ADAPTIVE MANAGEMENT FORUM, Part 1: Revisiting the Foundations and Planning for the Future

Click here for the Adaptive Management Forum Home Page.

Although the 2009 Delta Reform Act now legally requires adaptive management for all restoration efforts in the Sacramento-San Joaquin Delta; implementation remains somewhat challenging.  While adaptive management has been successfully implemented in some projects, for others, success has remained elusive.

The biennial Adaptive Management Forum presents an opportunity for practitioners and decision-makers to learn from the successes, promote coordination, and address the challenges as we work towards successful implementation in the Delta.  Hosted by the Delta Science Program, the Adaptive Management Forum provides an opportunity for the Delta community to share knowledge and promote collaboration on adaptive management of the system.

After reflecting on the first Adaptive Management Forum held in 2019, the planning team decided to focus on four goals: synthesis, education, networking, and dialogue.  To achieve those goals, the planning team invited presentations that synthesize information or share information that can be synthesized.  The sessions were designed to illustrate how adaptive management is practiced and how data can be transformed into actionable knowledge.  The forum’s agenda included time for interactivity to facilitate networking, mutual learning, and co-production

This session was hosted by Karen Kayfetz, Environmental Program Manager at Delta Stewardship Council.


Dr. Laurel Larsen, Delta Lead Scientist

Dr. Laurel Larsen then welcomed everybody to the forum, noting that adaptive management is at the core of realizing the vision of One Delta, One Science.

Adaptive management is by nature a collaborative system, focused and responsive to changing conditions, making it an ideal tool for dealing with the many challenges that make the Delta a so-called wicked problem,” said Dr. Larsen.  “It is also structured to maximally aid in decision making in the face of highly uncertain inputs and environmental conditions, which characterize the future of California. Adaptive management formalizes the application of science to on-the-ground management. It is a structured, systematic approach to learning and knowledge integration.”

She pointed out that adaptive management isn’t just applied science.  Many pieces need to be in place in practice and be well-coordinated to support adaptive management, including planning and permitting, monitoring systems, and funding. It also requires clear communication between scientists, managers, and stakeholders and the establishment of trust.

With respect to this communication, this adaptive management forum takes a big step in the right direction. One of the big questions that we face in the Delta is how to practice adaptive management, not only at the individual project scale, but also at a much larger landscape scale. In other words, how do we adaptively manage the dynamic socio-ecological system that is the Delta holistically with the objective of creating a resilient landscape in the face of major change?

In short, adaptive management is complex, and there are certainly challenges. This forum brings together scientists, managers, regulators, planners, and other stakeholders to explore our different roles in adaptive management and demonstrate creativity in addressing salient challenges. Building a community around adaptive management will strengthen our ability to confront our shared challenges and implement adaptive management at the landscape scale.

Susan Tatayon, Chair of the Delta Stewardship Council

Chair Susan Tatayon began by noting that this is the only forum dedicated to advancing the collective understanding of adaptive management and its benefits to the Delta region.  Adaptive management is an explicit charge of the Delta Plan; the Delta Reform Act states that Delta Plan shall include a science-based transparent and formal adaptive management strategy for ongoing ecosystem restoration and water management decisions. The Act also specifies that the Delta science program shall coordinate with Delta agencies to promote science-based adaptive management.

In light of global climate change and the Delta ecosystem’s ever-changing nature, successful adaptive management is more important than ever, she said.  The Delta Independent Science Board’s journal article, Preparing for a Fast Forward Future in the Sacramento San Joaquin Delta, states: as the effects of climate change gain force, the environment is changing more rapidly, and droughts, heatwaves, forest fires, and floods are becoming more extreme.  Ecosystems are being driven to or beyond thresholds of irreversible change.  As these climate-driven changes compound with ongoing changes in land use and other drivers, uncertainty increases. So as we grapple with climate change and rapidly changing conditions in the Delta, we must be equipped to learn and adapt more quickly if we are to achieve the state’s coequal goals.

However, Chair Tatayon acknowledged that creating an adaptive management plan is a demanding and time-consuming process that requires collaboration among scientists, managers, decision-makers, and stakeholders. And carrying out an adaptive management plan tends to be a slow process.

This dilemma leads me to many questions,” she said.  “One, what constitutes a successful adaptive management program? Two, are there ways to carry out the nine-step adaptive management cycle more quickly, without giving up on what’s best in the process? Three, are there tools available for concisely communicating the benefits of adaptive management to overburdened leaders and policymakers?

Chair Tatayon then closed by saying, “The caliber, diversity, and number of participants and presenters at this forum gives me hope that we will find ingenious ways to confront increasingly rapid change in the Delta and integrate our adaptive management actions. Thank you for your interest and your efforts in advancing our collective understanding of adaptive management and its benefits and challenges.”

PLENARY SESSION: Plenary: Revisiting the Foundations and Planning for the Future

The first panel discussed how adaptive management has evolved in the Sacramento San Joaquin Bay-Delta region from before it was a mandate in the Delta Reform Act to today, emphasizing lessons learned and the pieces of history that remain relevant to adaptive management now and in the future.

PHIL ISENBERG: Defining success for managing Delta resources (and defining failure as well)

The first speaker was Phil Isenberg, the founding Chair of the Delta Stewardship Council. Across his long career in state and local government, Mr. Isenberg has served in numerous roles, including Sacramento City Council member, Mayor of Sacramento, and a California Assemblymember.  As an elected official, he focused on land use planning, water and resource issues; state budget and fiscal matters; redevelopment reform; and healthcare.   More recently, Mr. Isenberg served as the Chair of the Delta Vision Blue Ribbon task force, a predecessor to the Delta Stewardship Council, which provided much of the structure for the major changes in water and Delta policy enacted with the 2009 Delta Reform Act.

Mr. Isenberg began by saying that, in his view, this is a public policy issue and not a technical issue.  His impression is that for all the other victories that the Delta Stewardship Council has accomplished, the one that is lagging behind and is more amorphous than it should be is adaptive management.  

The legislature mandated the development of a legally enforceable Delta plan and legally enforceable elements of that plan that relate to science, and that’s fundamentally important,” said Mr. Isenberg.  “As folks who are doing the hands-on work on adaptive management, you ought to consciously and deliberately spend your time thinking of ways to include within the mandatory provisions of the Delta plan in whatever it is that you want to do with adaptive management.”

Secondly, the council was equally and surprisingly, in my view, given the authority to develop the Delta plan,” he continued.  “The legislature explicitly adopted our so-called coequal goals to improve the Delta and make water more reliable for California without specifying who is included in that puzzle. That’s important because it’s the first statutory declaration that attempted to give a balance to both ecosystem and water use dependent on the Delta. As part of an enforceable Delta plan, that gives additional clout to environmental stuff.”

As a footnote, I think adaptive management spends too much of its time on ecosystem stuff because that’s where it gets a friendly reception, and not enough yet on water development and major water facility projects, which is where the vast public controversies exist.”

The Delta Reform Act explicitly stated that California’s policy is to reduce reliance on the Delta in meeting California’s future water supply needs.  The legislation also listed numerous environmental provisions that are part of an enforceable Delta plan and created the Delta Independent Science Board, the Delta Science Program, and the position of Delta Lead Scientist.

Best available science, whatever that means, is a mandated part of the legally enforceable Delta Plan,” said Mr. Isenberg.  “My own experience is that scientists are generally reluctant to say what’s the best available science; they spent a lot of time at least talking about it, and it’s become more real. The adaptive management part, it seems to me, is still a puzzle. Most of the other elements in the Delta plan and for the Council have been resolved through endless litigation, a lot of political battles, and some creative staff work.”

The point I want to make is that the process – litigation, mandatory requirements, system reviews – all carry clout, and that clout is helped by fighting and winning through on battles.  And I think that’s one of the problems – and one of the opportunities that adaptive management has to make.”

” … the most important thing in policymaking and government is not thinking up what to do and not passing laws or imposing regulations. It’s figuring out how real people on the ground behave.  If you can’t change human behavior, you’re not going to get any good results.”

I’ve always viewed adaptive management as kind of fuzzy,” continued Mr. Isenberg.  “It requires, as do all processes that are required to be done, clear and measurable outcomes. But we tend to focus on the process; we have hearings, we consult with people, and so on. And we haven’t really nailed down the adaptive management outcomes that we want to see.  More precisely, I don’t think there has yet been litigation to stop a water development project because of the insufficiency of their adaptive management plan. Until that happens, it’s going to be awfully hard to figure out how to do adaptive management without it just simply being left for endless conversations between scientists, occasional government lawyers, interest groups, and not being publicly understandable at all.”

Mr. Isenberg said that he learned a lot during his time at the Delta Stewardship Council and the years prior through the Delta Vision Blue Ribbon Task Force. 

I’ve learned that unless you figure ways to make actions actually happen, as opposed to being talked about, adaptive management’s going to eventually kind of bleed off into something that isn’t as important as it has to be.  After all, the most important thing in policymaking and government is not thinking up what to do and not passing laws or imposing regulations. It’s figuring out how real people on the ground behave.  If you can’t change human behavior, you’re not going to get any good results.

Karen Kayfetz then thanked Mr. Isenberg for reminding the importance of adaptive management, not only when it is welcome but when it is hard.  And for bringing everyone back to thinking about how projects that change the landscape or change our water management practices affect people and how we can use adaptive management to navigate controversial projects. He also provided a solemn reminder that even with legislation, but without litigation, we may not truly have the teeth to enforce the use of an adaptive management approach.


Dr. Mike Healey is a professor emeritus at the University of British Columbia, where his research focuses on the ecology and life history of Pacific salmon. Dr. Healey has also worked extensively to advance the application of science in water resources and fisheries management. Among these efforts, from 2004 to 2006, he served as a member of the Delta Independent Science Board, contributing his expertise as the board established its role in providing scientific oversight in the Delta. In 2007, and 2008, he was the lead scientist for the Delta Science Program. In 2007, he served as a science advisor to the Delta Vision Blue Ribbon Task Force. More recently, in his retirement, he has directed his energy towards raising awareness on the implications of climate change for society and the environment.

Dr. Mike Healey’s talk focused on the history of adaptive management during the CalFed years and into the Delta Stewardship Council’s early days. In 1995, he joined the newly created Science Advisory Board in the CalFed Ecosystem Restoration Program. Cal Fed was charged with using adaptive management to design and implement its water and environmental management program, but at the time, no one really knew what adaptive management was. Because he had worked with key people responsible for adaptive management, he was able to shed some light on what it was and how it might be incorporated into the CalFed program.

The origins of adaptive management

The idea of adaptive management began in the 1960s and 1970s as a new set of technologies came into play in environmental science and business that were based on concepts from engineering, such as systems analysis and optimal control theory; from economics and psychology, such as decision theory and structured approaches to decision-making; and from computer science, such as systems analysis and dynamic programming.

A new generation of professionals wanted to use these new technologies to make better decisions in business, and better decisions in resource and environmental management,” said Dr. Healey.  “Buzz Howling was one of those professionals.  In 1977, Buzz and Carl went off to the International Institute for Applied System and Analysis in Austria to consolidate their ideas with input from like-minded colleagues from various places around the world, and they produced the foundational book on adaptive management.”

The book, Adaptive Environmental Assessment and Management, was published by the Institute in 1978. The first section of the book lays out the intellectual and technical framework of adaptive management; the second half is devoted to how adaptive management can be applied and includes four case studies: the management of spruce budworm outbreaks in the forests in New Brunswick, the management of salmon fisheries on the Pacific Coast, land use planning for a resort area in Austria, and the construction of dams in Venezuela.

After several years of working with resource management agencies trying to implement adaptive management, Carl Walters wrote his own book on adaptive management titled, Adaptive Management of Renewable Resources.  It was a much more technical book than Hollings that is full of critical information about adaptive management.

I’d be really interested to know how many of you in the audience are working in adaptive management have read one or both of these books,” said Dr. Healey. “Neither is referenced in the adaptive management section of the Delta plan, but these are the foundational books on the subject of adaptive management. And in my view, there should be on the shelves of everyone who’s working in adaptive management.”

In Holling’s book, Adaptive Environmental Assessment and Management, he laid his vision for applying adaptive management. He envisioned a small team of adaptive management specialists who would lead the process and engage with decision-makers to do a preliminary scoping of the problem. The adaptive management team would then convene a workshop with decision-makers, specialists, and stakeholders to clarify the problem, discuss available information for dealing with the problem, identify the major uncertainties, and consider the management options.

The adaptive management team would then draft an initial version of the system model that they would use to explore how to manage or how to deal with the problem.  They would then reconvene, review the draft models with the workshop participants, discuss how the model might be applied, and work out any issues that were identified. After the workshop, the adaptive management team will work with specialists, scientists, economists, and sociologists to clarify policy options and the model. Then they convene a second workshop to share preliminary results and make further adjustments to the model.

Holling’s view was that this is a highly collaborative process, and the model was central to the whole process,” said Dr. Healey.  “These workshops were designed not only to share information but also to get all of the participants comfortable with the model as a tool for exploring how to manage the system they were concerned about.”

After the second workshop, they would further discuss the model, make any necessary adjustments, talk about what uncertainties remained, and how to address those.  The adaptive management team would update the model and the policy options and then use the model to test policy outcomes and report all of this to decision-makers. So at this stage, they would begin to get to what the expected results would be.  

Then decision-makers would choose their preferred policies. The adaptive management team would work with the managing agencies to design implementation and monitoring for addressing the problems they had been discussing. After this, the agencies would implement their aspects of the program with coordination by the adaptive management team.  As the monitoring data came in, the adaptive management team and the agency staff would use those data to evaluate the effects of the program, report those to the decision-makers, and recommend any program modification.

Dr. Healey said there are two important takeaways:

  1. Holling envisioned the adaptive management process to be built around a numerical model of a system used to game policy options and explore management outcomes.
  2. It was a highly collaborative process so that everyone engaged understood what was going on, including the policymakers and the public, so there was no mystery about what was being done and why it was being done.

Adaptive management in major US environmental programs and CalFed

In the 1980s-90s, adaptive management became very popular in the United States. The resource agencies and the federal government were all tasked with using adaptive management in carrying out their mandates. As a result, several large-scale environmental projects that started during that period were charged with employing adaptive management.  These included salmon restoration on the Columbia River initiated in 1984; habitat mitigation in the Missouri River, authorized in 1986; restoration of the Kissimmee River, authorized in 1992; and the Colorado River Glen Canyon adaptive management program, initiated in 1995.  Those programs are all still ongoing.  Other federal agencies, such as forestry and agriculture, were also employing adaptive management at the time. 

Something interesting is that adaptive management evolved differently in each of these projects, and hardly any of them unfolded in the way that Howling envisioned,” said Dr. Healey.  “In the Bay-Delta, there were several antecedents, both to the Cal Fed Bay-Delta program and the implementation of adaptive management.  Water management in California has always been contentious, and lots of user groups who want water are all competing to get a bigger share of the resource.”

Complex environmental consequences were generally considered secondary until 1992, when the Central Valley Project Improvement Act was passed, which required that fish and wildlife values be given equal consideration with other water uses.

So lots of things had to change, said Dr. Healey.  The first thing that came out of this was the Anadromous Fish Restoration Program.  In 1994, just as the Anadromous Fish Restoration Program was beginning to be implemented, government agencies, water users, and environmental groups agreed on an approach to water and environmental management in California called the 1994 Bay-Delta Accord.  Out of the Bay-Delta Accord, Cal Fed was established to be a coordinating agency to carry forward the goals of the Bay-Delta Accord using adaptive management.

My understanding is that adaptive management was specifically mandated for the Cal Fed program,” he said.  “That was about the time that I joined the Ecosystem Restoration Program Science Advisory Team.  I found out that they had to apply adaptive management, but nobody involved really understood what adaptive management was. So with advice from the external advisory group, the Ecosystem Restoration Program created a strategic planning team to provide a strategic plan for ecosystem restoration. And that planning core team included myself, Wim Kimmerer, Matt Kondolf, Peter Moyle, Roderick Mead, and Robert Twiss.  We met several times over about a four-month period to hash out a strategic plan for ecosystem restoration within the Cal Fed that was built around an adaptive management framework.”

The strategic plan that was produced by the core team included a detailed description of the process of adaptive management that was drawn from Hollings and Walter’s seminal works.  That plan was then incorporated virtually unchanged into the Ecosystem Restoration Program strategic plan published in 2000.  The adaptive management process described in that plan has been carried forward through successive Cal Fed and Delta Science Program documents. It hasn’t changed a great deal over that couple of decades since it was first put forward within the Ecosystem Restoration Program.

Dr. Healey presented the conceptual model flowchart for adaptive management published in the strategic plan. The key points:

It begins with a clear specification of what the problem is you’re going to tackle.  Although that sounds simple, it’s often not as easy as it sounds, Dr. Healey said.

It includes a clear set of goals and objectives for what is to be achieved.  “As Phil was saying, you have to identify what it is you’re trying to accomplish so you can measure progress.  I think Yogi Berra said it best, if you don’t know where you’re going, you have to be careful because you might not get there.”

It includes a conceptual model of how the system to be managed functions so that a qualitative assessment of how the management will lead to particular outcomes can be examined.  “The description of Hollings’s expectations for adaptive management is centered around a numerical model that will be used to explore and game policy outcomes.  When we talked about this with the agency people in the Ecosystem Restoration Program and within the core team, we were assured that there really wasn’t the expertise in the agencies or the people that they had for their consultants actually to build the kind of numerical models that Hollings envisioned.  So we settled on using conceptual models as a basis for at least making some preliminary assessment of how different management policies might unfold.”

The policy would be implemented, and the system would be monitored.  The monitoring data would then be used to assess whether or not the policy and management program was working.

Depending on the assessment, there were various pathways: If things are going well, you continue the policy. If things weren’t going well, you might reassess the model to see if you could improve things by having a better model, you might set new goals, or you might have to rethink what the problem was you were working on completely.

Some of this was familiar to resource management agencies, but some of it was not common practice, such as conceptual models. 

To my surprise, I found out it was not common for agencies to monitor the managed system very much or at least not in a formal way,” said Dr. Healey.  “Many of the funding sources for implementing management programs could not be used for monitoring, and that box on the left-hand side – the assessing, evaluating, and reporting process – was generally not a formal part of the management plan. So there was a whole lot in here that was pretty new and not part of the typical approach to environmental management in California at the time.”

We were assured that this is going to be a pretty hard sell with the management agencies,” Dr. Healey continued.  “I don’t mean any disrespect to the management agencies and the people involved in the state and federal agencies – it turns out that adaptive management is a pretty hard sell everywhere. And there are several reasons for that.”

In California, those obstacles included:

It was a new process, so there was no agency experience with adaptive management.   A lot of it was not things that the agencies routinely did. Agencies over time have worked out their own approach to doing the resource and environmental management, and adaptive management didn’t fit very well with that process.

There was no local expertise trained in adaptive management, so there were no experts locally that the agencies could turn to for help and advice.

A number of the elements of adaptive management were foreign to agency staff and decision-makers, such as the collaborative approach to problem definition, numerical models to gain policy outcomes, using management to generate information, and monitoring to assess management.  These things cost money, so it wasn’t that easy to incorporate these elements into the management process.

The common definition of adaptive management was learning while doing, so it allowed agencies to believe there really wasn’t anything new in adaptive management.   The agencies felt that was how they were proceeding anyways.

Agencies tend to redefine a new approach as what it already does.  Dr. Healey said the reason for this is that they’re often told to do something new without receiving any new resources to do that, and so the easiest solution is to redefine what you’ve been told to do as part of what you’re already doing.

Funding in some cases could not be used for monitoring, and monitoring is central to adaptive management.

Cal Fed had a lot of money that had to be spent quickly, so there really wasn’t any time to inform and train or build consensus about adaptive management.

So not surprisingly, Dr. Healey said the implementation of adaptive management was somewhat haphazard.

Nevertheless, several adaptive management projects did get underway during Cal Fed, primarily in the tributaries upstream because, at that time, conservation of chinook and steelhead was the priority; adaptive management projects got underway in the Delta as well. 

Some of the adaptive management projects initiated in the early days of Cal Fed were:

    • Flow and habitat restoration for chinook salmon on Clear Creek;
    • Flow and habitat restoration for chinook and rationalization of the Coleman hatchery in Battle Creek;
    • Flow and habitat restoration for chinook salmon in the Merced River and the Tuolumne;
    • Wetland restoration of Dutch Slough in the Delta, McCormack Williamson Tract, Liberty Island, and others. 
    • The Vernalis Adaptive Management Program, which looked at how the effects of flow and the Old River barrier on the survival of chinook smolts in the San Joaquin system.

Many of those never got much past the planning stage, said Dr. Healey.  “I think some of them are still sort of sitting in the planning stage, which comes back to you can’t just endlessly plan; you have to get on with it.  In fact, one of the purposes of adaptive management was to help you get past that planning stage. If adaptive management is done as anticipated, then you shouldn’t take more than a few weeks to get to the point where you can actually start implementing management.  But it requires courage to make those management decisions. And if there were other obstacles, as there are in California, acquiring land, getting permits, all of those regulatory things, some things just have a great struggle to get underway.”

Some of these projects were only loosely connected with adaptive management, which is okay – you have to learn while doing, Dr. Healey said.  Some of them follow the script quite well, such as the Clear Creek restoration project, which utilized consultants who understood adaptive management.

In confronting the difficulties of implementing adaptive management in all the major projects, Cal fed began to develop its own version of adaptive management.  “Given the institutional and expertise constraints within the state and federal agencies who are responsible for implementing adaptive management, the Cal Fed adaptive management model seemed to me to be not too bad a compromise,” said Dr. Healey.

He presented a diagram of the Cal-Fed adaptive management approach, noting that while it looks a lot like the adaptive management flowchart shown earlier, there some important differences, though. First of all, the Cal Fed version maintains the traditional separation between management and science.  Management retains responsibility for the operational implementation of management programs, while science was charged with dealing with uncertainties and complexities through things like targeted research and pilot projects to prove concepts.  There was collaboration but not quite what Holling had envisioned.  The preferred management pathway was defining the problem, considering the management options, choosing one and implementing it, and then assuming that it would have the kind of positive outcomes that everybody anticipated, he said.

The Cal Fed model included close attention to monitoring,” said Dr. Healey.  “There was the possibility of early on discovering that the management plan is not working as well as hoped, so corrective action could be taken either by redefining the management options or redefining the problem.  When this started to develop, I was fairly happy because I thought this captured most of the critical elements of adaptive management, even if it wasn’t exactly in the format that Holling anticipated.”

Adaptive Management and the Blue RibbonTask Force/Delta Plan

When the Cal Fed structure lost the support of the participating agencies and outside interests, the problems that Cal Fed had been created to address didn’t go away. So in 2007, Governor Schwarzenegger appointed the Delta Vision Blue Ribbon Task Force to develop a new vision for water and environmental management in the Delta.

While Cal Fed had looked at the entire watershed out to the ocean, the Delta Vision process focused only on the Delta, leaving issues upstream and downstream to be handled by the relevant state and federal agencies. The Blue Ribbon Task Force report specified that a healthy Delta ecosystem and a reliable water supply should be the coequal goals of managing the Delta. In addition, management had to be mindful of the importance of Delta as a place where people live, work, and play.

The Delta vision document made 12 specific recommendations that were codified in the 2009 Delta Reform Act.  The Delta Reform Act requires explicitly that adaptive management be the management framework for achieving the coequal goals.  The first Delta Plan, adopted in 2013, laid out the path for how the coequal goals would be achieved and how adaptive management will inform the management of the Delta.  

The adaptive management cycle described in the Delta plan is virtually the same one that was presented in the ecosystem restoration program strategic plan, although the design is definitely a lot nicer,” said Dr. Healey.  “I have to say it’s gratifying to me that the Delta Plan has held on to the original concept of adaptive management and continues to try to implement it.  However, one thing that I thought was a serious omission from the Delta plan is it neither Hollings nor Walters’s books are referenced in the Delta plan.”

What has been gained from Adaptive Management?

Some have raised the question, how will we know it is working? How will we know it’s a success? Dr. Healey said some specific expected products of adaptive management might make it worthwhile, but he acknowledged that the decision about whether or not adaptive management is worth the time and trouble must rest with the agencies charged with implementing it.

The products of adaptive management:

A clear specification of the problem, the management options, and the expected outcomes that everybody understands and generally agrees with.

A model, preferably a numerical model, that will allow managers to explore the consequences of management interventions.  This should be a model built in collaboration with all interested parties so that everybody understands the model, what it is expected to do, and agrees with it being a useful tool.

An improved understanding of how the system functions.  The main function of adaptive management is to design management interventions that generate information about how the system works so that knowledge and understanding are a product as well as the stimulus for management.

A more collaborative understanding of the system among the scientists, managers, decision-makers, and the public.

A way of using the monitoring data to update management learning while doing.  There should be a clear pathway from information gathered as a result of implementing a management program through to the whole reconsideration and evaluation of the program overall.

Most importantly, more effective management.  “If you aren’t getting this, then the rest of it, however interesting it may be, hardly seems worthwhile,” said Dr. Healey.


Adaptive management provides a structured, rational framework for the theory and practice of resource management. It’s a fully consistent system. But despite its attractiveness, in theory, very few management agencies have incorporated adaptive management into their standard procedures.

I mentioned earlier that there were an awful lot of programs were initiated, employing adaptive management in the 1980s and 1990s,” said Dr. Healey.  “Adaptive management survives in a recognizable form in a few of those, and the Bay-Delta system is one of the few where adaptive management is still being practiced. So I salute you for that.  But of course, you’re required by law to use adaptive management as the organizing framework for ecosystem management. However, there would be lots of ways around that if you didn’t want to do it.”

During the time that I was involved in the Delta, primarily from 1995, to 2009, many elements of adaptive management were incorporated into environmental management.  I would guess the new people joining the agencies would have no clue as to how things were done in the command and control process of 20 years ago. And finally, whether or not this has improved environmental management remains subject to debate.  We look up to those engaged in the practice to answer that question.”

As Susan said in her opening remarks, one of the charges of the Delta science program is to promote the use of science-based adaptive management,” said Karen Kayfetz.  “I’ve been working at the Delta Science Program on adaptive management for about six years now. But even just in that short time, I’ve noticed a great shift in how much we’ve had to promote adaptive management as a set of foreign concepts like you were outlining. I feel like now our job is less about getting overall buy-in and much more about providing advice and tools to help projects do the best job that they can do with adaptive management. So it does represent a huge shift from where you outlined we came from 20 years ago.”

DR. LAUREL LARSEN: Looking forward to an uncertain future

Dr. Laurel Larson is the Delta Lead Scientist and an associate professor in the Department of Geography and Civil and Environmental Engineering at UC Berkeley.  She closed out the session by looking ahead to an uncertain future.

Uncertainty is a necessary prerequisite for adaptive management being an effective strategy as indicated in this graphic prepared by the Delta Independent Science Board,” said Dr. Larsen.  “After years of fire, flood, mudslides, heat, and drought, the only thing that is certain about California’s future is that it will bring increased levels of uncertainty. Climate models predict statistically significant increases in the frequency of wet extremes relative to pre-industrial levels before we are halfway through with this century, and likely, but not as certain, increases in the frequency of dry extremes.”

At the end of the last drought, Daniel Swain, now at UCLA, popularized the term weather whiplash, which refers to the back-to-back occurrence of dry extremes and what extremes, which in turn creates storage challenges and conditions ripe for the occurrence of disastrous wildfires. These whiplash events likewise are projected to likely increase in frequency.

This increasing frequency of hydrologic extremes and whiplash events, together with the incomplete understanding of the processes that would need to go into creating forecasts of specific events, create grave challenges for water and ecosystem management,” said Dr. Larsen.  “Many water management decisions are based on historical averages obtained from decades with fewer extremes, and often do not consider the possibility of opposite extremes in the next water year.

On top of the increase in climate variability, we are also faced with the challenges of increases in mean temperature, sea level, and possibly annual precipitation totals, as summarized in the recent climate vulnerability assessment by the Delta Adapts program within the Delta Stewardship Council.  While not on the chart, equally important is the change in the form of precipitation, replacing rain with snow in the Sierras.  The timing of that precipitation will result in more delivery of water year totals to the Delta during the wet season, when it’s difficult to capture in storage, and less delivery during the dry season.

As reported in the vulnerability assessment, these changes will contribute to a host of socio-ecological challenges throughout the whole Delta system,” she said.  “One thing I would particularly like to point out is that while we often focus on the altered temporal sequencing of climate inputs as a major challenge for management, spatial heterogeneity, as exemplified here by spatial heterogeneity and social vulnerability, adds an additional layer of complexity. In thinking about management strategies for adaptation and system-wide resilience, a one size fits all approach is simply not going to cut it here.”

However, there is good news as well.  While spatial heterogeneity has its challenges, it also presents opportunities:

One is that because this is not a static landscape, attributes that we value within the Delta may have some freedom to move around within the Delta in response to these stressors without being lost, which is only a possibility because of the heterogeneity of our system,” she said.  “Heterogeneity of elevations, for example, while slight, may be conducive to some ecosystems moving to slightly higher ground in the face of sea level rise.  Because urban development and engineering hardscape are not as prevalent in the Delta as in the Bay, this may be an asset that we could strategically use to a greater extent here than our counterpart planners in the Bay.”

Second, in creating inundation maps, we tend to neglect that tidal wetlands are natural buffers against sea level rise, responding by accreting to higher ground,” she continued. “This arises as a feedback between water levels, vegetation, productivity, and decomposition rates; and vegetation sediment trapping capability. Though not all tidal wetlands can keep up with sea level rise, particularly if sediment supply is not sufficient, this is a landscape feature that should not be given short shrift in system-wide adaptive management.”

Finally, there has been a lot of attention among terrestrial ecologists in California given to the concept of climate refugia,” she said.  “The basic concept is that because of microclimates, varied topography, and other geographic elements, and local hydrology that create local areas of groundwater discharge, not all locations within a landscape will change to an equal extent in response to climate change. In fact, some areas may not exhibit much change in habitat characteristics at all, while others may change only at very advanced stages of climate change. These areas serve as local refugia for species present within a region and may allow them to endure for longer periods of time.

Though there has been a considerable effort recently invested in mapping refugia for terrestrial vegetation statewide, I am unaware of similar efforts to map thermal or hydrologic refugia within the Delta,” she said.  “Such an effort could fit very nicely into an adaptive management cycle that considers landscape-scale adaptive management needs.  In fact, all of the choices, challenges, and opportunities created by temporal uncertainty and spatial heterogeneity would benefit from a landscape-scale perspective. Moving towards this landscape-scale perspective is a key challenge that I would like to task this group with addressing in your discussions and ongoing work.”


Developing models

Question: Dr. Denise Reed noted that Dr. Healey made the point that there was resistance to spending time on developing numerical models early in the process.  “How much of a hindrance do you think there has been? We have this idea of having an agreed-upon expected outcome for the project to be important to adaptive management. I wonder, without numerical tools and quantitative predictions, how does that really limit our success fundamentally here?”

I think to an extent it does,” said Dr. Healey.  “But I’ve felt that during my time here, that it was critical to at least get the conceptual models worked out.  You were involved in that too – the DRERIP program.  This was not an easy task to do even just to get the conceptual models pinned down. And it was even harder to get them assembled somewhere and looked at from time to time so that they could be updated.  I think it really would have been virtually impossible to pursue a Hollings numerical model approach.”

But even getting the conceptual models out, and having some agreement on those I think was a big step forward, but it is still a hindrance to doing any kind of gaming with policies to determine what the consequences of a particular management action might be,” Dr. Healey continued.  “So I’d like to see the Delta Science Program and the Adaptive Management Program try to evolve towards those numerical models. Again, part of the problem was, on the core team, when we were reading the strategic planning document, they didn’t really understand the nature of the numerical models that Hollings was talking about.  They were thinking about very detailed, heavily information-based models, whereas what Holling and his crew did was they’d pick the conceptual models out of the heads of all of the people involved in a discussion and put those together, so that all of those models and sub-models are already familiar to the people involved. And by putting them together, they can see how they interacted. And that was a huge benefit in getting everybody on the same page.”

Now, Carl Walters and his students were incredibly skillful at doing that. If you could assemble a team that could do that, you could have numerical models of Delta processes put together pretty quickly. And I think those would be useful to the whole enterprise.”

Monitoring on an individual vs. regional basis

Question from Dr. Ted Sommer:  “I was wondering if we could perhaps differentiate between broader management of the region and individual projects.  My two cents worth is that it can be just fine for an individual project to have a really good monitoring program, including some experimentation, and not have a quantitative model. But I feel like the quantitative models are more useful to managers looking at the big picture across the region.”

The small local projects that are designed as an experiment, I’m sure that you can get by without a numerical model,” said Dr. Healey. “But if you’re thinking about the system as a whole, then each of those local projects should somehow be nested into the system as a whole.  So you need to know how it relates in a landscape sense to all the other things that are going on. And usually, those little local projects are stimulated by some uncertainty that relates to the larger picture. So I think at some point, they would need to be incorporated into the numerical evaluation of what’s happened.”

I would agree with that completely,” said Dr. Larsen.  “I think a lot of times the smaller projects are designed to test uncertainties that are related to incomplete knowledge of processes within the system. And those processes need further study to be incorporated into the numerical models used to scale up our understanding of the entire system. So I agree with both of you.”

Data uncertainty and model development

Question from Dr. Shawn Acuna:  Regarding uncertainty, another issue is the data going into these models.  “If we try to produce a quantitative model, what are your thoughts on the uncertainty if the data is not sufficient enough or felt not to be sufficient enough to create a model with sufficient certainty? That’s kind of subjective. What do you call sufficient certainty? But does it mean that you should definitely rely more on a conceptual model as opposed to the numerical model until you have enough data to reduce that uncertainty?”

You can put together a preliminary numerical model using even bad data or guesses if that’s all you can do,” said Dr. Healey.  “The first thing you can do is run a sensitivity analysis and see whether or not that aspect of the system is really having a big impact on the way things unfold. If it’s not, you can set it aside and say that we won’t spend a lot of time dealing with the uncertainty in this aspect, because at least in terms of our overall model, it is not having much impact. If it has a huge impact on the output of the model, then it tells you to get busy and firm this up.”

But I will say something else, too,” said Dr. Healey.  “One of the obstacles in other locations to making progress with adaptive management is people get trapped in the modeling phase; they’re always so concerned that the model has to be perfect. And we never get to any implementation.”

Dr. Larsen noted that when the weather forecasting community started making weather forecasts halfway through the last century, they were initially highly inaccurate. However, they were still being produced and used in making decisions.  “In part, it was learning from the mistakes that caused these forecasts to become a much more reliable operational tool at a very rapid clip. I think the statistic is something like the five-day forecasts now are as accurate as the one-day forecasts 20 years ago. So we can’t wait until we reduce our uncertainty envelopes to a minimal level before we take action because that will stymie progress. We learned by doing.”

It’s the job of the scientist to communicate clearly not just what the probabilistic uncertainty is with the model outcomes, but what are confidence and process-level understanding that goes into that the model is,” Dr. Larsen continued.  “Then it’s the job of the manager to take that uncertainty and that level of confidence into account in weighing risk and making the decision.”

The conceptual model is step one,” said Dr. Healey.  “You can’t do the numerical model until you’ve got the conceptual model. And even in the absence of the numerical model, the conceptual model can be a great tool for trying to get everybody on the same page so that everyone understands what all of the actors are thinking about how the system works and why they think certain things are important.

We do create those conceptual models quite often,” said Dr. Acuna.  “It’s the next step – a lot of us would have doubts on whether the data is sufficient enough to create those numerical models, while others think that we should at least try. And we tend to get stuck there and just rely on the conceptual models instead.”

I would certainly encourage you to try building some numerical models, even if there’s a lot of uncertainty in what you think are your key parameters, and then play with it,” said Dr. Healey.

I would remind everyone that current law requires the Delta independent Science Board to provide oversight of the scientific research, monitoring, and assessment programs that support adaptive management of the Delta, through periodic reviews of each of these programs that shall be scheduled to ensure that all Delta scientific research, monitoring, and assessment are reviewed at least once every four years, and report that to the council,” said Phil Isenberg.  “It just seems to me if we’re having trouble focusing the discussion of how to define and how to evaluate … the Science Board could start their hearings and say, how are we doing on the big project? How are we doing on a small project? Part of the role of good science and good evaluation from the Science Board is hectoring people in a very polite but meaningful way to point out that doing nothing is probably not a good idea.”

Evolution of adaptive management

Participant question: Dr. Mike Healey mentioned that the different programs have evolved differently in terms of how they embraced adaptive management. Evolution is part of adaptation, so I’m curious as to what you saw and how those different programs evolved and embraced this concept.

The biggest evolution was that quite a few of these big projects gave up on adaptive management,” Dr. Healey said.  “Sometimes that was a political decision; sometimes it was an administrative decision. In a sense, they all struggled in the way that the Delta has struggled.  How do you implement this process within the existing agency structure? The ones that have carried on mostly developed a process early on; it wasn’t exactly the same as what’s been happening in the Delta. But you know, they were 180 degrees different. And then they stuck to the application with that process, which is what I see happening in California.”

It’s not a wasteland,” Dr. Healey continued.  “The ones that I have some familiarity with, which certainly isn’t all of them, I don’t see anywhere that’s doing a better job than you’re doing in the Delta. So I certainly wouldn’t say you should be discouraged. I think it’s really a slow process to try to insert a whole new way of doing things into a bunch of different government agencies. In fact, I can’t imagine anything harder. So the fact that things have gone as far as they have, still changing dramatically, I think is a good sign.”

Progress being made in modeling processes

Participant question:  In the Delta, for large-scale restoration projects, particularly tidal Marsh projects and floodplain restoration projects … I think we’re doing better than we used to.   We have a lot of hydrological models or hydraulic models that we’re pursuing and using those to plan our projects. And then, we can use those models afterward to evaluate our projects as well.  The second thing is that, in addition, we’re doing sediment modeling and also looking at the context of our projects in the entire Delta. So I feel like we have made some great strides in our modeling processes that support adaptive management.”

That’s great that you’re using numerical models for aspects of your programs,” said Dr. Healey.  “I probably didn’t explain as well. The kind of model that I’m thinking about is a management model, not a model of some physical or biological process. And those sub-models might be part of the overall model. But it’s really the idea of putting together a numerical model of the system that you can use to game management options.  I think that’s a rather different model than the ones that many scientists are used to.

What I’m trying to explain is the interaction between those numerical hydraulic models and management decisions, so we have used those to redesign our projects,” said the participant.  “Then, in the end, we can use the same models because we have options and sizes of weirs and sizes of whether or not to raise the elevation in some places or another. And so using those models can help us; those 2D models, basically game the system, and then we make changes in our design and then later to evaluate the design.”

How will we know if we’ve succeeded?

Dr. Healey had a question for Phil Isenberg: How will we know if adaptive management has succeeded? So since we’re talking about outcomes, what would tell you that an adaptive management program has succeeded? Whether it be anything like the stuff that I gave, or would it be totally different?

It wouldn’t be convincing to me if it were based on the notion that all the participants are happy with it, which I think is probably the most common way public policy is evaluated,” said Mr. Isenberg.  “I’m interested in inserting failures into the evaluation process, articulating the grounds for the failure, and making it as public as we can, along with successes where it seems to be working. And I’m also in favor of moving toward non-environmental projects for that evaluation as well. Doing what the Delta Stewardship Council staff did on the twin tunnels proposal was a foundation that science can build on, hoping that future proposals of similar scope will not fall into the same trap of a meaningless adaptive management program.”

I like to count things because I understand and have a measure against which I can make judgments,” Mr. Isenberg continued.  “I think you have to narrow the boundaries of the discussion, so uncertainty and certainty get closer together. And you at least have ranges of probability, which become the equivalent of a management tool, and thus, adaptive management at its best. And also, by refining the definition, you’re really forcing project administrators to fess up. But what bothers me most is they don’t want to be told what to do by anyone. And they need more money to do it anyway. And that attitude is destructive of any kind of effective adaptive management.”

I agree, and the notion of failure, which is very hard for most managers to acknowledge, can be a great learning opportunity,” said Dr. Healey.  “And adaptive management in fact intended to provoke failure from time to time in order to learn.”

Consider what we’re all going through on the Coronavirus and the epidemic,” said Mr. Isenberg.  “It is kind of a classic illustration of where public policy public opinion, misinformation, scientific evaluation, and chaos is going on. And in that kind of context, if it actually happens, if, you know, President Biden said, we will make mistakes, and we’ll tell you them. Now, if that happens. Whoo, I like that. That’s a lot.”

Click here for the 2021 Adaptive Management Forum Main Page.

Print Friendly, PDF & Email