[00:00:00] Bridget Scanlon: Welcome to the Water Resources Podcast. I am Bridget Scanlon. In this podcast, we discuss water challenges with leading experts, including topics on extreme climate events, overexploitation, and potential solutions towards more sustainable management.
I would like to welcome Marty Ralph to the Water Resources podcast today. Marty is the Director for the Center for Western Weather and Water Extremes (CW3E) at Scripps Institute of Oceanography. His research focuses on precipitation forecasting with a big emphasis on atmospheric rivers and also how these atmospheric rivers or ARs translate to increased stream flow and tools to balance flood control with water supply.
So today we focus on atmospheric rivers and how we can optimize them to improve water resources management within the context of floods and droughts. Thank you so much, Marty, for joining me today.
[00:01:28] Marty Ralph: Happy to join you.
[00:00:00] Bridget Scanlon: So you have been involved in atmospheric research, atmospheric rivers research from the very beginning, but I guess the most recent family of ARs that we have experienced in California was December 2022 and through early 2023. That was a pretty exciting time for, for you guys. Maybe you can describe that a little bit, Marty.
[00:01:28] Marty Ralph: Sure thing. Yeah, in California in particular, I've had a lot of trouble with drought and flood and it had been a couple of years of drought and we saw on our predictive tools about 10 days ahead and indication in mid December that by late December around the holidays, there could, it could become active for ARs here and lo and behold, that happened.
And Christmas day, I think, or the day after we got hit by a really strong AR and then a few days later, New Year's got hit by another one, sadly, super serious and flooding and some fatalities. And then, we ended up by that point, we had started this period of nine atmospheric rivers hitting California in a three week period, including six of them as strong or greater in intensity.
Normally we'd get that many in an entire year and it came in three weeks.
[00:02:21] Bridget Scanlon: So that was very interesting, but it was fantastic that you were able to anticipate it with about a seven-day lead time that you could see these move across the Pacific. I mean, when I first thought about ARs, I really thought these extreme events, there's no way we can predict them at all. And so I think it's incredible that you can see them coming and anticipate them and then plan accordingly.
[00:02:44] Marty Ralph: Yeah, it's really a remarkable testament to science and prediction and all the people and organizations involved that that's even possible. So in that case, we had somewhere between 7 to 10 days ahead, we had a heads up that something might happen.
We don't usually count on it happening with that lead time because forecasts are often wrong at that far out. Atmospheric rivers, even before one has formed up the ingredients for it, are coming together in the models in a way that, that then form up the AR a few days out or a day before it hits or something like that.
And sometimes we see ARs out there already and near Hawaii or whatever, and it takes 'em a couple, three, four days to move into the West Coast. And they continue during that period. We're starting to see them even before the AR itself has formed.
[00:03:32] Bridget Scanlon: Wow, that's incredible. And that really helps with the management, water management, right?
But so you predicted the first ARs coming in maybe Christmas day or, and then the, can you predict the family of ARs? Can you predict that you're going to have so many?
[00:03:50] Marty Ralph: Well, it's like, imagine you have foresight that goes out, 7 to 10 days that you can see some fuzzy representation of something coming a few days out.
As it gets closer, you start to see it more clearly. And as the weather is evolving, we saw more and more storms coming. And there was one, one forecast around the third or fifth of January where we had five ARs lined up in the forecast product we use. I'd never seen anything quite like that. And I know it's happened before. I just haven't, we haven't had every, all the tools to look at it. And there we were with an onslaught of ARs coming, even though we'd already been through a few. So yeah, that's a really exciting moment. And there's a lot of people paying attention to this new information and it's quite interesting to see what's happening.
[00:04:35] Bridget Scanlon: And you mentioned that you had just experienced a couple of years of drought in California, and of course California is always either in drought or flood. But then as you get more and more of them, then it becomes, the flooding issue becomes problematic. So it's very difficult to manage these
[00:04:52] Marty Ralph: Right. Yeah, it's a land of extremes, and I know Texas is like that as well in its own way to have more insight or foresight in what Mother Nature is going to deliver in terms of extreme precipitation has a lot of implications and a lot of value. Of course, if the forecasts are wrong, that's a serious problem, and we have to be very careful about not expect perfection or, doing things not recognizing there's uncertainty. The idea that we can see, the hints of these, at least several days out has a lot of value.
[00:05:26] Bridget Scanlon: And so these ARs that came to California in December and early 2023, then. How were you able to translate those into increased water resources in that region?
I think particularly early on, you were working on the Russian river and Lake Mendocino. So what was the impact on storage and were you able to retain some of that storage for potentially upcoming droughts?
[00:05:52] Marty Ralph: Yeah, really good question. So our Forecast Informed Reservoir Operations, FIRO, pronounced FIRO, is a program we invented.
I came from NOAA to UC San Diego, Scripps Oceanography to start FIRO. We started on Lake Mendocino. And by the time this winter rolled around, we'd had a lot of experience. We'd had a couple of good outcomes already. And this year the reservoir operators were ready to look for whether or not there were ARs coming in the forecast.
And by the time we got into the the start of this period, the reservoir was about as low as it normally has ever been at that time of year. And there was a huge gap between the water level then and how much water would have to be in it in order to potentially create flood problems. It was so empty.
People weren't worried too much about that, but lo and behold, those nine ARs, they hit hard enough up there to put a lot of water in that reservoir and almost filled it up. And after the onslaught of ARs, the forecast tools I was mentioning a moment ago where we could see several ARs coming, it then started showing no AR coming.
And that allowed the reservoir operator after the onslaught of ARs had dumped its rain and the rivers had flowed and the reservoir was filling up almost all the way, they had their window of opportunity to release that extra water to restore the empty space for the next possible flood. And as they did that, they recognized there was no AR coming in the forecast for the next three to five days. And our FIRO effort had been able to officially designate about 10, 000 acre feet of flexible storage, where the reservoir operator was able to keep that extra 10, 000 acre feet, until the next day when they looked at the forecast again, and if there was no way are could keep it another day and another day and another day, so on and so forth.
So by the end of that, like, end of January, that was the situation, the reservoir had about 10 percent extra water in it compared to what it would have, or 20 percent really compared to the baseline. And that was a really good thing. But the winter wasn't over yet. And that's a whole other story I'll share in a sec if you want to hear it.
[00:08:05] Bridget Scanlon: Well, sure, go ahead.
[00:08:06] Marty Ralph: So, in February, the reservoir went into the month of February with the extra water safely there, knowing that if there was a forecast of an AR a few days out, the reservoir operator had time to release that extra water and restore the full capacity of the empty space, what's called the flood pool.
And lo and behold, by late February, guess what showed up in the forecast a few weeks later was ARs and some big ones and the reservoir operator then started to release that extra water to get it out of the way in case the reservoir was hit by a really big AR and as it turned out, it was hit and the reservoir rose in terms of its water level.
The reservoir operator then had time after that to release that extra water once again, down to that flexible level because they saw no more ARs coming in the forecast, and then they sort of rode that curve, so to speak, up into spring and ended up with, I think, 13 percent more water going into summer than they would have normally had, which is enough for tens of thousands of households for a year.
[00:09:15] Bridget Scanlon: Wow, that's fantastic. So, I mean, a lot of people talk about increasing water scarcity and that sort of thing, but my feeling is that in many regions, the biggest challenge is trying to manage these extremes, optimize water resources management to take advantage of these sorts of things, but it's not easy. FIRO is certainly a great example of how you're trying to push the envelope.
[00:09:39] Marty Ralph: FIRO holds a lot of promise. Yeah. Right.
[00:09:42] Bridget Scanlon: And so you're working with the US Army Corps of Engineers and Lake Mendocino is one of their reservoirs. I understand that they are strong proponents of FIRO and really looking forward to expanding this in other reservoirs in California, I think you mentioned, Prado dam and others on the Feather River, Lake Oroville, maybe you can describe the Prado Dam potential.
[00:10:07] Marty Ralph:. Yeah. So, Lake Mendocino was our pilot study and it became a five year effort to evaluate the viability of FIRO at that reservoir. And the results were very positive. And that led to a formal request of the Corps of Engineers to do what's called a major deviation, which would allow the reservoir operator from the Corps of Engineers during the high water storage periods to use the forecast information that we had developed and that they get from the National Weather Service, other parts of it and try out FIRO.
And, by the way, it's important to recognize the local water agency partner has been vital here, Sonoma Water Agency, which when the reservoir is below the level of the flood pool, they control how much water is released or kept each day. So it's a partnership between the U. S. Army Corps of Engineers and Sonoma Water Agency.
And it was their chief engineer, Jay Jaspers, who worked with me. We formed the steering committee and led the development of the program and in that period, another key individual, Cary Talbot, became the program manager of FIRO evaluations in the Army Corps. And we've all worked very closely together since then.
So about two years into the FIRO study on late Mendocino, it became really clear that the results were looking pretty positive, and that led to interest in seeing if it might apply elsewhere. And at the time, there were some folks who felt, well, we'd proven it on late Mendocino, so therefore it must work everywhere.
And of course, it doesn't. And after some modest discussion, it became clear to people who otherwise thought it would be that simple. They quickly recognized it's not that way. Essentially, every reservoir is a little bit like a person, and it's quite risky to assume you know every person because you know one.
So, each reservoir truly has a personality of its own, its purpose, its size, its climate, its structure. The stakeholders who need it, the environmental conditions, it's influences and as part of the politics around it, all of that stuff is very special to that one reservoir. So we knew that going in and yet we had to respond to this opportunity and pressure to explore it elsewhere.
So we did so. In a very measured way by going to a different location, and this time, instead of a very rural environment in Northern California, we went to a very urban environment in Southern California, neither one of which has a lot of snow impacting it. So that helped keep it a little bit simpler. Neither one (Mendocino or Prado) is connected by pipes to the major water supply system of California called the State Water Project or the Central Valley system. So they could sort of operate independent of all those things going on and we could poke around and see how it might work. It also had the benefit of like Sonoma Water Agency, it had a very progressive local water agency called Orange County Water District (OCWD). And they were already working with the Corps of Engineers. On Prado dam, which is right in the LA basin, It's right in the heart of the big 20 million person population center. And it's only a few miles from the river or from the ocean.
I mean, so it's travel time was going to be pretty short. We were optimistic FIRO could work there and it was already used. When there was enough runoff to fill it up a little bit, that water could be recharged into groundwater through Orange County Water District. But the dam as a whole, its purpose is flood control, and it's a big enough dam to provide flood control for 2 million people downstream.
But mostly it's kept very empty except for a little bit at the bottom after a storm that can be kept for a few days to then sort of drip it out a little bit to then. Orange County Water District can put it into groundwater. So that became a not too different type of system, but different enough to help us learn more about how FERO might or might not work on a different reservoir.
And it's turned out to be very successful there.
[00:14:19] Bridget Scanlon: Well, that's great to see the linkage between the surface reservoir and the managed aquifer recharge system that the Orange County Water District manages. And so, some people talk about surface reservoirs or subsurface storage or managed aquifer recharge, but here you're showing how you can link the two and optimize them.
And so that's a really nice marriage. So many of these reservoirs, they were built early on the 1950s, 1960s. And so their rules were determined back then and maybe haven't been changed that much since then. And so if they want to change the operation, then they have to get a deviation and you mentioned a major deviation for FIRO.
So incorporating forecasting skill then, I mean, the advances that you've made in forecasting into reservoir operations is a really big advantage then for water management and it's nice agencies take it up and go with it.
[00:15:15] Marty Ralph: That's right. So when we started FIRO in 2014, at that point, the Corps of Engineers rules were what they had been for decades with all of their dams is that the reservoir operator could manage based on what's called water on the ground, in a rain gauge, in a stream gauge, in the reservoir or in snowpack.
But a forecast of precipitation and runoff was not really factored in and to our good fortune, the Corps of Engineers rules changed and the engineering manual in 2015 to allow for the possibility of using forecasts. And that made it possible for us to pursue this even more. But that rule change didn't say how to do it.
It said that it was possible if, to use it, it was okay to use it if it worked out. So FIRO is essentially the way to answer the question of how to incorporate forecasting into reservoir operations in the Corps of Engineers. I need to make a sidebar here for a second and just mention that there are a lot of other reservoirs around the country that are not Corps of Engineers reservoirs.
And some of those have, like in the private sector, like private utilities and all, they have used forecasts and operating dams to their success in many cases, or in some cases they've had, there's not as many of those, but they exist and we're learning from them as well. But the Corps has 700 dams in the country, and they are a serious operator of reservoirs.
And this change in their engineering rules helped a lot open the door for FIRO.
[00:16:51] Bridget Scanlon: And do you think 2015, you were in the middle of a very intense drought in California. Did that have any impact on the Corps changing their rules then, or it just happened?
[00:17:01] Marty Ralph: I think there's two ways of looking at that. One is that there was already an effort underway that opened up the rules a bit to allow consideration of forecasts.
And at the same time that that discussion was going on, we had started exploring FIRO on Mendocino. So, they were parallel, whether someone talked to each other or whatever. I don't know, but it was an idea whose time was ready is I guess the main message there.
[00:17:27] Bridget Scanlon: Of course, you had another family of ARs in 2017 in January, 2017.
That was very much like maybe not quite as extensive as what you saw in 2023, but it was another big one rise that ended the drought really. And it's really interesting to look at the U. S. Drought Monitor data from one week to the next, to see its impact. It was amazing.
[00:17:51] Marty Ralph: Yeah, it truly is the case that in California in particular, and I said California a lot because that's where we've been learning, but a lot of this applies to some other places as well, especially on the West Coast and Texas, maybe that the extreme precipitation days account for a large portion of the variability from year to year, meaning if you get, here's the study in California, you take the 5 percent wettest days each year, and count those, add up how much precipitation is, then you do that the next year, the next year, the next year, and you look at how the 95 percent of other days or the other 95 percent of the days, how all that precipitation varies from year to year, it's almost flat. Those top 5 percent wettest days, which are usually 5 or 10 days a year, account for 85 percent of the variation in annual precipitation from year to year.
And in this world, this part of the country, those storms are almost entirely atmospheric river type storms. So it's not an exaggeration to say that the frequency and strength of ARs that hit California determines whether it's in drought or flood. That's what the science says, that's what Mother Nature delivers, and we have been working hard to better understand these storms and improve the observations and modeling and prediction of them with an eye towards usage by people in water management. Well, I mean,
[00:19:16] Bridget Scanlon: Well, I mean California was subjected to many multi-year droughts. I mean, way back 1987 to 1992 and, and 2012 through 2016 and stuff. And so you almost need this huge event then to break those droughts because you built up so huge deficit over those times, you need the flood to end the drought and Mike Dettinger's paper, drought busting ARs you do need to those huge events.
So you have developed sort of a formal procedure then to develop this program FIRO and viability assessments, preliminary and final viability assessments. Maybe you can describe that a little bit in the different agencies that are involved. I know you mentioned the Corps of Engineers and the Sonoma Water District, but also maybe Department of Water Resources and are there other groups involved in that?
[00:20:04] Marty Ralph: Yeah. So for each reservoir we're working on, we're working on like seven of them now, we have a team approach that involves, the Corps of Engineers, both the research side of the Corps and the operations side, plus the local water agency, who's also got some responsibility for that dam, as well as the National Weather Service in terms of the forecasting, as well as our center. I co-chair all of these to help ensure the science is there and solid and supportive of the operations. And then we also have some other agencies like the local water agency and California department of water resources. And also we've brought in folks who have a biology background for like fisheries related stuff, or even birds.
And we all get together in a forum and we create this ad hoc steering committee, two or three of us co-chair. And now at that point, we've developed the goal for that study specific to that reservoir, whether it's more water supply reliability in the face of drought or more flood risk mitigation capacity in the case of flood and environmental outcomes that we form a road map for how to evaluate that using our methods.
Very scientific methods about evaluating precipitation forecast skill, rainfall, or I'm sorry, runoff or streamflow forecast skill, environmental dimensions, and then work out, well, is there enough skill or not for FIRO to potentially be used? And our first cut at evaluating that is what we call the preliminary viability assessment.
And from that period, that's about a two year study. We then refine our efforts to carry out the full viability assessment, which usually means trying a couple of things we discover in the first preliminary assessment, and we realize what we need to do better at that point, and we pursue it fully. And then once that's done, we have established whether or not it's viable, and then that becomes the input into what's called a water control manual update.
Water control manuals are where the rules for the reservoir operator reside, and those are in some cases being updated already, and we're tacking on a FIRO study. Other cases, FIRO study is warranted, and a water control manual update will happen later, and other times they're sort of in sync. It's a very thoughtful process, very rigorous.
We do peer reviewed publications. We have our analyses reviewed formally, and it's all meant to be robust and it's helped us learn how to do this in multiple places.
[00:22:42] Bridget Scanlon: And so you are expanding this now up into Oregon, into the Willamette Basin, and also into Washington State. And of course, ARs play a huge role in the Western U.S., maybe a bit of a dipole, maybe when it's happening in California, maybe it's not happening so much in Washington State. But that's great that you can expand that to these other regions and develop these procedures and this template to follow will be very helpful for these other groups to get this on the ground.
[00:23:13] Marty Ralph: Yeah. So we're pursuing those as part of our program. I should explain that in Northern California, we now have a pair of reservoirs on the Yuba and Feather rivers. The two rivers come together downstream to a common point, which has a serious flood risk down there. So in that case, we're working with two reservoirs at the same time. And one is run by one agency. One is run by the Department of Water Resources. And that area, because it drains the Sierra Nevada mountains, snow is a really big part of the equation there. It wasn't part of the other two that we had done before. And this is results in added complexities as we're moving forward on this is two reservoirs at once and then snow brought in as well.
And also one of those reservoirs, both those reservoirs are much bigger than the other two we'd already studied, like 10 times their size. And then we also started one in Seattle, up in Washington, that is sort of similar in size to Lake Mendocino. And it's not that far from the ocean, so the travel time is short, which means forecast skill at three to five days is likely to turn out to be good enough for it to work there. But that one, one of the major goals is fisheries benefits and ensuring endangered species of salmon, for example, can thrive as much as possible and have the local water agency, Tacoma Water, be able to provide even more reliable water to its customers. So that's where we were at the end of the next phase of our FIRO process.
And we also in that start, we invented a way to do sort of like a weed out study across all of the Corps reservoirs, we call that the screening level process, and now we're in phase three of FIRO, where we're It's called the National Expansion Pathfinder.
And one of the first things we're doing is using that weed out method, essentially, called the screening level process, to down select from the 700 to a level where it's more reasonable, because the other ones, it doesn't even make sense to try FIRO, not because of forecast skill, but because how they're built, maybe they have no control structure that anybody can control how much water is released. And there are a couple of hundred of those summer locks on rivers and that sort of thing are between lakes. So we're going to be pursuing that over the next couple of years. So we'll really take a whole national look at the preliminary screening process is going to help us narrow down the field. In addition, we're going to be working on systems of dams where there are somewhere on the order of 10 dams, all that operate on one river system in some sort of coordinated way already.
And we're going to see how to evaluate FIRO for those all in one fell swoop. And these changes in our methods or expansions of them are allowing us to accelerate the evaluation process a lot. And our goal is to roll this across the country over time using the rigorous methods and tools we've developed to support the nation's need for additional flexibility in the face of climate change.
[00:26:19] Bridget Scanlon: And so are the reservoirs, the family of reservoirs in the Willamette Basin, the 11, I think, reservoirs there, is that an example of where you're saying their operations are coordinated and, and so you're going to see how you can apply FIRO in that context?
[00:26:34] Marty Ralph: Yes, and before we go there, I want to mention Seven Oaks Dam in Southern California on the Santa Ana River, which is upstream of Prado Dam.
So the same river, the Santa Ana River, where we were working, we've been working with Prado Dam, including this flood managed aquifer recharge. Upstream on the same river is a second dam, and it's substantial. And we're working with San Bernardino Valley Municipal Water District to evaluate FIRO with the Corps for Seven Oaks.
So that's two reservoirs in series. In sequence, yeah. And the other one, Yuba Feather, is two reservoirs in parallel. One isn't upstream of the other. Then Willamette in the Oregon area is also a place that's dominated by atmospheric river storms, and that is one where we are starting the process now to evaluate FIRO on 13 dams in that watershed.
None of them are on the main stem of the river. They're all on tributaries, essentially. And they're all run and operated by the Corps of Engineers. And there's an additional 14th dam that is run and operated by the U. S. Bureau of Reclamation. And our intent is to be in communication with them and explore coordinating somehow. Their dam is on a tributary farther downstream than any of the 13 Corps dams. So that's an exciting prospect for us. And I should mention that all of these ones on the West Coast have in common that atmospheric rivers are the driver of floods. It's just that simple. And it's not big thunderstorms so much. It's not, there aren't hurricanes out here other than the super rare thing that hit San Diego last summer, Hurricane Hillary, tropical storm and all. For the most part, it's all about ARs. And that's one of the reasons we're having success is because here is because the ARs are the storm type that matter and like we described earlier in the conversation here, there's some predictive skill for them that goes out a few days.
Now in other parts of the country where the storms that produce meaningful extreme precip for and I use precip as a short for precipitation, the meaningful extreme precip with relation to flood and reservoir operations sometimes comes from what we call convection, which are these thunderstorms that pop up, sometimes clusters of thunderstorms. Sometimes they can last a half a day and move a little bit or stall, and those really determine where the big floods are in those watersheds, predicting those skillfully four or five days ahead is much, much harder than getting the ARs right. Or right enough at least for reservoir operators to use. So that's going to be one of the future challenges is elsewhere in the country where that storm type is the dominant one, it's going to be a bit harder.
Another is that the rivers in the western coastal states, their origin typically in either the coastal mountains themselves or in another range of mountains just a couple hundred miles inland, the Cascades. Are the one mountain range in Oregon, Washington, primarily in the Sierra Nevada, in much of California. And then there's other mountains in Southern California, all of which are pretty darn tall, and it's not that far from the mountain to the ocean. So what that means is the distance the water has to flow in the river from the dam to the ocean is short, relatively speaking, and the river itself is sort of steep, so the water flows faster, and it means that shorter lead times are enough to be able to potentially have FIRO will be viable in that situation, you go to a place that's flatter and the rivers are longer.
That's going to make the travel time from a dam upstream to a flood prone community down river longer. And that travel time for the water coming out of the dam, if we've like kept it after a storm, expecting the forecast to be good enough that there'd be time to release it before the next storm, so that the higher water can get past the farthest downstream flood point, that might be two weeks.
That's going to be a long time, I think, before we have the kind of skill needed to make FIRO viable in a watershed like that. Maybe there's something I haven't thought of, or we as a team haven't thought of to change that, but I'd be It doesn't really look like a high probability thing at this point.
In fact, it's pretty low probability that it'll work. But there's a lot of range in between that. There's other places on the East Coast, there's shorter watersheds that don't have any other problems. I think we're going to find quite a lot of places where it works in the long run, and we're going to find quite a few where it doesn't.
[00:31:22] Bridget Scanlon: Right. So when you were saying there, the rapid transport in the West and the mountains, well, the mountains helped ARs form because of that orographic effect. And then the rapid transport. So you, five days is plenty of time for you for Lake Mendocino. So if you can predict five days out, you can release that water and you won't have a problem.
But going to other parts of the country may be a bit more problematic. So a colleague of mine, Ashraf Rateb, studied atmospheric rivers in the South Central U. S. And most of them come from the Atlantic and up through the Gulf of Mexico, and then some contribution from the Pacific at different times. But of course, here we have hurricanes and other things, and ARs are not the dominant driver. So most of them occur in the winter. And we met recently at the University of Texas at Arlington. So you were looking at the potential for the using FIRO in Texas. What are your thoughts on that or it's too early to tell?
[00:32:21] Marty Ralph: Well, it's a really important question and Texas has, there's sort of two parts of Texas.
There's the part east of the I35, which is pretty wet and the part west, which is often dry or it can be drier. And the answer might not be the same for both. When it comes to FIRO, that's one impression I've got. Second is that some rivers are pretty long there and pretty, and not very steep, and the lead times are going to be a couple weeks, two weeks maybe.
That's going to be pretty hard to have FIRO viability assessment be successful. I do think it's valuable to try them. On a few situations in Texas, which vary, there are going to be some that are shorter distance travel distances and times because maybe they're closer to the Gulf of Mexico or there's a big enough reservoir downstream that it can handle whatever the smaller tributary is putting out.
So I think it warrants evaluation just being eyes wide open that it might not work out nearly as well as it looks to be working on the West Coast, but that it doesn't mean there won't be some circumstances where it could be helpful. It's just got to be done very carefully, rigorously, with key parties involved, like I described in our viability assessments.
And we're going to be doing our screening process over the Texas area pretty quick. And that's going to give us a first look at how things might pan out for Texas.
[00:33:50] Bridget Scanlon: And so you've also been looking to kind of shift a little bit globally at ARs and originally most of the work in California, but now you're seeing them in many regions, Australia, New Zealand, Europe, west coast of Africa and other regions.
Can you describe a little bit about the global distribution and how that is impacting, has potential for water resources possibly, but maybe they're not that far along.,
[00:34:16] Marty Ralph: Yeah, sure thing. I think what I'd like to share is we're in the conversation here, shifting a bit away from the water management side, because that's all very guesswork at this point elsewhere.
But the science of ARs is quite interesting alone and looking across the globe. And what we found, first of all, is in the mid latitudes, like where we live in the U. S. mostly, that atmospheric rivers are the primary way that Mother Nature moves water vapor around horizontally. And particularly in the cool season, something like three or four ARs are present in the northern hemisphere at any one time, and they are doing about 95 percent of the horizontal water vapor transport for the entire mid latitude area.
So they're literally rivers in the sky, but they're rivers of vapor blown by the wind. And what that means is that they end up being a few hundred miles wide and a couple thousand miles long. Bye. The size of the planet standpoint, that's pretty small. Those three atmospheric rivers, if you cut across them all, they would add up to less than 10 percent of the circumference of the Earth.
And yet 95 percent of the water vapor transport is happening in those narrow regions. If you slice across one, an average AR is transporting horizontally as water vapor as much as 25 times what the Mississippi River discharges into the Gulf of Mexico. They're massive. That's also corresponds to 2. 5 times what the Amazon discharges into the Atlantic.
The Amazon is the biggest terrestrial river on earth. So an average AR is bigger and it's freshwater transport than any other river on earth. And they are the engine of the global water cycle. So when we look at the globe, then it shouldn't be surprising that where they occur elsewhere in the mid latitudes with some frequency, like the west coast of Europe or the west coast of South America or parts of Australia or New Zealand, even South Africa, even the Middle East sees them a bit, they are a major player in the floods and in the water supply, like they are on the west coast of the U.S.. And even in other parts of the country, like the Great Plains and the Eastern U. S., ARs are one of the key storm types. They're not the only one. Hurricanes and big thunderstorm complexes, like we talked about, are others. And that's true in other parts of the world as well. A lot of areas around Earth are influenced substantially by ARs, both in terms of their extreme precipitation and their water supply.
[00:36:48] Bridget Scanlon: And so in Australia, I was reading they had some major droughts and then the Indian floods and these floods are oftentimes caused by ARs. I think in March 21 and October 22 periods that they had major ARs in Southeast Australia.
And so do you think they're learning how to predict them and, and see them coming like you are in California or watching them?
[00:37:13] Marty Ralph: Well, we are working with folks in Australia to explore the applicability of what we've developed here to their challenges. And indeed, I think it was March of 2022 in central part of coastal Australia, they had a terrible flood and it became the biggest natural disaster in Australian history.
And it was in fact an AR, it was an AR 4 and a family of ARs back to back, and it really wreaked havoc there. So there's a lot more awareness now in that region to the importance of atmospheric river type storms to their stream flooding situation and their water supply. So yeah, there's a lot of potential there. Portugal is another really hotbed for ARs. There's some really good scientists there who've been working on it. Western Europe, inclusive of England and Norway and France, even the Mediterranean coast. That's very important there as well.
[00:38:13] Bridget Scanlon: Right, I recall in, in Australia at the end of the millennium drought, which went from about 2000 to 2010, they had major floods and that there were questions about how they operated the reservoirs, if they operated them appropriately.
And I think after going through such a long drought, you'll be very reluctant to release any water. And so, but if you understood ARs, then you would have a better feeling for what to do, I guess.
[00:38:40] Marty Ralph: I can't really second guess what happened back then, but certainly people can take better advantage of facilities to manage for extremes when we know better when an extreme precipitation event's coming.
That's what our center is about, our Center for Western Weather and Water Extremes is about extreme precipitation in particular, and ARs are a major component of that.
[00:39:04] Bridget Scanlon: And you developed a ranking system then for ARs, ranking one to five. Maybe you can describe that a little bit and its applicability globally.
[00:39:13] Marty Ralph: Sure thing. So about, I started working on ARs about 20 years ago after I read a paper by scientists at MIT, Zhu and Newell in 1998 that was one of about five papers back in the early nineties that coined the term. And it really clicked for me that that's what I was studying. And I started working on it more and studying it more and published a few papers.
And since then, the topic's really taken off. So about 10 years into that period by 2015 or so, the public had become aware of it on the West coast as a storm type that they paid attention to, that people sort of had a sense that they were due, they were. The cause of big floods, the understanding was not nuanced enough that, oh, there's weaker ARs that are mostly beneficial.
So, you don't have to panic or something, but, be prepared nonetheless, and maybe it's going to be beneficial for water supply. So we got word from National Weather Service in particular that they were getting inquiries from their customers, the public, that they'd like to know whether the next AR in the forecast was a big one or not.
And after thinking about that for quite a while, I've been thinking about it already. I formed a team to try to do something on that. And the fact is for a couple of years prior to forming that team, I had had a list of different attributes of a storm and the situation it would hit, like, is there snow on the ground or whatever.
On my whiteboard, there's literally 10 different parameters. I had yet, I knew I wanted to create a scale, but I couldn't. Really figure out how to simplify it. And then I went to a conference called AGU in San Francisco. It was on a Monday. I'm at a restaurant that had the TV weather on, and I saw the forecast, which included a sunny symbol on Monday and a little cloud on Tuesday, then a little light sort of friendly cloud and light rain on Wednesday, and then a scary looking cloud, big raindrops on Thursday. And I knew much better than that, that that was a potentially serious storm on Thursday. And there's no way a dark, threatening cloud and a few raindrops could ever convey the amount of impact that could have. So it was literally at that breakfast I decided I got to get off my duff and turn that 10 item list down to something simpler. And we had two parameters. I just published a paper a couple of years earlier that showed the strength of the AR. And how long it lasted over a spot were the two things that controlled the impacts the most. So, or at least in terms of the storm strength.
So, I formed a team and we wrote a paper, we did the analysis, wrote a paper, published in 2019. And we now have a scale for ARs from one to five. And then we did an economic study with flood damages and it showed like a factor of ten increase for each ranking in terms of the median damage is done. So our original scale was good at sort of representing where the biggest risks are, but we can still get an AR5, the highest one, that doesn't do much damage.
If it hits an area that doesn't have a lot to damage, or if it hits a landscape that's been dried out through the summer and the soils can absorb a ton of rain, or we can get, say, an AR5 after that, and maybe you get an AR2 after that, that AR2 can put the rivers over the edge of flooding, so the AR2s can also produce damage, but when you average all the cases, The median numbers ramp up by nearly a factor of 10 for each one with ARs 1s and 2s, not really doing much damage.
They're mostly beneficial, 4s and 5s batten down the hatches unless the soil is real dry and it's not an area that gets as much damage. That's the story of the AR scale, it's now used in the media a bit, our studies, there's a lot of, part of the reason to create it was for scientific studies, to normalize how people are looking at these storms, including for climate change, and it's been a pretty effective tool so far.
[00:43:01] Bridget Scanlon: And as you say, antecedent conditions are very important, and then if you have a family of ARs, that can be really critical. So you did look at some 400 years of precipitation with tree ring data and looking at these heavy precipitations in California. I thought that was an extremely interesting study and showing the importance of ARs over the long term.
Maybe you can describe that a little bit, Marty.
[00:43:24] Marty Ralph: Sure thing. A very exciting study, we've been working on it for a couple of years with a key colleague, Dave Stahle from Arkansas, who's a longstanding dendrochronologist is the word for a tree ring expert. We knew that the tree rings in California represented a good measure of annual precipitation.
In some cases, like 80 percent or 90 percent of the variation in annual precipitation could be, that's the correlation between the tree ring width on these oak trees or even some pines and annual rainfall. But we revisited that a bit where we asked the question about. The number of ARs and is the tree ring with related to the number of ARs.
Well, remember the little study about 85 percent of the variability and annual precipitation was due to the top 5 percent wettest days, which are almost all ARs. Well, that came true with the tree rings. Essentially, the tree ring width for this oak species in central California correlates like to 80 percent plus to annual precipitation and to AR count.
So the tree rings become essentially an AR measurer. Right. The total accumulation of AR activity for this, for the cool season kind of measure. It's pretty cool.
[00:44:39] Bridget Scanlon: Yeah, and then more, we've also been looking at the potential of climate change impacts on ARs, and maybe you can comment briefly on that, what you're seeing for the future, or what do you think?
[00:44:53] Marty Ralph: Yeah, sure thing. Yeah, we just have a minute here to go. I'll share that one of the key things to recognize is that atmospheric rivers are fueled by water vapor. That's what determines the amount of water vapor flux, is the wind times the water vapor. And a warmer atmosphere can hold more water vapor. So essentially, global warming is, CO2 and all that, is a way to make the atmosphere warmer, or is leading to a warmer atmosphere, which allows the atmosphere to hold more water vapor, which is adding fuel to the ARs, which is going to make some ARs bigger and stronger. It's that straightforward. There are some caveats on it, but for the most part, those are smaller. The other thing, though, is that the big weather patterns for the whole northern hemisphere, they can, in a warmer climate, get stuck longer in one position. And what that means is the dry periods can be longer and the wet periods can be wetter.
So for California, where we've done a bunch of studies primarily and there's other studies elsewhere, it's very telling that the annual precipitation total might stay the same more or less for California on average, but more of it is going to come in a couple of very wet days and the days between storms, it'll be longer dry periods as well.
So both heavier storms and longer dry periods. Not what you want to hear if you're responsible for managing for drought and flood, but that's partly coming full circle here on our call is where FIRO comes in, because that's the idea, partly that FIRO could address climate extremes by using existing water management infrastructure to more flexibly operate in the face of super wet, Or super dry.
[00:46:43] Bridget Scanlon: Well I think that's a good note to end and I really appreciate your time. Our guest today was Marty Ralph, who is the director for the Center for Western Weather and Water Extremes, CW3E, at Scripps Institute of Oceanography. Thank you so much for educating us on ARs and for the incredible research you've been doing for the past several decades.
I really appreciate it.
[00:47:06] Marty Ralph: Thank you. Very nice to be here.
