The search is on for a new Delta lead scientist to oversee the Delta Science Program and to consult regularly with the Delta Stewardship Council and other partner agencies. The position is secured through a multi-year appointment with the USGS. The Delta lead scientist will be appointed by the Delta Stewardship Council following a recommendation from the Delta Independent Science Board (DISB).
The DISB will be interviewing four candidates for the position over the next month; each candidate will present a brown bag seminar of their research and experience and how it applies to the lead scientist position. As part of their presentation, they will give their vision for the Delta science program.
The first candidate is Dr. Jay Lund, whose research and teaching interests focus on applying systems analysis and economic methods to infrastructure and environmental problems, including policy, planning, and management studies. His work primarily in water resources and environmental systems engineering, but he has very broad experience which includes work on solid and hazardous waste management, dredging and coastal zone management, urban, regional, and transportation planning, and urban ethnography.
Dr. Lund received a PhD in civil engineering and a master’s degree in geography from the University of Washington in Seattle. He has authored over 150 refereed, peer reviewed journal articles in a variety of fields, many of them include papers and reports and books on California water and the Sacramento-San Joaquin Delta. He is also an active member of the Delta Independent Science Board; he has served as the chair of the Board as well as Chair elect and past chair. He also has a very distinguished record with many awards, including the California Water and Environmental Modeling Forum’s Hugo Fisher award, and in 2018, he was elected to the US National Academy of Engineering.
Dr. Lund began by saying that his presentation would be about science through the perspective of integrated modeling because he is a modeler. He grew up sailing inland and offshore on the East Coast. He received a bachelors degree in International Relations and Regional Planning from the University of Deleware. He then went to the University of Washington to become an urban social geographer as his goal at the time was to become a professor of geography.
“That was my long-term goal and I failed at that goal,” he said. “But I had the most fun master’s thesis of anyone ever on the planet. I studied people that lived on sailboats and power boats in Seattle. It was an urban ethnography meaning I had to interview all those people, and almost every one of them offered me a beer.”
He then became an engineer, getting a second bachelors in engineering and a PhD in engineering from the University of Washington. Two weeks later, he came to UC Davis to become a professor for civil and environmental engineering.
“My field really is integrated engineering with engineering, economics, and operations research, mostly around water management, but a few other things,” he said. “I’ve worked on California’s system, and I’ve also done modeling work on the Missouri River, the Columbia River system, Panama Canal, South Florida, and a few other places, most recently, New York City’s water system.”
For this talk, Dr. Lund will talk about integrated modeling and give some insights and solutions for real water problems. “I view … the model as a central organ that you use to integrate knowledge. It’s not a bunch of conversations, it’s not a bunch of workshops; you actually have some stupid machine that you have to teach what you think is really going on, and then you test that hypothesis over time.”
First of all, he gave some thoughts on integrated modeling. “The problems don’t care about our disciplines,” he said. “We’re educated in our disciplines, we have to write dissertations on our disciplines, we have to theses on our disciplines, but once you get out to a real problem, it doesn’t care. Your discipline will be useful, but it is not sufficient. What can be important is if you combine your discipline with other disciplines; have a disciplined way of combining disciplines. This is not just getting everybody in a room and talking; there’s some structure to that conversation, some motivation of that conversation to have some integration and a useful outcome.”
Systems modeling is a discipline to integrate and test knowledge using computers for problem solving, he said. “We’re not really trying to understand the science of the problem usually as an engineer; we’re trying to figure out what do we know about this problem in terms of what solutions look promising or unpromising.”
Dr. Lund reminded of the quote about all models being wrong, but he said that modeling and the model process is used to improve difficult conversations. “We’re faced with these water problems that have 100 years to 150 years of difficult conversations back to the beginning of time,” he said. “What we’re trying to do is have the modeling to provide some disciplined idea of what we really know to help improve those difficult conversations to help us focus our learning and the modeling therefore should reflect both the science and the problems.”
“And by the way, the model is never finished,” he added. “There’s a great saying that, ‘models are never completed; they are only abandoned’.”
THE CALVIN MODEL
As an example of integrated modeling, Dr. Lund discussed the CALVIN model developed at UC Davis by many people over 20 years. The model was first developed in 2001 for a 2020 horizon, so he thought it’d be interesting to present just how well the model worked.
He presented a schematic of California’s water system. “It’s big, it’s everywhere, it’s got everything: Lots of reservoirs, groundwater, lots of people and lots of government. So we made a model of it and it looks like this. You try to get as much in there as you can.”
The CALVIN model has the surface water reservoirs, aquifers, vast conveyance network, 8 MAF of irrigated agriculture, 3000 governments, and more than 40 million people.
Dr. Lund noted the two pictures of Calvin on the slide which reflects the personality of this model: the mischievous 5 year old and the religious reformer. “We started building the model in the year 2000, and we decided to model the entire intertied system at a time when only the surface water or only the groundwater or parts of them were being modeled,” he said. “We modeled supplies and demands, and to keep us sane in our minds, because we’re just a bunch of professors and students doing this, we used an optimization model so we didn’t have to specify operating rules for every little piece of infrastructure because that would just drive you nuts.”
The model optimizes for economic value, considering the economic value of water deliveries to agriculture, cities, and hydropower and things like that; however, the environment is important, so it was represented as constraints for the environment that are required to deliver the water. The optimization model provides mostly water system operations over 82 years of that whole network.
“The real value of this is that it forces you to develop an integrated quantitative understanding of the system; the numbers are of secondary importance to what you learn putting the pieces together and trying to convince this really stupid machine that all those pieces fit together,” said Dr. Lund. ““The first time we ran it, it said ‘no feasible solution’ which actually I think is the truth. We should have just stopped there, but we did have to systematically debug it, take out all of the infeasibilities and now it gives us feasible solutions that nobody likes.”
Dr. Lund presented a schematic of the Tulare Basin, noting that it has a lot of little pieces there, and acknowledging that archiving and documenting the data was really important to the longevity of this work. For the economics, they had a whole separate set of models for that; the used the SWAP model which provides monthly economic value functions for water delivery to each of these regions over time.
“In managing California’s water system, there’s not one thing we can do – there are hundreds, thousands of different management knobs that we can twiddle and turn and levers to pull,” he said. “We have lots of local activities we can do. There are lots of regional and statewide activities we could do. We’re trying to represent as many of those in the model that we can.”
The model utilizes the portfolio concept, much like Newsom’s executive order directing state agencies to put together the water resilience portfolio.
“We’re trying to have an analytical way of coming up with what looks like a promising portfolio. Is it the best portfolio? Probably not. Is it a reasonable portfolio? Maybe getting close to there, but it’s a good point of departure for further discussions.”
Without an optimization model, it’s as if we’re in a thousand-dimensional dark room with a pen light looking at one solution at a time, he said. “An optimization model is flash bulb. But as you look at the whole thousand-dimensional dark space, just a flash tells you maybe the better solutions look like there over there or over there. And then you go to those places with your pen light or your discussions. An optimization model sounds scary but it’s not. It’s just a flash bulb.”
For the water management objectives, the first priority are the environmental flows and then the economics of water use. “We had to orchestrate all the data, hydrology, capacities, environmental flow constraints, values of urban water, agricultural values of water, operating costs, hydropower – all that has to be put into databases as well as the metadata or documentation of that data. We have a solver … and then when the model results come out, we see economic costs and benefits because we have the economics built into the model.”
The model utilizes conjunctive use and cooperative operation of the whole system because the model has been told to cooperate. “It doesn’t have to get changes to legislation,” he said. “It doesn’t mean the agencies are actually going to do that, but it does say how worthwhile would it be if they cooperated.”
An interesting by-product of the optimization model is that it gives the incremental willingness to pay for additional water at any location in that network at any month over those 82 years, so it shows the potential for water markets across this network between wet years and dry years. It also has the delivery reliability optimization operations, such as the frequency distribution of deliveries under economically-ideal operations; they can also constrain the operations to something less than the ideal if there are policies out there that constrain things.
The value of flexible operations can be quantified by adding or subtracting different policy constraints to an optimization. The model can also find and estimate the increase in economic value to the whole system if there is more or less capacity on any piece of conveyance or any bit of storage or any treatment plant in the system.
“The real problem with this is you get a big pile of numbers and then you have to sort it out,” he said. “It’s really simple math. There’s just a lot of it, and that’s why you have computers.”
The model gives engineering and hydrologic results, time series, delivery reliabilities and operating costs and it will give you economic results which is where there are some synergies from integrated modeling that wouldn’t happen if the hydrology-engineering modeling was separate from the economics.
“You get to see these things together, which is kind of what people care about,” Dr. Lund said. “They really don’t care about hydrology and water deliveries for their own sake; most users in the system care about the economics or for the environmental case, and we’ve got that as well.”
It took about 5 years to build the first generation of the CALVIN model, and they’ve used it a number of times for a lot of different problems throughout the state, as well as many PPIC reports which have been based in part on these model results. The CALVIN model has been built by a lot of students over the last 20 years.
CALVIN’S 2004 CLIMATE STUDY
Dr. Lund then presented a slide showing the results of a climate warming study they did in 2004, noting that if you were to plot climate change scenarios today, you’d see a similar scatter. The dark line shows the historic hydrology including the spring snowmelt bump, and there are drier and wetter scenarios, all of which have the climate shift with different temperature changes.
Back in 2004, the model foresaw the importance of groundwater storage and conjunctive use, and agriculture to urban transfers, such as what did occur between Metropolitan, San Diego, and the Imperial Valley. The model predicted water quality exchanges and a lot of flexibility in trading, due to the tremendously variable hydrology and crop prices. It also predicted some agricultural land fallowing, particularly in the drier scenarios, because when there’s not enough water to go around, the lowest value agriculture usually gets taken out of production. Also new technologies, wastewater reuse, and urban water conservation and water use efficiencies.
“Basically there’s a mix of responses, not a silver bullet, but an integrated mix of responses that you can derive analytically, at least it puts you in the right ballpark,” said Dr. Lund.
Dr. Lund than ran through a series of results, starting with scarcity costs by sector, noting how agriculture really gets hit hard during the dry scenarios. The second slide shows groundwater storage, noting that the drawdown-refill cycle of groundwater basins can be much longer than surface water reservoirs, due to the increased amount of storage in the aquifers.
For new water source technologies, the model predicted 1.6 MAF in the dry scenario of additional waterwater reuse and a little bit of desal in some cases.
He presented a slide which showed the ‘shadow cost of an environmental flow’. “If you increase an environmental flow by one acre-foot per month, where does that water come from? Sometimes it’s just there, it was a wet year. It has no cost to increase that environmental requirement. In dry years, it might come at the costs of some fields that are now been partially fallowed, or some additional groundwater pumping that has to be done someplace else in the system. And you can see for each year of the study on the record, how that has fluctuated and what you see is in the dry scenario, we had to go to a completely different scale because it was now affecting very high value crops and urban water uses. But in the other years it was high but not horrible.”
Lastly, he presented a set of results illustrating what is the economic value of facility capacity changes both for reservoir storage and different canal conveyance for different climates, noting that climate is important, where the infrastructure is is important and what it does (conveyance or storage) is important.
Dr. Lund than gave his overall conclusions from CALVIN effort. “The quantitative analysis identifies promising portfolios of actions, identifies important things that we don’t know or worry about, helps us explore options and impacts, and hopefully it makes the discussions more productive, because it’s something to ground the discussions on.”
“Optimization models helps you explore many different options, it helps you identify promising combinations almost automatically, and the bottom line, I think the most important thing, integrated modeling helps us learn.”
DOES CALVIN WORK?
In 2001, the CALVIN model was used to predict the water system in 2020. “Here it is, 2020, and we did the first modeling for year 2020 demand – and it worked,” said Dr. Lund. “All the big water transfers that were suggested by the economically-driven optimization model in 2001 have now been completed. The model worked. As an academic, I find that amazing. It was just a bunch of us that had never done this before, and we had just started.”
One of the results that came out of the 2001 study is that EBMUD and Contra Costa Water District should build an intertie, because in wet years, EBMUD has a lot of really good quality water and Contra Costa Water District is still pumping from the Delta; and Contra Costa could save a lot of treatment cost if they just took EBMUD water. In dry years, EBMUD struggles with water supply and Contra Costa Water District has Delta pumps that can be used for trades; EBMUD might not get the water quality they’d like, but they would get water.
“We presented those findings in front of the head of Contra Costa Water District, a fellow named Wally Bishop,” said Dr. Lund. “He was kind of a feisty guy. He didn’t like model results and gave me all kinds of grief about it. The next day, people I knew in both East Bay MUD and Contra Costa emailed me, asking for the data. I believe there’s an intertie there now. And I was told a few months ago, at the time that I gave that talk, they were starting early discussions on negotiating that intertie. The model was smarter than I thought.”
Conjunctive use is active and integrated across the valley. There’s not a huge value of massive amounts of expanded surface storage, and finally and important, the overall flexibility of the system provides a lot of strength, he said.
“The ability to change your operations between wet and dry years, different sources, different demands, and water transfers gives tremendous strength, and we certainly saw that in the recent drought,” he said. “Integrated modeling for California works.”
Dr. Lund then turned to the Delta.
Issue: The Delta has a lot of diverse problems that overlap with agency missions.
“Certainly if a problem has one agency with that mission, it’s pretty clear – the agency can just do it and you know who to hold responsible,” he said. “We have a lot of problems in the Delta that cross over different agencies’ missions. Regulatory side, project side, different regulatory interests – this is a real challenge for science.”
At the university, if a scientist has a problem, there are people around with different expertise that can be brought together to figure it out.
“Agencies have and they have to have more territorial missions, it’s like deans at universities. It’s not my department, I don’t care. The real problems are networks. There’s no perfect solution to resolving this problem, but it’s something we have to struggle with or struggle against as much as we can.”
Issue: Brutal incrementalism
“In the real world, I’m not a believer in revolution. There’s a wonderful quote that says, ‘revolution is the opiate of the intellectual,’ and I firmly believe that. Incrementalism is unavoidable and it can do pretty well if we’re organized and a little bit more forceful about it. But my impression of the system is that incrementalism has become brutally slow and too small in increments. We’re not really able to move things along as fast as they need to happen, given the changes in the system.”
“The Delta is changing faster than science and management can keep up, but if we’re brutally incremental, we’re not going to be able to keep up,” he said. “We’re learning in many tiny bits alone and slower than things change. Anybody that proposes things has to go through many agencies; it’s like the university getting something through the academics. And then we have sort of a Zeno’s paradox of incremental slowing where we’ll get through that increment, let’s make the next increment half as big, and the next half as big as that, and we’ll never arrive.”
One Delta, One science
Dr. Peter Goodwin, the Delta lead scientist from 2012 to 2016 set a great direction with the ‘One Delta, One Science’ concept for integrated science, planning, and activities and the development of the Delta Science Plan.
“These actions plans bring people together,” he said. “We’ve done a good job of establishing that those exist now, but we have to get them out of being brutally incremental. We need to bring more people on board and take bigger steps and that’s a big job; it’s hard to do.”
“Part of that I think will require integrated modeling, partly because that’s what I do, and of course, being a good academic, I think whatever my expertise is, the world needs, but I think in this case, it’s actually true,” he continued. “Not for everything; there’s a lot of things aren’t modeling science. We need a lot more interagency collaboration and staff development. All the agencies are going through the same problems. They’re losing senior staff, new people coming in with these responsibilities but they haven’t had time to really come together and come up in it. If you have interagency collaborations, maybe you can bring all these new staff in together, educate them up in the same common problem and the same common approaches and they’ll all get along better because of it. We have to continue to look at science as the middle ground where discussions can occur and we can move forward.”
“We need to continue to pay more attention to the science-policy interface, something that Peter was tremendous at, he’s both effective and nice,” he said. “It’s important to be as nice as you can, but it’s ultimately important to be effective. Peter is a great example of how you can do both. Bottom line, we must move more firmly in these directions.”
Science for common problems
The Delta has many problems that spans several agencies. Dr. Lund suggests building multi-party task forces, bringing together multiple agencies and non-agency participants to add some independents. The task forces would focus on things such as hydrodynamics, water quality, systems operations, data management expectations, synthesis review, and solutions development. Then try to bring these task force results into the policy discussions.
Dr. Lund then gave some examples of where this has been implemented successfully.
The first example is New York City’s water system. The green areas on the map shown on the lower left are the watersheds that New York City uses for its water supply. The Delaware River is shown to the west. All these are components of their model because in running the water system for New York City, they have to be concerned with salinity intrusion in the Delaware River, hundreds of miles away.
So they’ve developed the Operations Support Tool which has about six different system operations models. “Different agencies have different models; different programs in different agencies have different models,” he said. “They have one model that’s used for operations planning on an hourly timescale to system planning and policy planning on a decadal scale. They all run it differently, of course, but at least once a year, they bring all the modifications that have come out of the different shops using them for different purposes, and they’ll upgrade the common model. They get more cross-fertilization and more commonality in understanding how the system works. It took them about four years to develop the initial version of this model and about $8.5 million, but they revise it annually. One model for everything. Water quality, data systems, all sorts of good things.”
The second example is the Chesapeake Bay program which is led by the US EPA and their modeling program has the goal of stakeholder understanding of the system by developing an understandable model through an inclusive process with better and more local input data and more monitoring data.
“I asked them, what keeps this from just fragmenting into a model that you never show the public because you’ve never gotten every stakeholder buy in on every little detail? And they said, ‘we have a timeline. The model will be released and used on that politically driven timeline, and that enforces some discipline and forces some simplifications in the development of that model. You can always improve that model later for the next timeline, but you will complete and have model results on that timeline.’ And that I think is the key to it. EPA money keeps everybody in line.”
The model they’ve developed has fall runoff, atmospheric deposition, transport of all sorts of things, nutrients and the like, land uses, BMPs, 3D hydrodynamic models, water quality models of the system, all different ways to look at things.
They developed the model through an inclusive process that has a lot of different workgroups involving lots of different agencies and independent review folks on every component of the model, so long as its done on that timeline, he said. They have a Scientific and Technical Advisory Committee with 41 academic and agency members that has been carefully developed so that it has enough academic and independent members so that it’s not just what the agencies can agree on.
VISION FOR THE DELTA SCIENCE PROGRAM
Dr. Lund then gave his thoughts about the Delta Science Program. “We need to try to organize science and policy discussions a bit more,” he said. “As academics particularly, our academic background hinders us often in terms of policy making as we tend to concentrate on things we don’t know. The things we don’t know are certainly important. But, for real decision making, you also have to have some faith in things you do know. Often we have sufficient knowledge to make a reasonably good decision, and then we can look at the uncertainties as what other contingencies should we prepare for, so you’re not neglecting uncertainties and you’re not letting them hold you back, but you’re not using them as an excuse not to make a decision.”
“An important aspect of the future work around here for this system and other systems is to talk about the first order uncertainties, the things that help us develop a common understanding in the system,” Dr. Lund said. “And then to communicate this across agencies, across interests. So we have some common ability to talk about things.”
Dr. Lund recalled how when he was in the Netherlands trying to learn how they do flood risk analysis, he noticed everybody was quite familiar with flood risk analysis. “If you have a discussion here about floods, everybody is on a different page. Everybody talks about something completely different. If you go the Netherlands, they’ve had this problem for 150 years and everybody talks about flood risk analysis. Enviros talk about flood risk. Agency people, engineers – everybody has the same framework. They will disagree about things and that’s fine, but their disagreements occur within a framework that everybody understands. It helps the conversation around them and makes things a little more cost effective.”
We need to take bigger incremental steps, he said. “Dr. Peter Goodwin’s lead set a common direction with persistent incremental progress in as big as steps as you think you can get away with, but we need to make the progress in bigger increments than we’ve been doing lately. Staff development is really important in all of this to develop a coherent core. We have to start from the core of things we know and build out. If you start from all the things we don’t know and try to build in, I think there’s just too much entropy going on.”
“We have to have a little broader science appreciation from the elected officials, the agencies, the higher level agency folks, the legislative staff and the like. They aren’t going to fund it if we can’t explain it to them and if they don’t have an appreciation for it. So I think we need to have pretty regular and maybe a little more organized discussions around the core of knowledge and the core of problems in our science and policy discussions.”
Dr. Lund then gave his conclusions. “We know a lot about this system. There’s a lot we don’t know, too. And we need to know more and make it known. All of our science work needs to be focused on the problems and solutions, some of which will involve fundamental research. Cross agency, cross party organization is really important for helping us all know more together and helping us be more effective as policymakers and managers. It’s not just us communicating with the policymakers, it’s also being better organized to digest science. The science appreciation becomes important. It takes two sides to have a conversation. We’re not particularly good at purging ourselves with the jargon and the like, but sometimes the processes that we’re trying to help might be tweaked [if we adjust the words we use when we talk].”
QUESTIONS AND ANSWERS
Question: I appreciated your comments about science appreciation and the importance of connecting with policymakers and the legislature. This is an area where I know you know a lot of the policymakers because of your career and work there. What are your thoughts on how to increase science appreciation on the part of policymakers might be achieved?
Dr. Lund: “I was told by a wise professor many years ago that in this planning business, you have to accept that most of the time you will fail. I think this is also true in science policy communications. Most of the time you’re going to fail. A lot of times, you’re just trying to establish lines of communications for when a flood, drought, or lawsuit comes up and they’re really interested in paying attention to you. You have to be persistent, that’s the main thing. I think building it around the core of things we do know is a lot better conversation than talking about uncertainties, because decision makers don’t want to hear what you don’t know, they want to hear what you know. We’re scientists; we’re supposed to know stuff. Sometimes we know that we don’t know perfectly but we do know somethings that are useful for them.”
Question: I know you’ve been a player in these discussions for many years as you described. The frustrations that you described are bountiful, what makes you drawn to that now in your career? Why do you want to persist … ?
Dr. Lund: “It’s kind of the ultimate challenge for an academic. I have had an unexpectedly productive career. I’ve kind of been successful in doing this from the outside. This job has opened up in the past, I’ve usually thought, well, I think at this point I can be more effective at informing policy from the outside than from the inside … I think I can be useful. I like to be useful. That’s the most interesting thing to me. The people of California have paid me for 33 years now to do good things, and this is just another continuation of that.”
Question: Sometimes the approach we have is if we just throw data at people, if we just show them how good the models are and how good our understanding is, they’ll understand, and sometimes it seems like the reverse is true and that makes people fight back less. Do you have any suggestions or thoughts about how we as scientists might be able to communicate that information better?
Dr. Lund: “Show them the data and the model results, but explain it in terms of their checkbook, their car, and their house. Because they can understand their checkbook, their car, and their house. If they can see the similar logic between that and what you’re talking about, it increases the absorption capacity immensely. And if you can make a joke, it’s even better.”
Question: You’re suggesting this integrated modeling approach can be improved and offer great benefits to the system. It’s pretty famously problematic to evaluate environmental outcomes using economic models and I’m wondering what kind of thought went into the process and social outcomes as well. How you develop and defend those when you’re talking to people within the context of the model?
Dr. Lund: “This is an imperfect process. Any analysis, any science is imperfect. Everyone with a PhD is basically trained as an assassin on anybody else’s science. That’s the environment we live in. You’re all trained assassins, you know that on some level. If you’ve ever had a paper reviewed, you know there are trained assassins out there.”
“In our modeling of the California system, we had a really interesting discussion and a very important decision early on. There were a bunch of academic economists that said, ‘you shouldn’t represent environmental flows as constraints; you should have contingent valuation studies of economic value of those flows for ecosystem services.’ We thought about it’ yeah, we could take that flow in that stream at that month and say, change in that flow will produce so many more salmon or so many more smelt, and then we have a societal value on each salmon or each smelt of amount of dollars per fish. Intellectually and mathematically I can see that. If I put that into a model, and then I present the results, what’s everybody going to talk about? ‘Why did you pick that value for the fish and why did you pick that production function for the fish and the function of flows?’ It’s not a useful discussion.”
“The purpose of the model was to help people think about operating the system. So it was just far easier to say, we’re going to give this much water to the fish, and it’s a constraint. They get their water first. If you think that number is too low, we can raise it and see what it does to the rest of the system. You have to organize problems – science problems, analytical problems, so that they can be solved. We have a lot of people discussing problems in ways in a framework where you’ll never get a solution out of it. There’s abundant empirical examples.”
“Our luxury as scientists, certainly at a university, is to cast the problem in a way where we think we can provide some insights. And in that particular case, it seemed to us pretty clear that if we cast it in the way the ecosystem service economist wanted us to cast it, the discussion wasn’t going to be helpful to the problem of managing the system. It would all become about the details of the model. The details of the model should not be important. They are important, I certainly understand that, but for the discussion, you want to help the discussion move along. You don’t want to set up the model, the analysis, and the science to encourage getting lost in the weeds.”
Question: The example in New York as I understand, what you were managing to as an endpoint, the New York example was this salinity intrusion in the lower Delaware River … I would like your thinking on the end points that we can reasonably approach with a model like you’re describing for the Delta system.
Dr. Lund: “I think it has to be driven by the problem. What do people think the problem is that you’re trying to solve? There’s sort of a social contract between science and the society: they pay us; we give them insights. If we don’t give them any insights to their problem, they’re probably going to send us less money, some day eventually. So the way they see the problem helps us to find the end points. Most of the water use in California, the end point for most of the diversions is economics which is why we built the model mostly around economics. And so city of New York, they had some end points that some of our regulators are interested in, which is salinity on the Delaware River, but their primary end point, objectives in the model, is turbidity for the most part moving into their UV disinfection plant that’s delivered to 9 million people.”
Question: Here we are in 2020. When you think about the CALVIN model, and think back about lessons learned form that model, can you think of ways where you would have changed what you did to effectively communicate the results of that model in order to better effect the trajectory of water system improvements? When you think about those lessons learned, how would you take some of those lessons learned and apply them in the Delta system now when you’re thinking about integrated modeling going forward?
Dr. Lund: “We had the luxury in 2001 of developing this model from an academic perspective. We had money from the state to do this from the Resources Agency, they were interested in what’s the value, what would the investors be willing to pay to help us expand capacities in different parts of the system. When we started to do this modeling, people were interested but didn’t think it could be done, so I don’t think they paid much attention to us. It’s probably still true today; they can’t believe this can be done.”
“We could be a little bit more of the mischievous 5 year old Calvin then the religious reformer Calvin. If you were the Department of Water Resources and allowed to do it today … you’d just get nothing but grief for an optimization model that said what to do. I think you’d have to do this as part of a bigger process that involves multiple agencies, gets you a little more coverage, and enough independent people so you can blame them for it, too. Then put it in a bigger context that addresses some of the other environmental issues and land use issues, environmental justice issues that come up with this model. We have to make it a bigger problem in order to organize the insights and supplement it with additional insights.”
Question: Some of the issues you are talking about are science communication issues. Since we have a science communications unit here, I was wondering what you see as a vision for science communication moving forward to address these issues, not only in the science program, but in the Council as a whole. Partially, it’s bringing all these agencies together, how do you see that working?
Dr. Lund: “Collectively in the community of scientists around the Delta and California, we’ve made a real leap in science communications and I think it’s been very good for us as scientists to think about what’s important and how we’re going to communicate it. Where I think we’ve not taken full advantage of that is to think about how do we orchestrate the messages. There’s some common core to the messages that we want to get across, some common perspective or common framework that we want to get the decisionmakers to understand. Just communicating ‘so and so found this thing in their study and so and so found this thing in that study’ – this is how we tend to do most science communication. We write a paper, a science communications person writes something, we have a thing for these policymakers to think about. And in fact, if we give them lots of weeds, they are going to be confused with us.”
“The hardest part and the most important part is giving them a more coherent framework to understand science and to put the pieces together so they have the impression that we kind of understand something about what we’re doing here. They have a role in it; this common framework can help them think through what are very difficult problems for them. If you’re a policymaker, agency head or elected official in this system, you’ve got numbers of different masters out there that are all throwing tomatoes at you and things like that. If you can bind to a common understanding of the problem and the directions for some of the solutions, I think it helps everybody. It helps direct our science as well.”
Question: In your remarks, you talked about needing to make bigger, faster incremental changes to keep up with what’s happening on the ground, and you also talked about integrated modeling and more integration and more coordination. I often come to the old saying, ‘if you want to go fast, go alone; if you want to go far, go together.’ But I think those two things eventually push against each other, we have to go fast and we have to go far. So how do you think about balancing those two factors and getting the right mix?
Dr. Lund: “For Chesapeake Bay, they had a mandate from the federal legislation with EPA and had EPA money that said, ‘here is a set of timetables that you have to have for your regulations and your TMDLs. And we’re going to give a structure so you have to move forward together to do it’ But if you want to go slowly, getting everybody to spontaneously organize is a really good way to go slowly. Sometimes I fear that we have used consensus words as an excuse for not doing anything controversial or important. We have water action plans, we have a lot of non-action plans.”
Question: There’s well over 500 water agencies in this state. You touched on the fact that they all are territorial as one might expect, they have to be, and they also have different authorities and different inventories of things based upon when they were formed and how they do that. So my question is a little broader. What’s the currency that actually makes a plan work here? Is it political muscle, is it dollars, is it convincing people it’s a good idea to do these things? Which has not actually worked well in recent years. All these agencies, and I must admit I was part of that, we were very adept at stopping things. So what’s the currency that makes this thing move?
Dr. Lund: “When I was first thinking about applying for this job, I talked to someone who said, one of the things that has me worried the most is, I’m coming into an environment that’s very good at saying no, a gauntlet of no. I think you have to have, in the case of EPA Chesapeake Bay, they had federal legislation, money, and mandate to do it on time, all together. If you didn’t do it all together, then EPA like with the Delta agreement back in the 1990s, then the EPA could come and do it by itself, and nobody in the state wants that. That’s one way to do it. You orchestrate some need to converge in a timely way and a framework for bringing people along.”
“But what happens if you don’t do that. Well, you’re either stuck with brutal incrementalism, or you’re stuck with doing incremental in most areas so you pick a few areas where you say, I think these people and these agencies are ripe for making progress in this area, and that’s where we should invest. We’re going to rely partly on the good example of hopefully success there in bringing people along. … Often, in this system you’ll spend ten years of your career doing studies and talking to people, and then when flood, drought, or lawsuits and then they are really interested in what you have to say. … It’s an imperfect world, and it’s going to be incremental. It’s not going to be magic, it’s not going to be revolution, it’s going to be incremental. We’re going to perceive it as brutally incremental but we ought to be able to do better than we are now.”
Question: In your presentation, you mentioned that the Delta is changing faster than science and management can keep up with. As the lead scientist, how do you propose dealing with that issue?
Dr. Lund: “I think you have to move faster. A lot of this change is exogenous. New invasive species show up, the climate is changing, changes in the price structure of crops, new listed species – all kinds of things. We can’t stop change. All we can do is organize ourselves to be better prepared for it and to better respond to it. I think science ought to have a vanguard role in that. We ought to be able to tease out what those changes are, the most important of those changes. There will always be uncertainties about what those changes are, but we kind of know what most of them are qualitatively and in some general orders of magnitude of how big they are likely to be, so we’d better prepare for them.”