system, such as and so oscillations are often the largest contribution to sea level change on a local scale. These climatic and ocean circulation signals, of course, not independent from the change in the global climate system, and much is yet to be learned by integrating our understanding of ocean and climate dynamics with measurements of sea level. And in particular measurements of relative sea level on local scales. And we've seen much convergence between the geophysical and the oceanographic and atmospheric communities in terms of understanding this complex multi scale form. Now what this meeting focuses on is the role of the solid earth and how our understanding of the long term deformation of the lithosphere in the asthenosphere. And in fact, the convective transport of density anomalies in the asthenosphere can help us contribute to addressing this society, societally relevant challenge of understanding sea level. And in the afternoon we then heard from Pippa White House on what used to be neglected processes the role of solid earth and controlling ice sheet contributions to sea level change. And there we have a number of important interactions to consider. Now, the big signal on global scales is that we're moving out of a glaciation and the reduction of the ice loads over time scales of 10s of thousands of years over Asia and over Canada leads to glacial isostatic adjustment. Now this removal of the ice loads of the 10s of thousands of years does not just lead to uplift and reduction of local sea level but it also leads to increase of sea level in other places. And the global response of the earth that is to do with the redistribution of the ice masses into the oceans. But also the change due to the removal of the attraction the gravitational direction of the ice loads that plus the flexure of the elastic lithosphere leads to global fairly complex expression of local ice loading. So each ice sheet over Greenland and over Antarctica has a specific spatial pattern and this approach of using the fingerprints of the ice loads as promising in terms of understanding where the melting happens, contributing one puzzle one piece of the global complex system so we have relative sea level we have satellite observations. We have an understanding from the global response of the earth that is very much to do with the structure of the lithosphere and the underlying asthenosphere that tells us about how sea level changes. Now besides these global effects, where the structure of the earth is telling us about how different sea level pushes things around we also have very important regional effects. They include that the stability of ice sheets, which is controlled by the behavior very much underneath the carving side of the ice sheet is controlled by the motion of the grounding line and the grounding line distinguishes between the region of the glacier that accumulates ice in the region where it loses its ice. And if that ice loss can be then accommodated by a visco elastic response of the underlying solid earth, then if that response is fast enough on local scales it can actually serve to stabilize the ice sheet. So regionally then on scales of 50 to 100 kilometers on those scales is then very important to understand the flow behavior of the asthenosphere underneath the ice sheet, because if the viscosity is low enough. Then the solid earth response can serve to stabilize the ice sheet and therefore in the end lead to less extreme increase in the ice associated mass in the oceans and in a different estimate of sea level. And it turns out that these viscosities are of the order of which they have been determined to be based on rock mechanics experiments post seismic deformation and other ways of getting at the flow behavior. So there's an amazing convergence between traditionally distinct communities mental dynamics rock physics structural seismology and now glaciology ocean atmospheres modeling to understand this very complex problem of sea level change. And in our last talk, yesterday we had Jackie Austerman tell us about using paleo sea level records to image Earth's internal structure and decide decipher drivers of sea level change. And what Dr. Austerman showed us then is that this this global problem of having sea level and ice and glacial isostatic adjustment, change the loading of the earth, we can then turn around to actually use the loading to say something about the internal of the earth. And with that infer properties including viscosity, which, which is what what stabilizes the ice sheet in the first place. And so we have an amazing cross scale connection going between regional melting to regional response in terms of the ice solar and the action to global really distribution of water, which where this global distribution of water can then in turn tell us something about the interior of the earth, perhaps even including down to the common boundary 3000 kilometers down into the earth. And, and where we're sea level can be used as a as a as a driver, which we can then measure the earth's response and say something about the long term evolution of the planet. So it's a, it's a fascinating field where we still have much to learn. And one of the particular important points is that when we are asking about the impact of sea level change on on communities on on on coastal communities. Then it is very important to not just understand what the sea level is doing with respect to a solid earth that's assumed fixed but also what the supposedly solid earth does in terms of actually moving up and down. And these vertical crustal motions that I'm important to understand for relative sea level is what we're going to focus on today, and we're going to have a discussion of verdict of geodetic constraints on these vertical coastal motions followed by general discussions with all of our speakers, where we will be able to revisit these cross scale complexities. Before we begin, I have a few announcements. This session is being recorded, and will be available on our website within a few days. In addition to questions from our speakers and committee members we plan to take questions from the audience. As noted in the previous slides, simply click the q amp a button at the bottom of your screen type of questions and click send. The questions you submit maybe read aloud included in our video recording. In the interest of time will skip, skip committee introductions and buyers of our committee can be found on the Academy's website. Importantly, I want to thank all of our speakers and the audience again for taking the time to join us yesterday and today. It's great to hear from each of them, and I'm excited for the day. Before we begin over, Jeff Reimler will introduce today's speakers and moderate the discussion. Thank you, Torsten. We're right on time and we'll we'll stick to our time so we should be right on the schedule for those of you who are looking at the agenda. I'm excited to hear from our two speakers this afternoon bill Hammond and Manu Shirzai. And I know they're going to give us very informative and interesting talks. We'll have time for questions and then we will reconvene after the break for discussion with all of our speakers from yesterday and today. So hopefully we'll have some time after each talk for some questions for each individual speaker and then a little bit of time at the end for questions for both. And then we'll take a break. So first up is Bill Hammond. Bill is professor of geodesy and geophysics at the University of Nevada Reno and a member of the Nevada Geodetic Laboratory Research Group. Bill uses space geodesy to study active processes inside inside the solid earth a whole range of things, including tectonic and seismic cycle deformation mountain building mantle flow geophysical loading of Earth surface tectonic controls on geothermal resources and interactions between tectonic and magmatic systems. Bill is yours. Okay, can you see and hear me. Okay. Yes, looks good bill. Okay, I'm going to share my screen. I should be used to this by now. Okay, so you can see my presentation. Yes. Okay. Great. Thank you everyone and thank you to the committee on seismology and geodynamics for asking me to speak here today. It's a great opportunity. I'm going to talk today about vertical land motion from regional to global scales. It's constrained by GPS data mostly vertical land motion is very important for sea level rise that has been mentioned a number of times in the workshop so far. So I'm not going to get too much into the reasons why we're doing this but I will say that that it is. A one to one impact on the on the impacts that sea level rise has on coastline so if for example sea level is rising three millimeters per year. And the land is going up three millimeters per year that that virtually nullifies the impact of that sea level rise but if it's going the other direction, there's subsidence of three millimeters per year it can in fact double the impact of the sea level rise. So, so we need to know this we need to know what the land is doing in order to get a full assessment of the impact of sea level rise. So it's one of the fundamental observables, and it's important for studies on all scales both local to regional to global. It also helps close a loop between other kinds of data which constrained sea level. For example, geocentric measurements of the rise of the sea surface and connecting that to the shoreline where tide gauges collect data vertical land motion is essential for closing that loop. So, we, we have a lot of GPS stations not just near the coasts. So what we're going to look at is vertical land motion is constrained all over the land, not just at the coast and this is important for recognizing the processes that contribute in the solid earth to moving the land up and down. When solid earth geodynamics are the root cause we need to understand the underlying processes as best as we can. So the solid and the partially solid earth vertical land motion exhibits highly variable spatial and temporal scales from things that are virtually not changing over time over the time period of GPS observation to things that are changing rather rapidly to seasonally. There's a diversity of processes that contribute to vertical land motion. And we're going to see a bit of those next. I think Manu is going to do this too. So it's good that we're both kind of hammering on the message that the GIA is not the only factor that contributes to vertical land motion of coastlines. GIA is certainly the biggie on earth that has the biggest spatial scales. And I'll discuss this more later but it has a very big impact on the overall figure of the earth that changes over time. In regard to GPS measurements, GIA is a very steady process. It doesn't change much so it goes into the trends of the data that we're going to look at. Another contributor to vertical land motion that we know about is mantle flow. There are other aspects of mantle flow besides GIA that is in dynamic topography. This can include things like slabs, moving through the mantle, tectonic slabs, drips, delaminations, pieces of lithosphere that detach and sink down into the mantle can be removed from the lithosphere and allow it to move vertically through buoyant forces. And when it comes to what we measure with GPS, we might be interested in how these mantle effects are changing over time because we're not looking at the static part that supports topography but the part that is moving and changing and gives a velocity over time. And then we have other things on earth that move mass around the surface such as sediment loading which is also another long-term process which can involve viscous relaxation of the mantle and potentially in this example there's subsidence in the Gulf Coast on the order of several millimeters per year associated with a process like this. Other things like tectonics, long-term motion of vertical motion of Earth's surface results in mountain building. We know there are lots of mountains on earth and these were built largely through tectonic processes and they can have very significant impacts on coastlines. In this example I'm showing the terraces that are built in Southern California. How that coast has been uplifting over time is very apparent in the geology and this is a very long-term process. Interseismic tectonics is the motion of the Earth's surface that occurs between very large earthquakes. So this is a shorter term. This is something that's very accessible with geodetic data. It's a medium-term process on the order of hundreds to tens of thousands of years can go by between large earthquakes. And we can see in this example for subduction zones. This is Cascadia where the coast is going up interseismically because of convergence of the Wanda Fuca and Gorda plates into North America causing a contraction which is temporarily rising the coastline. And of course then at the end of the seismic cycle or the beginning, depending on your perspective, there will be a large earthquake where the coastline suddenly drops and there will be a very bad day, potentially tsunamis. In the case of the Tohoku earthquake, the vertical land motion on the coast was about half a meter even though the earthquake was way offshore towards the plate boundary. So that results in very significant, very rapid vertical land motion that is part of the entire process. Other things that are potentially more time variable include the effect of aquifers and groundwater basins, how water content in the Earth can affect the surface motion. This is very accessible with GPS and INSAR, I think Minou's going to talk more about this after me. These can happen on very short time scales. They can be seasonal to decadal and they can often depend on what people are doing and how they're pumping water out of the ground. Sediment compaction is another, this can be a natural process that occurs over very long geologic times. It can be related to compaction and aquifers are similar to it, but it can result in vertical land motion that can have a broad impact on subsidence, this example in Southern California. And then we also have the effect of water in another way, how it can load the Earth's surface as water moves around from one place to another. The terrestrial hydrosphere can load the Earth's surface from the top. So this is a force from the outside of the Earth that's loading its surface or unloading it in the case of droughts. In this case, in the Sierra Nevada and coast range, in California where drought, long term drought of a number of years help drive uplift in the Sierra Nevada and coast ranges, Southern California, and also responded to groundwater pumping in the Southern, Central Valley. So those are some processes that are there. It's not a complete list, but we need to recognize that there are a lot of different things going into the mill of vertical land motion and we would like to be able to just measure it, see what it is and then see what we can do about using this information to interpret and understand its contribution to sea level. So regardless of the project or the process, if we have precise measurements, we can measure the vertical land motion. This is an example GPS station from the NSF's Network of the Americas, it's P534 near the coasts, north of Santa Cruz. And this GPS station is collecting data all the time and we can get precision and latitude, longitude and height to the order of a millimeter given 24 hours of data. So about every day we can find the position of this, the face center of this antenna to about a millimeter. Moreover, we can put that into a global frame of reference and a global frame that we use is the International Terrestrial Reference Frame, version 2014. And this frame is very useful, it's global, and its origin is aligned with the Earth's center of mass to a precision of about 0.2 millimeters per year. So that's about how well we know where the center of the Earth is, even though it's a highly inaccessible location. So rates of motion in this reference frame are with respect to that Earth's center. So this is great, however it requires a ground station, some money to install things like this and look at the data. And preferably we want these stations to be placed in locations that are representative where the motion is representative of a large volume of Earth's crust, so we don't have to have too many of them. So if we have a 20 kilometer spacing, for example, that those things are measuring similar signals that are happening in the solid Earth. Okay, so data from these stations can look something like this. This is an example of vertical positions over time. Each blue dot in this time series represents one days worth of data. We can see there's a little bit of wiggling in the vertical motion, but overall there's a trend. In this case we've collected about 20 years worth of data, and it seems to be describing a downward trend. If you fit this line, the subsidence is on the order of 0.56 millimeters per year. That's just a little more than half a millimeter per year, and I would say that because we've collected data for so long, we can be fairly certain that that subsidence is real. The data are somewhat convincing that the subsidence is going on largely because we've collected data in a stable place for so long. Despite that there are some some digressions from from perfect linearity where the time series wiggles just a little bit and the time series underneath is is the velocity for sort of a moving average of two and a half year window looking at how that velocity changes over time. And for the most part it's pretty stable but there are some digressions where for brief periods the velocity changes by an order of magnitude and then returns to its background rate. So there are some wiggles. We have to contend with that. But the trend seems representative of long term motion and that's been borne out in many studies where we've compared the geodetic results to to geologic processes the movements of plates separates on faults and the like and and for it's very common for the geodetic rates to match up with geologic rates suggesting that they are sensitive to long term geologic processes. Okay, so Ben Hamilton showed this map briefly yesterday. This is a clip from our website that shows the stations in the United States. There's something like 5000 of them in this map so we have a lot of data. At the Nevada Geodetic Lab we're currently processing we're just shy of 19,000 stations now globally distributed that come from hundreds of different networks around the world that that provide open data. So, so this is a lot of work going out to to the different FTP servers and network web pages and like and pulling in going collecting and finding data and processing it all in a uniform way using the gypsy software and products from the Jet Propulsion Laboratory. So this is a similar kind of map. It's the same database just done with a GNT map and it's the global distribution of GPS stations in the NGL archive. And in here we have time series that are up to 27 years long. In some cases we get solutions less than two hours from the time the data collected so they're fairly up to date. And this includes everything that we know of and the number of stations is still increasing. The rate of discovery of new stations is still positive and the number of new networks that are going in is still positive. Okay, you'll see right away that the geographic density is highly variable so obviously we're going to have very good constraints on vertical land motion in some places and not so good and others. So a theme of this talk is to try and show what sort of signals are there. Where are we doing well where are we doing not so well and how can we quantify that a little better. Okay, so this is an example of how we use, how we analyze the data. We're interested in doing robust measures of trends that are not very sensitive to lots of things that affect GPS stations and receivers. So these can cause outliers and steps in the time series that are not necessarily known about beforehand. Wiggles from seasonality and the like so so we use the Midas robust trend estimator. And this finds the trend in the time series with very little supervision and very accurately, and it's very robust to problems. And then we also do something on the spatial scale where we, we do a robust medium weighted spatial filtering of the data, which does a great job at removing speckles and very station level noise in the data. So that can be used on a pixel by pixel scale basis to to interpolate the data into these coherent maps of vertical land motion. This one shows the United States. So there's a log scale. So I'm going to change that and put it into a linear scale which is better for seeing some of the details in the near zero area of vertical land motion so there's a lot of stuff going on in these maps. The vertical scale of motion is minus three millimeters per year downward in blue and up three millimeters per year in red. And the big thing in in Canada is upward motion from the glacial isostatic adjustment. It's huge. And then there's the hinge line as it goes through zero is white. And the and the four bulge collapse. Thorsten Becker mentioned this the four bulge collapse is a lithospiric flexure that goes pushes. Most of the central United States downward in response to this uplift in Canada so this is an effect that's on the order one to three millimeters per year and it goes from Alberta all the way to the southeast Atlantic coastline of the of the eastern United States. So there's other things besides GIA in here we can see, for example, a lot of coast, a lot of subsidence along the Gulf Coast from water and hydrocarbon extraction in Texas and Louisiana. We also see some interesting uplift signals in the Western tectonically active Western United States. We can see some magmatic uplift in Yellowstone, some post seismic relaxation from earthquakes in central Nevada. We can see elastic strain accumulation on the Cascadia coastline, which is going up from the process I mentioned before of convergence, the seduction zone, and then we can see uplift in Sierra Nevada and coast range subsidence in the central United States. There's a rich variety of signals here, all of which can, if they're near the coastline can affect the impact of sea level rise. Okay, so, so we want to know how well are we doing, not just on the US scale but everywhere on earth where we have this large database of GPS stations. So this is one of the steps to try and assess the database and see how well we can estimate vertical land motion around there's coastlines. So this map here above is is based on the on the locations of the GPS stations what's for every point on earth the distance to the nearest GPS station so this is the spatial part of the uncertainty. In blue we have a short distance, you know, most of North America is is within few kilometers or maybe 10 kilometers of a GPS station places where we're not doing so well Asia. So this is a function of our own database, not necessarily all the data that the Russians and the Chinese have, but just what's in the NGL database and then Africa has a huge empty quarter in the middle in the northern desert up there the most extreme desert in the world is right near the intersection of the boundary of Egypt and Libya, Chad and Sudan so that's the that's the point on earth furthest from any GPS station, not including oceans. Okay, and then we can look at the statistics of this. The blue bars are for the for the land area and the red bars are for linear coastlines, and we can see that 74% of the coast is within two degrees of some GPS station. And 73% of the coast is within five degrees of some GPS station which which is pretty far if you're interested in a very fine scale but if you're interested in the global scale. That's not such a big deal. Okay, change there we go. When looking at the data we can rely on things other than just the, the distance to stations as measure of quality of VLM estimate. We would also like to know what is the statistically speaking what is the spatial wavelength of the vertical land motion signal, and how does the distance to the nearest station compare to that distance. That's a better measurement of quality of your vertical of your VLM estimate. So we can look at the data and there's, there's, there's some a lot of detail I'm going to skip in the development of these curves, but these spatial structure functions describe sort of how the similarity and signal falls off with distance. As you're looking at the vertical land motion data. So some continents have relatively long wavelengths, some areas have really short wavelengths, and we can use the local wavelength when when doing, when doing the analysis. The map below shows how the distance to the nearest GPS station compares to the, the fall off distance or the or where the spatial structure function passes through point five so it's one near the station and it's zero at great distance, where the SSF is point five is about where the similarity falls off so so we can see that that North America is blue SSF is about one everywhere there and Asia we're not doing there's some where there are GPS stations you get good constraint but elsewhere, not so good. Africa is surprisingly good around the coasts because most of the GPS stations that are there are near the coast where the people live. Africa is highly variable Australia is highly variable. So, so that is an assessment of the quality of the vertical land motion estimate that is going to be built into the uncertainties of what I show you. Okay, so here's here's the actual estimate of GPS vertical rate as a on a global map. And color scale is before minus three millimeters per year down three millimeters per year up and red. And you can see the huge red signals in northern North America and Europe, which are the GIA signal this is the single greatest vertical land motion signal on earth and it dominates the northern hemisphere. We also see a lot of a lot of other little pockets of different sorts of things going on. We can recognize them through various studies that have been done. We can label put labels on what some of these things are. And that helps us interpret how constant the motion may be, and how applicable it may be for doing projections and the like. So, I already went through a list for North America. I don't have to go through very many of these but there are some interesting signals for the subduction zone and in Chile, for example, Tibet, and then Nepal area going up from hydrological and loading earthquake tectonics Australia going largely downward that's a topic for another day perhaps this groundwater subsidence signaling in Angola. There are lots of things that can go on and and it does take work for each one of these locations to try and interpret it and understand whether it's going to be a long term to dynamic effect or whether it's a short term aquifer effect. Usually more work needs to be done. But we can do things that that help just based on the GPS data alone. For example here, I've removed the predicted signal of the glacial isostatic adjustment given one model that's available online. The Peltier at all I6 GC model I've removed here. And that takes away most of the uplift in North America and Europe. It changes things in Antarctica quite a bit. And it may not be doing a perfect job because there's there's some residual signal here in Canada where it seems like the GIA removal has that signal has over corrected a bit and now there's another subsidence North America. I'm going to talk a little bit more about this in a second and how how we can assess the what is the impact overall globally of processes other than GIA and the overall energy in the in the vertical land motion signal. Okay, so one thing we would we would really like to know is is whether the vertical land motion that we're sensing is going to be a long term process or whether it's it's highly variable over time. Can we depend on it to continue with that rate over the period of projections. So I showed you before how I used the GPS time series to to come up with a time series of velocity. We can assess the stability of a GPS station and how much the vertical land motion varies over time by looking at the GPS data itself and developing statistics on the on the velocity time series. In this case, I'm plotting globally the the median absolute deviation of the the velocity time series from every station and then use GPS imaging to image that field so we can see where on earth, the velocities and the vertical component are highly variable and where they're relatively constant. So we can see this. There are places that the Greenland coast experiences, not just an uplift but an accelerating uplift because the ice loss has continued to increase in rate. So it's, it has a red a red color because it's a highly variable time variable motion that is occurring there. So, while these measurements may be dependable we know that they're they're not constant over time. So we might want to treat the East Coast of Greenland a bit differently than we do say the East Coast of the United States where things tend to be relatively constant overall. So other places that that tend to be highly variable over time and the GPS data are places affected by water, such as Amazonia where there's a huge amount of seasonal precipitation, the, the seasonal motions here are extremely large up and down, but the GPS imaging technique that I'm applying filters out seasonality so this is not, this is not the seasonal up and down but this is the change in the inflections in rates and how they change over multi annual periods. So, so Amazonia is changing its uplift rate over time, not just in a seasonal way but through inflections in its rate. We can see pockets of this in California and the southeast. Again, the red is in the California Central Great Valley where aquifer effects are dominating the vertical motion, and they change not just seasonally but the inflect to droughts and management changes and the like. Okay, so that's, that's useful for assessing the vertical land motion that we have. So we have these this temporal variability and we have spatial variability. It's clear from these maps. And we can start to, to interrogate this and try to understand the significance and the source of uncertainties in the vertical land motion. So this map on this plot on the left is just a plot of the nearest neighbor spatial variability of GPS vertical rate. The horizontal axis is the temporal variability from that view time series. And we can see that there's very poor correlation between these two things. So that is telling us that that that it's not just one thing going on the earth is very is highly variable spatially and highly variable temporally, and these two things are not related at all. So we really need to understand both both aspects of the uncertainty. If we're setting vertical land motion in any given location. And this plot here is just, it shows the temporal variability as a function of the sparseness of the data that we have me some GPS stations don't don't perform perfectly so there's gaps in the data. So that that V you rate is actually rather robust and and the V you time series don't change that much in terms of their variability, just because the time series are sparse, given the way that we're going to get about a two minute warning for you minutes. Okay, thank you. So, so we can map the, the vertical and motion field we're very interested in the tide gauges of course been talked about this, and Chris talked about this yesterday so we can use the process to map the insert the vertical rate to the tide gauges. And that's what that looks like. So we can do this now. And then we can use the, the various measurements of uncertainty that we have to determine quality measurements for the GPS based vertical land motion for each tide gauge. And this it's based on very various measures and in the criteria, the thresholds that I've set we got about 52% of these over 2000 tide gauges have pretty good vertical land motion estimates. So at least 7% are downright bad. Those include the ones that are on islands far from anywhere, including GPS stations and then medium is in the middle. They don't fall into either the good or the bad category so we can categorize these. We can also get very, very granular where if an analyst or let's say a manager, a city manager, in this case in San Francisco was very was interested in knowing what is the vertical land motion at a tide gauge near the Golden Gate Bridge in the San Francisco Bay. We can pull these numbers out of the analysis and for San Francisco tide gauge has been there close to 100 years I think the GPS stations have only been there for a lot less time, but we can look at their vertical rates and see that they're all contributing slightly different rates, but they vary between minus a tenth of a millimeter to minus one and a half or 2.4 millimeters per year. And we can do the weighted median strategy to look at these different GPS stations how they contribute and come up with an estimate for the vertical land motion tide gauge so all of these are built into the process so that all these data contribute to vertical land motion estimate at that one location. And since time is short I'm going to leave that and go back to it if there are questions. But I wanted to show sort of on broader scales how imaging over the whole land surface helps us understand things about attributing the vertical land motion to process. So if we look at North America. We can see this huge uplift the subsidence from GI from four bulge collapse and we can integrate that that vertical land motion field to estimate the flux of material that's rising in cubic kilometers per year under North America and how much is exciting and look at the balance of that. So if you do that, integrating the up part of the field gives you 36 cubic kilometers per year and integrating the down part throughout all of North America gives you minus 16 cubic kilometers per year so there's actually much more up than there is down the balance of that is about 20 cubic kilometers of net up. But because we have these models of GIA we can subtract that out and show that for North America if you do that. You actually get a in the up areas you subtract from GIA and you get minus 5.4 kilometers per year so it looks like that there's either. GIA a little error in the GIA model or there's processes that are contributing to vertical land motion on a continental scale that are net subsidence. The same is true in the area that is subsiding from GIA that the model doesn't fully account for that so the net is downward so so if you forget about GIA. There's actually quite a lot of vertical land motion and it's net subsidence in North America. We can do this this becomes highly uncertain on a global scale because we know so little about the interior of Asia and Africa but we get a similar kind of result net upward. Worldwide the up flux with GIA is 80 cubic kilometers down as minus 60 and we get a net up but if you take away the GIA. You get a net downward around the world that is measured right and we're only measuring in certain places on the continent so this could have implications once we start being able to see the vertical land motion on a global scale. We might be able to say well if continents are are going up overall because of GIA maybe the sea floor is going down net overall because those are the places that are observed. Okay, so that's a little fanciful at this point because we don't have wonderful constraints everywhere across the continents but in summary I'll just say that that. Because we have this large database 19,000 locations on earth we can constrain vertical land motion. Well in some places, not so well in all places but this contributes a lot to measurements of relative sea level rise. There are many processes that we need to be aware of when looking at this GIA is the biggie on earth but there are plenty of others. So we can use GPS imaging for identifying and characterizing the physical processes for estimating and understanding the uncertainties and for asking questions about the budget of up and down across the whole planet. Okay, and sorry I want to go over here. Thank you. Okay, thanks Bill. Yeah, I think what we're going to have to do is hold questions until after both of the talks so we'll we'll go to Manu next, and then we'll have a little bit of time for hope for questions for both of the speakers at that point. So, so next I'd like to introduce Manu Shirzai. Manu Shirzai is a geodesist, geophysicist who's made significant contributions to the field of crustal deformation monitoring and modeling from space. So he specialized in spaceborne synthetic aperture radar and using that measurement tool to study groundwater hydro geodesy seismic and seismic faulting processes volcanoes induced seismicity and fracking and the impacts of relative sea level rise on coastal areas. So, Manu is now at Virginia Tech, and he's also a member of the Center for Coastal Studies at Virginia Tech, and has also been associated with the Southern California Earthquake Center NSF EarthCube, and a variety of other programs so let me turn this over to Manu please begin and Thanks Jeff. Hi everybody. I hope you see my slides. So my presentation is a follow up to conversation that we had yesterday and today. And specifically is a continuation of Bill's great presentation. I focus on vertical land motion from a slightly different perspective. I would touch on different monitoring, modeling and projecting approaches that allows us to learn about the contemporary rate of vertical land motion and its future evolution that needed to evaluate future hazards at the coastal area. So, this figure of variation of this figure has been shown frequently today and yesterday. I want to draw your attention again to this important parameter which is a relative sea level rise. Relative sea level rise, very broadly defined as a difference between the elevation of the sea surface and elevation of the land. And this is arguably, this is evidently the most important parameter for evaluating hazard at the coastal area. To elaborate that I'll show you this figure here. So in the literature, it suggests that that 20 centimeter rise in the sea level would increase with double the frequency of the flooding around the coast of the United States. So here what you see is a very first order projection of the sea level rise at the Galveston tide gate, which is in the south Texas, and the yellow line. So you see that 20 centimeter threshold will be reached about 2014. However, if you add the present day substance rate to this value, which is about 2.2 millimeter per year at the location of the tide gate, that threshold will be reached almost 15 years earlier. So this suggests that accounting for sea level rise, it's a critical factor for future adaptation plan and strategies to deal with climate change. So, in the next few slides, I summarize what is discussed related to the factors that drive land substance fields, discuss the few of these drivers and the presentations yesterday. We also had quite a bit of discussion. My purpose here is to summarize those conversations and put them in a forward looking perspective. So the top panel is across section across the North and North and America that goes from the West Coast of the United States through the East Coast. And based on the geological setting and technique settings, there are different drivers that become important into the present day observation of the virtual land motion. So on a vest to begin with, we have subduction zone that oceanic plates dive under the continental plates. And during the inter seismic period, which is a bit too major earthquake, the overriding plate is squeezed, peaked and causes the uplift. During an earthquake or co seismic event, that accumulated stress is released. And as a result of that, we have uplift and subsidence. And these are different subsidence depend on the location of the shoreline with respect to the patch of the slip, sleep, sleep on the fall. The second mechanism that is of great importance is the compaction of the aquifers. So as fluid is removed from the aquifers, poor spaces close and the entire aquifers compact, which manifests itself into observation, observed subsidence on the surface. This process in contrast to the technique process is very local. However, the rate of this subsidence can be very fast and temporarily nonlinear. The third process that's very carefully discussed yesterday is the GIA effect of the rebound of the crust and of the mantle due to removal or addition of the load. So GIA is one and also sediment loading is another. This is the process that causes the global change in elevation of the ground at the rate of one centimeter per year where ice mass is lost and the subsidence of about one to two millimeter per year at the perimeter of that old ice. The first process that is mostly concerns the basins is the compaction of the sediment, but also specific type of faulting that called gross faulting. So the compaction of the sediment is also very localized but can impact very broad area depending on the history of the sedimentation and thickness of the sediment. So in this slide, I summarize the rates and the time scale of the operation of these processes. So in the middle here, we have the contemporary rate of the sea level rise and its projections to put everything in a context for you. So sea level rise at the moment is about 3.3 millimeter and it's projected that it would increase to up to 10 if we don't do anything about the global warming and management of the ice sheets. So everything that you see to the right of this red row that is SLR are the processes that have faster rates than sea level. So obviously fluid extraction and sediment compaction have rate that is order of magnitude faster than that of sea level rise. However, other processes that we see to the left, such as GISIA and Technic, are often slower than sea level rise. Of course, if there is EarthSphere that would over, that would surpass all these rates and for a short period of time causes the very rapid subsidence or often depending on the location. So we have two major challenges when it comes to majoring coastal land subsidence. We need observations that are especially extensive, hundreds to thousands of kilometers, you know, coast of the United States, West Coast, Alaska, Gold Coast, East Coast and Coast Coast. So they are very extensive and we need data that can cover that. And the same time we need something that is has a management resolution management relevant resolution, we need observation that are relevant to size of a building or streets, tens of meters. And all this observation must be related to a global reference frame to be able to compare with the observation of the sea level rise. So these challenges bring us to a great opportunity, which is integration of the GNSS and interferometric synthetic aperture radar. The GNSS is discussed very carefully by Bill, and I will touch on the siren interferometry and I'll describe in very brief how that works and how these two data can be combined together to address these challenges. Currently, the era of the siren interferometry in the commercial and civil applications started in the 90s by missions such as ERS-1 and 2 and it continues to the present day and it's hopefully continues into the future with the next couple of decades with the planned missions. So the radar satellite offers a different wavelength say CL and expand. And among these missions, I want to just draw your attention to these two, which is Sentinel-1 and 8 operated and maintained by the European Space Agency and a NISAR, which is a joint venture between NASA and Israel, Indian Space Agency, that's hopefully going to be launched this year or next year. These two are game changers, the products to a golden era in the sense that they have open access data policy and provide worldwide coverage at no cost. Anybody and everybody can download this data and do anything that they want to do with it. In the next few slides, I put together a few cartoons to show how siren interferometry works for those of you that are not familiar with it. It's very simplistic. I left lots of details, but the general concept is present. So think about this island. This is Hawaii Island. It comprises two main volcanoes, the Mauna Loa and Kilaoa. Kilaoa is among the most active volcanoes in the world. You want to monitor these volcanoes, volcanic islands with siren interferometry. The satellite flies over the area, transmits the signal, collect the back of the radar signal and generate this siren. Then it goes around the Earth and come back sometimes later at the location. It is close to the first one, but never the same. It transmits the radar signal and collect the back of the siren and generate the second one. So now we have two images, two solid images taken at two different times over the same area from slightly different siren interferometry. By multiplying the first image with the complex conjugate of the second one, we generate an interferometry. This interferogram comprises mostly signal due to the change in elevation and due to the topography. We are interested in the change in elevation, which is the formation. Therefore, we use a digital elevation model such as SRTM-DEM as a reference and subtracted from this interferogram and as a result, it generates a differential interferogram. What you see here is mostly the formation due to change in the signal due to the change in the surface elevation, but also there are other effects such as two first freak and island freak effects that can be corrected using modern and advanced across different things. One thing that I have to remind you is that SAR observations are not vector. So in SAR observations provide us with measurement in the line of sight of Saturn, so not vertical nor horizontal. It contrasts the GPS that provide us with the 3D vector. So this is actually an opportunity that by combining the two, we are able to solve for the 3D displacement field at an unprecedented resolution, and I curious as I will show you in the next slide. If you take our messages. So land subsidence exacerbates the hazard and risk associated with the sea level as we all agree. Then several national and pathogenic factors drive. So that include technique, aquifer reserve or sediment compaction, GI analysis. In SAR enables measuring the contemporary rate of subsidence at management resolution, relevant resolution. So the next topic is the projection of this land subsidence. All we discussed until this point is the measurement of the contemporary rate of the land subsidence. However, we are interested in knowing what happens in the future to develop adaptation plan and resilience plans. So therefore we need to know how this land subsidence evolve in future. In theory that seems to be a trivial task. We develop a model, we calibrated using the contemporary observation and we do a projection to the future. The impact is extremely challenging task. All the physical and socio-economic factors that drive subsidence are now the station and may vary over the time and space. I will show you a few examples. So this is the simulation of the uplift and subsidence associated with the earthquake cycle at the location of the GPS station P403 which is located in Cascadia. Currently this station is rising uplifting at a rate of about 2 millimetre per year. However, that area has been subject of the major earthquakes in the historic time. So the most recent one was the beginning of the 17th century because about half a metre of the subsidence at this location. So we know that it's going to happen but to predict when this uplift would turn into subsidence we have to predict the earthquake. And that's the topic of active research and I don't know if ever we will be able to predict the earthquake. So therefore knowing how the land subsidence and uplift evolves is a challenging associated with the technical processes. This is another example for different factors. So this is the aquifer compaction at the Yongchang aquifer in Taiwan. So a very sophisticated model, a mechanical model is developed to simulate the behaviour of the aquifer for 2007 through 2010. And then the same model is used to predict how the aquifer system behaves in the following year. And note that in the first couple of years, the collaboration years, we had normal precipitation or no droughts. While the 2011 there is a prediction here, we had droughts. And just because of that and changing the behaviour and probably pumping rates, you see the model does a poor job predicting the behaviour. So this suggests that predicting the future of the land subsidence due to aquifer compaction is also extremely challenging. GIA, I hope you all agree that is not a challenging task to predict the future. In case within 21st century, in the area that is ice is not lost today. So we can assume the rate of the GIA, the vertical and motion due to GIA is steady through 21st century. Compaction of the sediment is another factor, another process, sorry, another process that is maybe a challenge. So compaction rates may change in space and time due to the geology sedimentation rate and so on. But here I show you a model simulation. So you see the compaction rates colour coded for different age of the sediment, horizontal axis and different thickness of the sediment, vertical axis. You see that the rate changes over the time, but the change for the time scale of 20, 50 and 100 years can be considered steady. And this is really what we care for management and resilience plan and adaptation plans. A few take home messages. So among factory driving land subsidence, the technical aquifer reservoir compaction process are not steady. GIA can be to a good extent monotony. To develop climate adaptation strategies and flood resilience plan, we need future work to develop one type of objective land subsidence forecast models. These models must integrate different factors, physical processes, socio-economic factors and climate process. And all these models must be calibrated using contemporary rate of the observation of the land subsidence. This is a field of active research and some work has been done in the recent years, but there is a lot more to be done. A few case studies now. So first of all, I will talk about the entire California case study where we monitor the 1,300 kilometer cost of California using combination of the data acquired by ALUS, ALUS is a Japanese satellite operated at the L-band and operation period was 2007 through 2011. As well as the Sentinel, A and B, C-band satellites, we use the data between 2014 and 2019, but Sentinel is still operational and use our GNSS data available to UNR website. Same data as Bill presented to you in his presentation. And you see here a special distribution of different datasets, as well as the location of the faults and faults in the region. Combination of the INSAR and GPS data provides us with this map of the vertical land motion along the coast of California. So the pixel size is about 50 meter and we have about 30 million pixels. So compared to 10 kilometer spacing of the GPS stations, we have almost thousand fold increased in sampling rate. And the features that you see here are very interesting. The signal is dominated with long wavelengths of formation, subsidence in the south and central and between the two is the zone of uplift and also in the north of California we have uplift. The uplift in north California is likely due to the subduction. That's where the sense of deformation changes from the strike in sleep to reverse faulting. The sounds of California is likely to be due to the subsidence is due to compaction of the sediment. As well as in San Francisco Bay Area, we suspect that most of these due to the compaction of the sediment and bay mud specifically here. This object signal is very interesting. We associated that to the technical processes, but not San Andreas fault. These are faults that are sub vertical to San Andreas fault and have reverse sense of motion. And also we have this compaction here in the basins that drive the most of the subsidence, but also there are some pull apart basins that contribute into the observed subsidence. We see all different processes in play along the coast of California. I would like to draw your attention to the zone of the very rapid uplift right here. One is in LA, south of LA, one is in south of the bay and the second one is east of the San Francisco Bay Area. These are the aquifers that are rebounding. But at the end of the recent drought ended in 2015 and aquifers are replenished thanks to the water management plans. And aquifer shows uplift, which is very interesting. So to compare or validate this result, we use a subset of GPS observations. These GPS stations have very small standard deviation, less than two millimeter per year. And as you see here, each circle represent two observations. The field color is the GNSS observation GPS and the edge is the inside measurement. So there is a good agreement, but also there are some disagreement. For example here, and here you see there is disagreement between the two. Most of that disagreement is because the observation periods are different. Otherwise, where we have overlapping observation period for GPS and INSAR, we achieve very good comparison between the two with the standard deviation of difference less than one millimeter, which is remarkable. We did a little bit more with that. We did some exposure analysis. We estimated how many people along the coast of California are exposed to subsidence. This is not flooding. Flooding is an excess of just how many people are exposed to flooding. And we see that hundreds of thousands and actually millions are affected or exposed to subsidence. For example, in San Francisco Bay Area, about 800,000 people are exposed to subsidence at a rate faster than one millimeter per year. And the number for Los Angeles and San Diego is over 2.3 million, which is a very important number to pay attention to. Because subsidence can cause other hazards and damage to infrastructure in addition to exacerbating the sea level rise. Some take home messages for this part of the conversation. So combination of the multi-track INSAR and GNSS data enables measuring BLM along the coast of California at a thousand kilometer extent, 50 meter resolution. We estimate something within 4.3 and 8.7 million people are likely exposed to subsidence rate faster than one millimeter per year. And the last part is very important, last take home message. So our data are available globally and we demonstrated that technologically it's feasible to process and compute this data at a larger scale. However, very little work is done in this field. So this is something that requires special attention by finding agencies and my colleagues to do this for entire world coasts and provide data that is management resolution at tens of meters special resolution. A few cases studies know about how this subsidence allows us to shed light on the future in our nation has. So I zoom into the San Francisco Bay Area where I just show you the subsidence. I mask out the uplift signal so everything that you see here is subsidence. And most of the subsidence occur where we have Baymont and Sediment that are compacting under their own weight. So the combination of this observation of the present day subsidence together with the projection of the sea level rise that are available for the region enables us to tell something about the future of the inundation in the region. But to do that we have to project land subsidence forward. So assuming that most of the observed subsidence is due to the compaction to build a mechanical model that tells you about the compaction rate of the sediments for different age and different thickness. We found out that for the typical thickness of sediment in the Bay Area, which is 10 to 30 meters, the compaction rate is near steady. So assuming a linear rate would cause only 14 percent error in our future projection, which is negligible compared to other sources of error. So without any significant worry, we project extrapolated subsidence forward using a linear model. Combination of those linearly extrapolated subsidence with observation of the sea level rise provides us with many different scenarios of inundation for the Bay Area. I show you here only one example and that's the combination of the sea level rise under the RCP 8.5 scenario by 2100. Assuming on a sea level rise, we estimated about 168 square kilometers inundated. Adding the land subsidence to that increased this amount to 218 square kilometers. So I would skip this next slide, which is a zoom into the positive city and San Francisco International Airport to highlight a few examples. And I would talk about next case to study along with Gold Coast, which is very recent. It's just a paper Megan Miller, my former students currently at JPL submitted to GRL. So what she has done in this study, combine all INSAR data, including Sentinel and ALOS without the vision of the LiDAR and projection of the sea level rise for the Houston and Galveston area, right here in this white box. The observation of the vertical line motion for the area shown here. So what you see here is mostly subsidence at a rate of few millimeters for most of the Houston area and also Galveston. And the driver of these subsidence is mostly compaction of the sediment, residual compaction of the aquifers, some tectonic signal as well as the salt directors. They roll and I believe there is some effect from GIA might be very weak, but still it's likely to have some contribution from the GIA effect. I'm not 100% sure what the last one. So assuming only land subsidence, so magically somehow we stopped the sea level to rise, sea level rise only due to the land subsidence. We expect to have 76 square kilometer inundated by year 2100, which is mostly at the Galveston island and also Galveston Bay area. So adding to that effect of the sea level rise under the scenario of the RCP 8.5 and also adding to that effect of the storm surges. The storm surges are important factor in both the coast area and as they are subject to destructive hurricanes and cyclones frequently every year. And we believe that frequency and intensity of those cyclones are amplified in recent years due to the climate change. So we see that large part of the Galveston Bay and Houston area will be inundated under this worst case scenario. I note that we don't distinguish between inundation and flooding. Inundation is the permanent flooding while the flooding itself is temporary. So the storm usually causes flooding and under certain conditions that will result into permanent inundation. This did not make the distinction here, but please pay attention to that factor. So we estimated under this worst case scenario, something like 1200 square kilometer will be inundated by the year 2100. So if you take home messages for this last chunk of the presentation, it's important to remember even without sea level rise, flooding had not may increase due to continued coastal land subsidence. In San Francisco Bay, we estimated that something between 125 square kilometer to 429 square kilometer will be inundated due to sea level rise and land subsidence by 2100. And in Houston area, something about 186 to 1157 square kilometer is subject to inundation by 2100, given SLR, a different SLR, different storm surge and land subsidence scenarios. Thank you very much. Okay, thank you, Manu. So we've got a few minutes, I think we'll use a few minutes for questions from the committee for Manu and then we'll bring Bill back in and we'll take questions from the committee and from the attendees on that. Let me start out with a question for you, Manu. In terms of, what do you think are really the biggest opportunities for advances here in the future? Is it simply really getting more data from more places or do we need some modeling or theoretical advances to go along with it? I think we need both. So, you know, GPS data are looking great, but we need a lot more GPS station, but also we need to exploit all this open access, open publicly available archive of the SAR data and process them in combination with GNSS data to produce a community model, a community vertical land motion for the entire coast of the United States and the world. So this is the opportunity is available, data is available, we need just resources to get that done, but also we need models. So, for example, sea level rise scenarios are very well established. They have significant uncertainties due to different processes, but the mechanism and processes very well known. Still, we do not have, for instance, a predictive model for the aquifers that take into account different behavior of the, you know, users, you know, different climate factors, which drive the, the, the, the ponding of the groundwater. These are things that needs additional work, future research, and it's interdisciplinary. We need different disciplines, different experts get together to build those multi-objective predictive models to be able to generate something that is comparable to a future projection of the sea level rise. Okay, great. Let's see. Next, we have a question from Cindy Evinger. Yes. Thanks so much for a really interesting talk. I wanted to follow up on a point that you made in your work with Miller, your student Miller. You inferred that you were also using differential LiDAR, or were you comparing the inside of LiDAR, or do you have experience with differential LiDAR that would be relevant to this problem? Right. We did not use differential LiDAR. We used LiDAR digital elevation model. So to, to, you know, INSAR provide us with a higher resolution and to, to generate a map of the foundation that is, again, management relevant, management resolution relevant, we need a DEM that has high resolution and high accuracy. So this LiDAR DEM is the tool that we use, provided by NOAA, I believe, and USGS. So, however, differential LiDAR, yeah, go ahead. Oh, so differential LiDAR can be used to, to study how the, how the coastal communities are evolving, you know, how the infrastructures evolve and so on. But for measuring deformation, I'm not quite certain that differential LiDAR provide us with the enough accuracy for measuring deformation rates at the, at, as slow as one or two minutes. Thank you. Thanks. I think a related question to that, which actually comes from one of our attendees is how sensitive are those the predictions and this would be, I think, the inundation and flooding predictions to the DEM. And do we have coastal DEMs everywhere that are good enough to do that, or do we need better data? Right. So let me answer the second question first. We have LiDAR DEMs with good accuracy for most of the United States coast and Australia and some area in the Europe, but Asia is lacking that in most places. And if become, I believe we have none. So the accuracy is good. So the latest DEM, LiDAR DEM that I've worked with has actually better than couple of centimeters in vertical direction. But any error that we have in the LiDAR DEM will be directly translated into that inundation hazard map that we have seen. And we have very little control on that one. We do always do diligence. You know, we do the error analysis. We provide everything with a certain uncertainty ranges, but we cannot reduce that bias or error in DEM. Okay, great. I'd like to bring Bill back and I've got a couple of questions here. One specifically for Manu and then one I'd like to see both of you address. So the specific question for Manu is, could you say a little bit more about how you combine the INSAR line of sight and the GPS 3D measurements? And then if you and Bill could both talk about the issue of time dependence and time variation, you hinted at it when you were talking about some of the aquifers. And so I think, yeah, if you could say a little bit more about that, and then I have a more specific question for Bill on the same topic. Sure. So the answer to the question is a little bit technical. We build a common filter structures, a structure that combine INSAR line of sight with horizontal GPS displacement. We do not combine the vertical GPS displacement. So, and the advantage is that we have multiple viewing geometry. So, for example, ascending descending from one satellite and ascending from another satellite that provide us with three observations. And GNSS provides us with two horizontal observations, but at the lower resolution. So we oversample GNSS horizontal displacement at the location of INSAR pixel that provide us with five observations and three unknowns. And we build something in the error analysis and adjustment, we call it generalized model, which is a stochastic model. And we use least squares to invert that model and solve for those three parameters in space. And then we use a common filter to reduce the temporal noise. So it described in the supplement of the paper, both paper, the science advances paper in 2018 and 2020, and I have a software available for anybody who is interested. Okay, great. And Bill, are you, if you could turn your video back on. Oh, there, he's allowing me to start my video now. Okay, great. Yes. Yeah, so, so yeah, I think if you could talk a little bit more about the time dependence and in particular I'm kind of wondering about different time scales for time dependence. I mean, I think you use like a two and a half year window. And you'd see more or less variation if you used a longer or shorter window and so on. And if you could maybe address that, and we could hear a little bit about us, particularly for this INSAR GPS combination where we're trying to get a rate from something that might actually be a little bit variable in time and how, how we can deal with that. Yeah, so it's a complicated topic but the, both the GPS and INSAR do allow for estimation of time variable motion right over whatever the period of observation is. Both techniques can be used to look at the variation in rates in various ways, even when they're combined together. So that those kinds of motions are accessible from the techniques. So what so many people are interested in and what so many of the models they are relying on are trends. And so a lot of effort has gone into isolating what is the trend in the data that is most, it's just like climate, you know, we got, we know it wiggles all over the place, but we want to know what trend is it going, we're getting hotter or colder on average. So, so with GPS, your capacity to do that is dependent very much on how long the observation is. And we have stations now that have been around for 27 years, that's great, but most of them have not been around that long. So, so the future is bright, as long as we can convince people to maintain their networks for a long time, where it's good just going to get better and better and better and better. And a lot of the gaps that you saw in the maps are because not not because there's no station there but because the station hadn't been collecting data long enough for me to include it, you know, I have a threshold, it's gonna be there a couple years or something. So, so, so you station may stations are there that haven't had a long observation history so the trends are not very well understood though they they're good perfectly fine for capturing an earthquake or something like that. So we do better on different types of process, depending on status of the network. When it comes to these, these other areas, you know, Asia is not completely blank to the Asians, they have collected a lot of data. There's just not in our database necessarily the raw Rhinox observations are in our database but they're that they're out there and and certainly tables in US journals can be mined to to improve the overall model of trends, because there's data there that that could could improve the constraint overall. And then of course work can be done to develop collaborations with people other countries. Geodetic diplomacy can be employed in order to get more data accessible to labs like ours that are willing to process it. And we put everything online so you can go to our webpage and and we'll we'll process any data for free from a continuous station, as long as the operator is willing to agree that the products can go online for everybody that's that's the deal. So we're willing to do the work if, if the operators are willing to share the results. Okay, great. Let me go next to Torsten Becker and I'll just encourage the attendees also go ahead and put your questions in the Q&A box, because we are looking at those questions and kind of bringing them to the speakers as well as the ones from the committee so Torsten. Thanks to both of you for great presentations I guess, asking about the integration right we've heard that it would be great to have a GPS station and every, you know, tight gauge, for example before. I just wonder if both of you could comment on what are the bottlenecks in integrating INSAR and GPS, you know, I think, you know, for cross-segment processes is clear that the closer to the fault you are you have the correlation and what do we need next to really merge things for an integrated vertical product. And also, you know, how can we fold in, say, the hydrological observations and, you know, things like from grace follow up to sort of correct for, you know, remaining signals. You want to go first, Manu? Please, you go first. I mean, I talk a lot so I'll go next. Yeah, so I mean these combinations can be done between INSAR and GPS very effectively, I think, where the GPS network is great. And so you're dependent on that. Even having one GPS station in the field is an improvement because you can use that as a calibration. You can look at the pixel in the INSAR time series and compare that to the one point where you have a GPS station and that is an improvement. So you're always better off combining, in my opinion, and that goes, you know, that's a lesson for both the GPS people and the INSAR people. And they're both demanding techniques so we're not all great at both of them and we have to get together specific projects to do the integrations to best effect. So I mean, that said, the devil's in the details because you might have a great GPS network now but all your INSAR data was collected between 10 and 5 years ago. And so you have excellent products but they're collected over different time periods. Now we have global Sentinel, so that's less of an issue, but really the Sentinel time series is on the order of, I don't know many of what, six, seven years. About six years now. Yeah, so if we had a GPS station for six years, we would say that's good enough, we'll get a vertical rate from that. INSAR is like a coverage but has greater uncertainty in its individual pixels in terms of rate. And so you might want an even longer time series for INSAR if you want to get down to the rates that we come to expect from a GPS, continuous GPS station. So we don't have very, very long time series unless we start stitching together these different systems, going back to NVSAT, USA. So we're doing all, like you already mentioned this, combining all the data into one giant analysis. It's very demanding computationally, but certainly within the bounds of computer systems we have now, it can be done. Right. I second what Beal said, I don't think that at the moment question is the technology and methods. So, and the combination of the GPS and GNSS has been done by adding more GPS and INSAR has been done by adding new satellites and new GPS station. We just make the outcome better. But the technology is out there. What we need is infrastructure and somebody who want to do that. So we need a push for that in a systematic fashion and with resources made available to that kind of performance. Okay, thanks. Next question from Diego Melgar from the committee. Hey, I have a quick question for Bill. You know, in geodesy we obsess so much about the monument, right, to make sure that the station is measuring what the earth is doing and not the monument sinking into the soil. So I wonder, you know, we've talked about the importance of tide gauges so much. And whenever I've seen a tide gauge, it looks like a downwards pointing radar antenna bolted to the edge of the tide house at the end of a 100 year old wooden pier. So when you're making the tie between the GPS and the tide gauge, I wonder if the stability of the tide gauge monument is ever an issue that people contend with. Sure it is. I'm not a tide gauge expert. And, you know, I've built some GPS stations. So I usually know what's up with the ground when I'm installing. So I can develop a sense of confidence or lack of confidence about what it might measure. Tide gauges, I don't know a lot of the details, but it is, I know it's a well understood technology. It's been around a long time. And of course there are, there's lots of things that go into building a good tide gauge and I'm sure they're not all the same. So the folks who are in touch with these kind of data, I'm sure there might be someone out there who's better at answering this question than me. Of course it's a concern. And the water's going up and down all the time, but they're measuring the average level on a given day or a given month that is something that's insensitive to waves and the systems are built to deal with that. So I don't know, maybe there's someone out there in the audience who can comment more on the gauges themselves. But the one thing that tide gauges do have in their favor is longevity, right? Build the thing and it sits there and it can be there 100 years or more. And that's an extremely stable kind of measurement if the thing is taken care of. So NOAA in the United States and other agencies in similar agencies in other countries do do leveling so they'll do leveling from the tide gauge to network of tidal benchmarks which are back on off the pier essentially and on the ground. And so those are done and that's, in theory, those data are available whether those are easy to access in terms of the long term stability and other changes is another question. But that's the purpose of that is to really be able to distinguish between motion of the tide gauge that is not representative of motion of the broader area. In cases of that, Jeff, with just comparing GPS and INSAR to tide gauges where the tide gauge was sort of the odd man out in the measurement of vertical land motion. And so we, you know, you've got two that look the same and one that looks different you tend to trust the one that the two that agree. So multiple techniques are of course, advantageous. Yeah, so I take a question that actually was posted back during your talk bill. So you showed in when you're showing North America there was a sort of a very mild or moderate uplift that was east of the Mississippi River mostly in Tennessee and Alabama. And it wasn't entirely clear when you remove the GIA model whether that was still there as uplift or or not but what is, what do you think is the cause of that and I guess more generally when you see features of that sort of scale. What's the process for trying to figure out, you know, on a discovery basis what what might be behind those, those anomalies. That's one of the things I I love the most about about doing the work in this mode and discovery mode just look at the whole field and see what's out there and see what's not explained yet. And that that signal in Mississippi is not one we've studied specifically we've gone after some other specific up to signals and to really dig in and what is it. And I can, I can say it's, in my opinion it's not GIA, because it's way down by the Gulf of Mexico almost. And it's a blob on the map that doesn't look like the broad GIA kind of signal. So my, my guts say it's not a GIA kind of thing. It's also not a plate boundary. There's no, there's not many there's there's a few earthquakes out there you know the new Madrid seismic zone is not very far away. But we don't feel like it's post seismic relaxation from some big earthquake that happened not too long ago because we know such a no such earthquake and we know of no such plate boundary that would cause that. So we do know that there's aquifers there and that there's there's use of the aquifers and there's a lot of management of that. And we haven't gone into it to more depth than that but my gut feeling is that it's a it's a hydrological unloading signal of some kind. And what we would normally do is go in and start looking at the seasonality component to see if if the stations go up and down with greater amplitude in that area. If the rates are inflecting in ways that coincide with grace measurements of terrestrial water storage. And so there there is sort of a sort of have a quiver of tools that we used to go after signals like that and pull them apart and see how much of it is correlated with water. First of all, in the absence of other explanations like tectonics and the quakes. Okay, great. Thanks. We've just got a few minutes left. I just to call everybody's attention to it. Torbjörn Tornqvist put sent a link to me. It's a paper on tide gauges and how well they capture vertical land motion so I just put the link in the chat. If you're interested in following that up. I do have another question that came from Torbjörn as well. Which is for Manu. What, what are the sort of the progress in getting INSAR to work in coastal wetlands when you get into the areas that are sometimes inundated sometimes dry sometimes kind of in between I guess in terms of being able to get the land surface motion out of that is that something that you think that there's some good prospects for. That's one of the really challenging part because you know radar signal is a radar or INSAR is a coherent imaging technique so we need to observation that are coherent in time and changing in the surface power in wetlands actually defy that requirement. So one one approach is to use a corner reflectors and install them on the wetlands. But the issue with corner reflectors such as similar to GPS is that they need their monuments and monument will be anchored to a certain depth. So the shallow subsidence might be lost. So for that reason, we might want to use other instruments such as ARSET. Which I can remember stand for what but is a tool that's used for measuring very shallow compaction of the sediment in wetland and in fact Torbjörn is one of those has used that a lot in Louisiana basin and Mississippi earlier. So the combination of the two in a corner reflectors and this ARSET instrument, I think would help us to fill that gap to measure the full spectrum of the deformation in the coastal areas in the wetland area. Okay, great. It's almost 145 I think it's within seconds of hitting 145 so I think in the interest of minimizing everybody zoom fatigue I think we'll thank you very much to both the speakers will take a break and reconvene in 30 minutes. The chat can disappear I don't I think this will I think this meeting will will still be there anyway. I'm not sure but if you're interested in that link I would for all of you I would I would click on it from the chat before the chat goes away. So thank you to the speakers will take a break and reconvene in 30 minutes and when we come back Maya Tolstoy is going to moderate a discussion with all of our speakers, both the two from today and the four from yesterday so thanks to all the attendees and we'll see you back here in 30 minutes. All right, welcome back. I'm excited to moderate a discussion panel with all of our speakers. So we'll take questions from both the committee and from the audience. And as a reminder, audience members can click on the q amp a box at the bottom of the screen type in their question, and remember to press send committee members just raise your hands. I'm going to just start with a brief question of my own and I want to also start by saying again what a marvelous job all the speakers did. In particular, I really appreciated the sort of in depth understanding from multiple perspectives that quite how complex this is, and how relevant Solid Earth Sciences are to it. You know, Pippa in particular talked about the importance and complexity of the different feedbacks. And I want to sort of think about how do we help. As a community how do we help foster the necessary communication between the different, the different disciplines. And one of the things that this committee does is try to bring together people from different perspectives different agencies different disciplines to help sort of provide a bridge for people to have these discussions. So I'm curious. As a sea level community. How do the how do you as the panelists feel things are going in terms of broad cross communication between the different disciplines and are there more things that that should be happening could be happening, whether it's from a funding perspective, or you know we always want more data and more funding, but also you know meetings forums, what kind of things are working well at the moment and what more should be done to sort of enhance the important communication on this on this really vital interdisciplinary problem. And whoever wants to go first can go first or I can start calling on you. I can go ahead and speak to that. This has been Hamilton. So you make a really good point that there's this need for this interdisciplinary research to tackle some of these issues so one effort that that I'm involved in is the NASA sea level change team so this is an effort a science team that was spun up by by NASA recognizing that a lot of these problems are interdisciplinary and the need to bring scientists from different disciplines together to tackle these issues. So Steve Nerum was the first team lead on the team lead of that team now and we've kind of been finding our feet and how to really engage across disciplines and really start to do these investigations bringing different pieces together and it's certainly a work in progress. So there are I think funding the agencies recognize a need to address sea level in this way in an interdisciplinary way and bring the pieces together I still think a scientist we're trying to figure out exactly how to take advantage of the different areas of expertise. I should also note that's, I mean that's a group of 70 ish sea level scientists. Additionally, there's all these other international efforts and other pieces going on that we need to pull together so it's a really a big organizational effort international organization effort to to pull the necessary pieces together so I think that funny agencies recognize this and are working towards this again to still work in progress I'd be interested to hear other thoughts from other panelists in their experience. Thank you. Who else can comment on this. I'll jump on after Ben so I'm also participated in this round of the sea level change team and like Ben said this has been a current effort of NASA's to fund this sort of interdisciplinary work and that's fantastic. One thing about the NASA sea level change team is it's very centered on, if you will, certain timescales, certain kinds of observations which viewing the whole realm of sea level science is very limited I mean not surprisingly the NASA sea level change focuses a lot on satellite data, which if you're talking about sea level is the last three decades or so. So something like the NASA sea level change would not facilitate the kind of interdisciplinary work to understand 20th century or common era or Holocene sea level change so I mean in an ideal world. There would be interdisciplinary interdisciplinary programs like this not only at NASA, but through agencies like the National Science Foundation, you know, funders that have sort of in mind in their portfolio sort of broader timescales and that sets different driving questions because again the NASA, the sea level change is increasingly driven by sort of these, I don't want to call them impacts but applications. And that necessarily favors a certain kind of analysis a certain kind of question. And so again, maybe broadening the number of opportunities both domestically and that's also sort of, you know, in other countries internationally. And I think that's really necessary and perhaps the sea level change team is one data point that could be, you know, towards a blueprint of how to do that well. Thank you. May I add also to this conversation. Please. So, I have been outside the NASA sea level change team and I really enjoyed that. And I, but I think that the type of questions that I'll ask should be different, you know, the question should be driven by the stakeholders that are affected by sea level rise and hazard associated with that and the scientists try to provide input to solve that so my my dream team would be to have scientists next to social scientists you know the basic scientist social scientists engineers, lawyers and policymakers together to come up with the with the with the framework that ask questions that impact eventually the ordinary people or economy and so on. What we do as a scientist and we enjoy and don't take it wrong. I have a great time. But sometimes I have the feeling that there is a gap between my outcome and really the person who is the end of the shame that is going to be affected by the sea level rise, eventually in the house the person that loses the house that the industry that's going to be affected, or the habitat coastal habitat that's going to be lost. So, I think we, you know, future report shouldn't integrate all this, if possible, and I think if we want to make impact we should make it possible. Thank you I think that's a really important point and I think you know it's something. I mean I think everyone's thinking about this now but also the racial inequities in the impacts of climate change are really profound and you know sea level absolutely as well so I really appreciate your making that point. Can we hear from the other three panelists what what what their perspective is on this. Yeah, I can, I can add a little bit I mean, there are some initiatives that try to kind of bridge those gaps more and Pipper is leading one of those. And Jeff and a lot of people here are aware and involved in this, which connects kind of the solid earth dynamics and the ice sheet dynamics. So I think that you know GA modelers and ice sheet models I think there is a lot of interactions but I think I think that could be more interaction and and platform for interactions towards the other geophysics communities seismology, geodynamics mineral physics. I think you've got from both in my and Pipper's talk that those are really critical links, and there are definitely connections and, you know, I think every you know GI model has probably collaborations across different, I think the thinking about workshops or platforms that really fosters those interactions I think those exist less. And I don't know if you see this differently. And I think to Manu's point I think that you know the co production of knowledge and how we drive that forward it's another really important important aspect and we can also talk about I mean it's, I see it's a little possibly a separate aspect but that I'd be really interested of talking about more as well. Yeah come in there. And the program that Jackie refers to it. It looks at interactions between ice sheets and the solid earth specifically. And it comes under scar, which is the scientific committee for Antarctic research. And we essentially get funding to put lots of different people in the same room and see what happens. Which it can often be a little bit quiet to begin with. But we've got some amazing discussions going on and I'd say it's actually really influenced the direction of the science in the last decade I gave a talk at the start of this program and at the end. And I checked my slides from the first talk. And there's stuff eight years later which I didn't even mention in the first one. So it's yeah it's difficult to get those people in the same room. Something I'll also reflect on is trying to program the recent scar conference which ended up online unfortunately. It's all to do with Antarctica. And there's often many, many different sessions running in parallel. And this year we specifically tried to reduce the number of sessions and essentially make people make a choice of well I should go to something. And that one, it's not quite what I do. But actually I wanted that person to be in there and listen to something different for a change. So just using that approach in programming some conferences is something to think about. It's a really smart idea. Thank you, Bill. I agree with what has been said so far. I think we're also seeing currently a positive impact of having a lot of data streams being open available for anybody to use without any kind of permission or access or to reduce the boundary to zero. So people getting access to data and lots of it in places where they didn't even know it was present before and giving them the ability to generate to the awareness of it. First of all which relies on people like us going out and telling everybody we know that it's there. And what can be done with it and what potential exists and encouraging people to contribute their data in ways that are open is sometimes they're not interested and they do convincing. That's that's a very general statement but I think we're seeing the benefit now of people kind of getting on board with that and thinking it's a really good idea and then the net gain is worth whatever potential risks are in sharing data and results from analyses and getting them online and and explained and easily accessible to the maximum number of people. Another really important point I couldn't agree with you more on that and and thank you you brought that up in your talk also I think and with something in the marine seismic community we've we've started to get on board with and it has made an enormous difference it's really really important. Okay now I've got a bunch of raised hands from the from the committee so I'm going to call on Steve next you want to unmute yourself. I mean this my question kind of relates back to my as an initial question and somewhat what Bill just said but it's probably mainly directed for Jackie and Pippa but you know we heard about all these great new data sets relatively new we all have the defamation of the solid earth in there somewhere maybe not exclusively but like grace data we didn't talk a lot about grace but there's a lot of great GA signals and grace bill was showing GA signals and yeah I don't see and maybe it's just that I'm not kind of an active member of the GA community but I don't see those data sets being used extensively to improve GA models or improve isolating models. And what's so important is that because these data sets have multiple signals mixed in or or am I have I just missed something here. So I'll go first on that and there is one way in which we're using those satellite data to get at the GIA signal. Specifically in areas which still have ice sheets and that's where we're combining the grace data with altimetry data. And both of those data sets contain a signal due to ice mass change or ice volume change and solid earth rebound. And so by combining those data sets with their different sensitivities to ice mass and GA. There are approaches that back out the GIA signal there. So that's a specific one which is carried out in Antarctica. On a global scale. I think that there are efforts. I'm not not so much involved in global scale GA modeling, but there are efforts to use the grace data and improve our understanding there. Yeah, one of the big questions is that grace is so good at measuring everything. So you need to have a good understanding of the different components that go into it. I guess I can add to that. I think it's mostly use grace and the GPS data in an article in Greenland. And I would say there it is used to test and look at the GA models. And then of course, yeah, for the Laurent refining the Laurent had ice sheet history and the finish scanning ice sheet history. It's used. But I would agree with you actually that it's probably not used as much as it could and should be and I really enjoyed the talks today and took a lot of notes because I found it actually really inspiring so that you know this is potentially areas that could be bridged and incorporated more but but it is also I mean I six G includes GPS data and gravity data as constraints on the ice sheet model. So they are they are being used, but maybe not extensively and as widely, you know, as they should. All right, Jeff, do you want to ask your question. Yeah, so I think this question is more aimed probably at Chris and Ben. So I mean today, Bill and money we're talking about the variety of timescales in the vertical land motion signal and of course there are in the in the oceanographic components that the signals that purely come from the ocean. One of the things that is striking when you look at the altimetry, the sort of spatial map of present day sea level rise or sea level change rate is that even with 25 years, it doesn't look like global mean sea level at all I mean there's very very strong differences and what my question is when we look at the multi decadal and maybe even centennial scales what do we know about variation within the ocean. Do we understand what is is driving. You know, let's say if we had 100 years of altimetry how close would we get to GMSL versus looking at longer term patterns how much do we know about those longer term variations within the ocean. Sure, now I can start and then you can certainly add to add detail or whatever I've missed but. So there is work being done to assess the extent to which the altimeter trim map represents what we call the forest response of Steve and John Fasulo have been doing quite a bit of work on that. And you say you have to start bringing in model projections of sea level in order to make that assessment so given the current length of the the altimeter record, it's still reflective of natural variability certainly and a lot of locations. So as the altimeter record lengthens with the launch of Sentinel 6a and Sentinel 6b, we would expect that trend map to keep shifting in some areas. But as John Fasulo has shown in certain areas it's possible that forest response. So I mean that I guess to maybe put this another way we don't expect the ocean to be flat obviously the longer or the sea level rise to be flat the longer the record. So that's the racial distribution of these trends associated with the fingerprints the GRD response and also a stereodynamic signature as well so we can make that comparison with models and in some locations we do start to see, potentially see that signal emerge, typically where you have less interannual to decadal variability that's where you'd expect to see it first so. I don't know I mean certainly maybe Chris can provide another answer but I think all this is to say that, and we've heard this multiple times continuing these records the satellite records. It's really critical to try to make that assessment what is forced sea level change what we might expect to continue on the future what is just reflective of still natural variability and the record so it's an important research question and again really points to the need for for continuing these records that we have. Yeah, so I'll add on to that a little bit I like that then made the distinction between, you know the so called forced response versus the component of sea level variability observed say by an altimeter that reflects sort of a noise process, say natural in what Ben said is valid there but I'll I think what's important to reiterate also something that Ben said is. The other speakers can correct me but I'm unaware of any process in the ocean in the solid earth, involving the crisis that would lead to a horizontally uniform change in sea level at any time scale in the deep past or in the present so I don't think we're ever going to see that in the uniform change but I think a deeper question that you point to is a really good one, which Ben been hinted to which is, you know, if we have that 100 year altimeter record. What would it look like, I know that Ben has done some work trying to reconstruct that pattern also carling hey has has looked at things like this and more recently Thomas Frederick so well you know what is what is the pattern to the best of our knowledge look like. This is the 20th century. And there's a there's a lot of uncertainty there for instance to the extent that we know changes and land ice mass, we can have a good estimate of what the associated spatial patterns of sea level change are. But one thing we don't know very very well is know what patterns are related to to ocean circulation changes in other words do, if you have that 100 year record. Now, in the 30 year altimeter record if you look at a map like like Ben showed it really is the most charismatic features really are ocean dynamical a lot of people have focused on those really small scale. Signature features that are associated with music scale eddies and instabilities of strong Western boundary currents. I'm not sure what that would look like in 100 year record it's not clear if, if sort of the more grd the gravitational rotational and deformational signal to become, or not the deformation but the gravitational rotational signals will become more apparent we don't know. So I think there's a lot of work we have to do as physical oceanographers to understand the low frequency variability spectrum of ocean circulation, certainly we can interrogate models for that there are important questions about how realistic the models are on those long time scales. So there are a lot of open questions, I think, so it's a long way of saying I don't know. Yeah, that's that's good. Good to hear where the things are I think maybe a quick question that one of you could answer is, in terms of the steric forced steric component of that what what sort of magnitude does that reach in terms of. Most outside of the proximal areas most of the fingerprint from sort of gravitational redistribution most of those are relatively small outside of the proximal areas to where the ice is really changing. But, yeah, are there centimeter per year level steric variations and what, what, yeah what's the real magnitude of those. Yeah, I mean it depends on timescale I mean to Ben highlighted an important region yesterday and his talk is if you if you look at the center record and you look at somewhere like the Western tropical Pacific. These trends that you see again really enhanced rates, then I'll correct me if I'm wrong but you're looking at something on the order of a centimeter per year for the last three decades. That's primarily steric specifically thermo steric related to changes in ocean heat content either related to local addition of heat or changes in ocean circulation or transport that converge heat in that region so certainly on the multi decadal timescale on those regional spatial scales, you can see magnitudes of that magazine values of that magnitude. But as I you'll probably get sick of hearing me say this I'm kind of like a broken record but it will depend strongly on space scale and timescale if you zoom out to the global average magnitude on that much as we heard said, the last you know decade a decade and a half that steric signals much smaller on the global average and contributes about one third to the total, roughly similar values if you zoom out to the 20th century a little lower magnitude so so it again it is a question of spatial scale. And, and it even works in the other way I mean as you go to sort of smaller scales. And, and shorter timescales, you'll expect even larger magnitudes. So. Thank you. Another important point it's it reminds me of old politics are local and apparently all C level is local. With that I'm going to call on Thorsten. Thanks Maya and thanks again to the speakers. I wonder about the importance of open and co located data sets right and the importance of really doing even more to bring the communities together and discuss these interdisciplinary program something that can be set for for many, many fields of study but I wonder about the dynamics and the modeling aspects of some of these interactions and I think they're there is perhaps even more of a challenge because the communities that have run say numerical models in the solid earth realm are traditionally distinct from the ones who modeled the cryosphere and the climate, even though we've been saying for decades, there's commonalities between how rocks to form and ice to form. And so, and I think the community is also at a different level, right, we have a sort of strong support structure and car and others is, you know, assistance for the climate part and I wonder what what our panelists think about, you know, the community and the resources that might really help to accelerate the research there and I think you know Jackie already commented on some of the, some of the questions there from the GA modeling right where models with lateral regions recently become available, but there's a need for benchmarks and we've, and I just wonder if you could sort of build on that. Where, where would you see most sort of synergies and how could computational community infrastructure help with these conversations across the disciplines that we'll probably need to continue to have. I would. Sorry. I thought I have Lourston is is when when looking at the talks, or and when preparing my talk I, I saw there's a lot. There's GA models out there that they're accessible and can be downloaded and people can think about them, maybe not all of them are there but they they really help us understand what what's the plausible kind of signal that we're going to get at it at a GI and these things are available and we also have models that are accessible plate tectonic movement like horizontal motions of the plates, you know, there are vertical models that are accessible we can get that. But we don't have a community model for vertical land motion associated with plate tectonics, like the subduction zones, Mount building in in convergence, conversion boundaries. But I mean if we had a new class of model that that could be available for for everyone to access that was just the tectonic part of vertical land motion that would be a, I think a very achievable bite that could be taken and worked on by a program person. That's that's one thing that could be done in this in the spirit that I think you're suggesting. And there are other things like like the hydrological loading that's accessible through grace, we could we could have vertical land motion that's, you know time series of that people working on that now so so in the future I think we're going to have from better perspective on atmosphere and hydrological loading effects at the time series level that can be accessed and compared to all these other things to see if they are the explanation for the signals we're seeing. Just a bill like let me push a little bit more there for example you mentioned the predictions from some of the GA models are available but you know, often, it's not the whole workflow that is available and we've heard how things are happening so it's might be derived with this costume models that are inconsistent with some of the things that people want to do and I guess, is there an opportunity to, you know, to sort of not have just the final model but also some sort of framework to to get at a community understanding of how you get there and I think similar things apply in all kinds of aspects of this problem. Are you asking me. Yeah, yeah. Sorry, I was just going to chime in here because I actually it was really great to hear that you thought that GA models are accessible and and available, because I actually think that the GA modeling community, compared to the other kind of computational you know I think is behind on sharing code development as well as sharing model output. So there are some available but I think it's quite limited and it there's also no centralized platform. I think it will be fantastic to have you know like iris like a centralized platform that has the different ice models that are available or different GA predictions and those could be, you know associated with the different publications and I think there's definitely a trend towards making that more available as supplementary information, but it's not been done traditionally and for example for the ice sheet reconstructions it's really a limited set of reconstructions that are available. And, and I think this and you know and then limits my work. And I think the same could probably be said for the GA model output and that limits other people's work so I think there is a lot more data sharing that could be done. And from the model output I think in parallel on the observational data. And I think there's a movement towards that which I think is great but on the observational data there's definitely been a strong push for that in the last five years and the Holocene databases and last interglacial C level databases are being more and more compiled and synthesized and standardized. And that's super important and just really really useful. Thank you so I think Mark has a has a follow up. Okay, can I actually build on this but I think this is a really important topic and I'm liking what the other speakers have said so far so I'll build on this. Yeah, I think there's a lot of room here a lot of room to make models and the workflow more accessible more public more available for the things like GitHub and other open source things. So, I'm not sure if this is obvious but my perspective on this is that yes we need to do this as people are suggested with the GA models and the electronic models and the geophysical models but there's also a lot of room here for probabilistic models, you know, having probabilistic frameworks that you know as we start collecting these these new data sources and new models having ways to synthesize them and making that making that available because you know with all these communities with both the geophysical community and the probabilistic modeling community. The room tends to be dominated by a couple personalities right and I and I would love to see more accessibility and equity there so that more more folks feel able to contribute to that and have access to that work will not just the results. And I wanted to do more things actually really quick one is I think we need we need to prioritize funding people to be doing these things is because commonly, I mean, there are certainly programs to the exception but but often, you know, take this is all labor intensive right sort of creating generating data making it available taking the time to sort of, you know, mark up your code well and I mean this is this is all time consuming and I think that I mean is sort of my idealized world that there would be, you know, the recognition of funding agencies that these are priorities that these speed science in progress and that in that fund should support scientists to do that so that we don't feel completely stressed like when we're publishing the paper like I don't want to take another week, put together my supplementary material, even though that would really help the community. This is a very important point to, to note right where there should be funding support and there should also be perhaps better infrastructure to help the individual the eyes with that task so that we don't say oh we should wouldn't be nice and somehow it has to be supported. If I could just come in with a comment as well thanks Maya. I think we've sort of flagged up a common issue with the word model in the last few minutes. And so one is being able to run a numerical model and one is the model which is the output. And I, I agree with Jackie I think the GIA community is a little bit behind in providing a model that everyone can download and run. That's there. There are a few out there. But that's an area where a little bit behind places like glaciology and ice sheet models were sort of started to be developed at that point where people were sharing code and they're much more accessible. And just a quick comment about GIA model output. I think it would be feasible for for us to have a repository where people upload their model predictions of the global GIA field. I just liked it's a little bit of an issue with me about how we define GIA. It's the response to past ice sheet change past ice sheet changes stuff that happened last decade. And if you define it, take it to that extreme. So for somewhere like Alaska, if we're looking at the maps that the bill was showing, if you take out something like ice six G which looks at the LGM ice sheet to present. It doesn't account for recent ice sheet change on Centennial scale. And that actually should really be counted as GIA. So that's part of the reason why we're a little bit cautious to release our models because we sort of know they're going to be wrong in those areas with recent ice sheet change. And just one more point it gets a little bit technical and it comes back to Steve's question about grace. In those areas where we have contemporary ice mass loss, and we're trying to ask the question of, okay, what in the signal is GIA and what is ice mass loss. If we're in an area with low viscosity upper mantle, the ice mass loss could be triggering a viscous response. Usually when we account for the ice mass change, we assume it excites an elastic response. There are areas like Alaska, like the Antarctic Peninsula, and the Amundsen Sea area. We're actually that standard approach of separating the GIA and the ice mass needs to be thought about a little bit more carefully with low viscosity mantle areas. That was just a personal little point. Thanks. Thank you. Mark, do you have a follow up, I believe? Yeah, it was actually more just kind of a comment on one of the points Jackie made that, you know, it's an interesting time right now in terms of computational geodynamics, geophysics, because two of the big programs in this area, CIG and CSDMS are both basically up for renewal within the next year. And I was just wondering whether there had been any conversation in the community about, you know, sea level could be a great new working group for either of these two NSF funded organizations. And I was just kind of curious whether there had been any discussion about pushing for that in the renewal of either of those programs. I can talk about this just very briefly. So CIG has a sea level code, a 1D sea level hosts a 1D sea level code. And then they also host aspect which is which they're incorporating 3D GA. So it's on their radar. I haven't heard about establishing a new working group around it because I guess the working groups are often around specific codes, but maybe not necessarily so, and this would really build on other codes. But I think that is really interesting. We had a meeting planned for this year, which is now postponed to next year where we actually have a paleo sea level meeting where we invited some people from CIG to talk about open software development and acknowledging code development more and funding it more. So, you know, we're talking about those connections, but it's interesting that you bring up of actually making a formerly a working group a part of any of those. And I haven't heard any conversations in that direction. I mean, I think that would be a really exciting direction to go because I think you could gain more traction in the community. I mean, absolutely. That's right. Aspect is there, but the kind of thrust of aspect has been, you know, mental convection and then trying to build and let the dynamic deformation long term tectonics. And you know, not all of the working groups at CIG have always been that way like the magma dynamics working group started before there was any code there at all. So, you know, and it just might be a way of getting more traction. And I think the same for CSTMS, which focuses on the surface processes bringing in connections to the groups there as well. Anyway, it was just a thought in terms of kind of trying to move the ball along a little bit. All right, thanks. Cindy. Yeah, well, can I keep going a little bit with with that idea about. So, there are two prongs of what I wanted to talk about is that many of the CSTMS and CIG have education components, a large, a big group within a big push within those groups is to to maybe even enhance or increase and also cohort building so you know, the CIG level community has scattered smallish groups without and so you need the global cohort. Well, same problems in other areas. And we, we all recognize the need to communicate across disciplines. So, so one part of the question was about joining forces in the education about C level and communication as a, as a really sale or push also for the climate change issue. It's an easier sell. But the other thing that I'm wondering how easy it would be to put together a toolkit for local communities. This goes back more to Manu and Bill and I was just thinking about sample data sets demonstrations that we can take to high schools in coastal areas that we can take to university students and go beyond. Oh, if the ice mountain. There are some examples on, on UNAFCO, but they're not the same as the compelling stories that we're talking about with inundation levels in Houston or examples that start to link people to the process and to the challenges so I guess those both are kind of related to education so I'll throw that out there. May I have a comment. Please, please. So, well, we, we tried that a few times to, you know, present some of those results to local people. But the point is that it caused often panic. Then actually there we, my, my experience was that the way we presented that as scientists was not the best way of doing it, you know, so we asked all of the question, what is the rate of, for example, subsidence or what is the rate of sea level rise? What is the area that's going to be inundated? For people the question is what I should do when I have this amount of sea level rise? And we were not able to answer that question and it caused more confusion because, you know, for, for ordinary people, this number has very little meaning. So if sea level rises in this area, five centimetres in total, what's going to happen to them and what is their option? What is the health they get? What is the damage and so on? So I think taking this maps and this data to classroom probably had educational advantages but I'm worried about the consequences if it's not translated properly. Sorry, I didn't mean to say, to take your science, I didn't mean in that way, I was trying to use as an example how compelling the challenges are and you know where I'm in New Orleans and Tours and Tulane as well with Tours. But I just wanted to ask whether there were ways that we could teach high school and university students some of the challenges that we have and link these to people and consequences. I think many of us are trying but maybe as a community we could have some tools or some kids and also maybe excite some students in going in this and working in this direction for research as well. So I mean one comment here, I think Manu is, he outlined a lot of the challenges of trying to communicate the science to decision makers. I think what you're referring to is a little bit of different terms of education of high school, college students, things like that. So prior to coming to Philippine and California at the Jet Propulsion Lab, I was a professor at Old Dominion University in Norfolk, Virginia, and I was teaching a 100 level understanding global climate change class, and Norfolk, Virginia where Old Dominion is located is really impacted by coastal planning similar to not in the exact same way as where you are but certainly there's issues there. And I taught this course several times and it was only after trying to bring in those local examples and connecting to their everyday life. I mean there were days where there was flooding outside it impacted their ability to get to school to come to class things like that. So I started to resonate with them. So I started to try to tailor almost everything, all my discussions about global climate change to the local experience and a lot of the things we're discussing here we can link back to that local experience. So you can convey that that science to their everyday life in a pretty clear way. Certainly that's going to be more challenging for someone living in the interior of the United States or a different country but so maybe to get more to your question. So I started to make these resources or go through the process, the time consuming process of getting these resources, these teaching resources, but I found that other professors, other teachers and other locations were struggling with the same thing. So there was kind of an informal network that was being set up of people sharing ideas and materials for class. So, so I would say that the interest is there and exactly what you're describing I think there needs to be mechanisms to and funding to try to establish these links in a more formal way and allow people to be connected. I think the resources are out there I think that they're just sitting on one person's computer or one person syllabus, as opposed to being shared more broadly so I think your comment really resonates with with my previous experience and hopefully there are some funding programs to develop these kits and to extend our science in that direction. Yeah, I certainly understand and if anybody wants all kinds of pressure or temperature. As the hurricane, I goes overhead and passes. More than welcome we've had quite a few so yeah. But no, it's a lot easier to teach when they've just been through one. Well, I think that's a great place to end it that we are unfortunately out of time but thank you for for all that input and, and then that's particularly helpful advice and I think it also goes back to my new original point about engaging stakeholders and I think it's really important that we communicate with local community leaders and local community NGOs and find out what is useful for them as well. So, thank you for for ending on that very important points Cindy as well. And now I will hand it over to Steve to wrap things up but once again huge thank you to all our speakers and right Steve, take it away. Right. That's not Maya. So I'm going to try and wrap this up it's kind of hard to just still down. You know, three hours, six talks that we've heard the last two days but I'll give it a shot here. So first I think we learned that satellite measurements are providing a wealth of new information about seeable change, including sea surface height. And that are Jason three, but also in about a week now we're going to launch Sentinel six Michael Freilich from Bannonburg go continue that time series. Also ice sheet height. I said to and cross at our examples there and I sheet mass changes from grace follow on, as well as inside observations and a variety of other tools that we have. They show, you know, collectively, they show that seeable has been rising since the early 1990s at an average rate of about three millimeters per year. But the rate of rise is accelerating by about point one millimeters per year every year so back in the early 90s was about two millimeters per year and now we're over four millimeters per year over the last 30 years. And the same satellite data and other data sources have shown us that this is we know for certain that this rise in sea level is caused by ice melt and thermal expansion of the oceans and the acceleration is being largely driven by the melting ice. And all of this is due to the warming of the earth. So when there's, there's considerable regional variation and sea level change due to the, the fingerprints of the ice mass loss on sea level, and then variations and where the oceans absorb the excess heat. And so understanding and ultimately predicting these regional variations is going to be key to understanding future local impacts of civil change. The melting ice and thermal expansion not only cause a change in the volume of the oceans but also lead to changes in ocean circulation that can impact sea level as well, especially for the western boundary currents. But the largest uncertainty for projecting future sea level changes determining how quickly the ice sheets are going to melt. And so their understanding ocean ice interactions is one area of current research that will be critical here. Understanding how the solid earth responds to loading and unloading of the ice sheets is going to be critical for understanding ice sheet dynamics. So this latter topic is going to require a better understanding of the internal viscosity structure of the earth. So, in addition, studying past sea level change and how the solid earth responded in the past to ice loading and unloading can help us understand how the earth will respond to present day changes, and also help us figure out what will happen in the future. So, and this will require better estimates of the past isolates and their spatial distribution. Measuring vertical crust emotion is what today is an important piece of the puzzle, both for understanding the internal structure of the earth as well as for mapping areas of land subsidence that will exacerbate the impacts of sea liberalize dense GNSS networks have been helpful in developing these variations as well as satellite based interferometric SAR measurements. And so these two tools have provided a new global and regional estimates of vertical land motion that we can use for these studies. And critically understanding the physical processes driving these estimates of vertical crust emotion is important if we're going to be able to predict future vertical crust emotion that can be incorporated in projections of relative sea level change. So in summary, much progress is being made, both due to the availability of new data sets and new modeling and data analysis techniques. Critically on the understanding civil changes a multidisciplinary undertaking that involves oceanographers hydrologists, placeologists, climate modelers, geodesists, solid earth geophysicists and a variety of other fields. So advancing our understanding of how sea level change in the future will require all these disciplines to work together as many of the challenging problems lie in between the disciplines. All these changes take place on a deformable solid earth and so understanding how the solid response to sea level and ice mass changes is a critical piece of the puzzle. So I'll stop there and just say that on behalf of the committee, want to thank all of our speakers and also the audience for joining us for the meeting. As a reminder, the recordings will be available on our website within a few days also be sending out a short evaluation on the meeting next week. I know we all hate doing evaluations and polls but if you have a moment please consider responding. So thanks again and everybody enjoy your afternoon.