April 28-29, 2001 (WSU - Vancouver, Student Services Center) Meeting Summary Meeting Minutes Meeting Agenda Presentations Attendees Meeting Summary SEI Science Panel Modeling Workshop, April 28-29, 2001 Main Points NMFS agreed that, assuming the project goes forward, use of an adaptive management approach is appropriate. The challenge remains of how to define that approach. In developing a conceptual model for estuarine conditions, the Science Panel in general felt that: The question of what constitutes an 'appropriate standard' (i.e., a 'recovery standard' or a 'no-net- loss of current conditions standard') is a policy decision beyond the scope of the panel's charge. The baseline for evaluating information should be in terms of incremental impacts to current conditions and not to the historical development of the estuary. Given the diversity and complexity of the life histories in the estuary, the Science Panel in general: Called for an emphasis on some of the more vulnerable species (e.g., ocean-type chinook and chum) as indicators of the impacts of changing estuarine conditions. Endorsed the approach used by Bottom et al., which argues for a simple management objective of maximizing the amount of diverse habitat. NMFS has identified five hydraulic parameters of concern: salinity, surface water elevation, depth, velocity, and temperature. The Science Panel in general felt that suspended sediments needed to be moved onto that priority list. In a discussion of the meaning and implications of an 'ecosystem approach,' NMFS concurred that the agency is looking for 'a finite list of a small number of factors to be evaluated and a recognition that they interact.' Regarding the WES salinity model used by the Corps, the Science Panel in general felt that: While some outstanding issues remain concerning the adequacy of the model, the model nevertheless provides valuable information on a range of physical factors from which inferences can be made about their effects on fish. Since the model makes the case that the project would have little impact on the channel, 'physical reasoning' dictates that even less would occur in the periphery. The model alone, however, does not necessarily translate into salmon habitat attributes, and therefore, more information on the periphery is needed. NMFS, however, must define a finite set of parameters of concern for the periphery. In addition to the WES salinity model, the Science Panel in general felt that: It should be compared with the Baptista model, using the same inputs. (NMFS stated that if no appreciable difference is discerned between the two models, that it would be comfortable moving forward with just one of the two models.). Other estuarine physical processes ¾ specifically, depth, temperature, and velocity ¾?should be modeled to assess the ecological effects of the project on salmonids. (The Corps, however, argued against modeling for temperature, stating that the effects of the project will most likely be a lowering of overall estuarine temperature, given that the project will allow for colder water to penetrate farther up into the estuary.). A worst-case scenario for low-flow conditions for the period August-September should also be modeled, given the potential gravity of the current water shortage. The Science Panel expressed interest in learning more about the area above the head of the salt wedge, the narrow part of the freshwater tidal estuary. Meeting Goals Provide an ecological framework for evaluating estuarine processes in the context of salmonid functional needs. Determine estuarine processes that should be modeled to assess the ecological effects of channel deepening on listed salmonids. Define the modeling requirements, model outputs, and magnitude of hydraulic/salinity changes that would be required to assess ecological effects on listed salmonids. Resolve the questions about salinity modeling and potential hydraulic/salinity impacts raised in the December 1999 NMFS Biological Opinion and NMFS Science Center review comments. Goal Attainment The group made progress during this meeting on developing an ecological framework. Essentially, the framework is a matrix that fleshes out the relationship between various physical parameters and their potential effects on salmonids and salmonid habitat. Filling in the individual cells of the matrix is an ongoing process that will be further refined in future meetings. In addition to salinity (already modeled), the Science Panel felt the following physical parameters, or estuarine processes, should be modeled to assess the ecological effects of channel deepening on listed salmonids: temperature depth velocity The Science Panel charged the modelers with the task of determining prior to the May meeting the feasibility of modeling the additional physical parameters of temperature, velocity, and depth. Panel members also expressed interest in modeling a worst-case scenario of low-flow conditions for the period August-September, given the potential gravity of the current water shortage. In general, the panel felt that the Corps salinity intrusion model provides valuable information on a range of physical factors from which inferences can be made about the effects on fish. Further, they felt that, assuming the model structure is adequate, it makes the case that salinity intrusion is not an issue in the channel, and therefore, is not in the periphery, as well. However, they did feel that some issues remain to be resolved about confidence in the model structure. They encouraged a comparison of the WES model with Antonio Baptista's model, using the same inputs. Meeting Highlights Uncertainty and Risk Analysis NMFS agreed that, assuming the project goes forward, use of an adaptive management approach is appropriate. The question remains, however, how to define that approach. The presenter, Anne Fairbrother, said that regardless of the project decision, adaptive management should be used; a decision not to go ahead with the project is still a decision, and therefore, requires an adaptive management approach. There was lengthy discussion about the irreversibility of the project, but Karl Eriksen of the Corps pointed out that the locations the Corps proposes to dredge and deepen "are primarily the same locations where we continually do maintenance dredging on a regular basis. So if an impact is identified, the decision is not irreversible. Many of the areas would fill back in to the current channel in just a few years. Conceptual Model Discussion centered on what constitutes an appropriate standard ¾ a recovery standard or a no net loss of current conditions standard. Steven Courtney of SEI said the group's charge is to evaluate potential impacts on current conditions and future goals; the issue of the appropriate standard will need to be addressed by others. On the discussion of the baseline for evaluation, the panelists felt that NMFS was using too broad a scale by which to evaluate information. Panelists felt the baseline should be evaluating information in terms of incremental impacts to current conditions and not to the historical development of the estuarine system. Overview of Salmonid Estuarine Ecology A general discussion ensued on the high degree of uncertainty about how an individual fish will respond to changing conditions in the estuary. The panel suggested stochastic modeling as an approach to help reduce uncertainty. Panelists agreed that it is a difficult choice to concentrate research on just one or two species (in this case, ocean-type chinook and chum). However, they concurred that the diversity and complexity of the life histories in the estuary call for emphasis on some of the more vulnerable species as indicators of the impacts of changing estuarine conditions. Given the complexity of the system and the variety and diversity of ESU's (Evolutionary Significant Units), the panel agreed that it would be "a fool's errand" to try to develop a management matrix. Instead, the panel liked the approach used by Bottom et al. which, in acknowledging the complexity and diversity of the estuary, argues for a simple management objective of maximizing the amount of diverse habitat. Overview of Estuarine Physical Processes The panel expressed interest in understanding the biological foundations of sediment and sediment transport. How would dredging, for example, affect suspended sediments and could this be modeled? Antonio Baptista of the Oregon Graduate Institute felt that modeling sediments is not practical. The conversation then turned to the potential effects of dredging on the location of the ETM (Estuarine Turbidity Maximum) and the characteristics of it. In looking at the modeling input-output table, NMFS identified five hydraulic parameters of concern: salinity, surface water elevation, depth, velocity, and temperature. Panelists felt that suspended sediments needed to be moved into that priority list, given that the estuary is a turbid system and the presence of suspended sediment reduces nutrition. The question of the extent to which changing channel morphology will affect suspended sediment distribution was deferred to the June sediment meeting. Overview of Corps Modeling (Corps presentation) The focus of discussion was on the historical changing morphology of the channel and the potential impacts of dredging on that change. Since the north channel has been gradually filling in, does that mean there is greater penetration of salt water up the south channel, and how will dredging affect that? Resolution of the question was unclear. Antonio Baptista of OGI pointed out that the Corps' study is "a preliminary study with preliminary resolution in the estuary." He recommended looking at actual bathymetry, which he would do in his presentation, scheduled for the following day. Overview of Corps Modeling (NMFS presentation) Panelists wanted to know what NMFS meant by an "ecosystem approach." Much concern was expressed too broad an approach could be intractable. Ed Casillas of NMFS explained that for the agency to really grasp the biological consequences of the project, more than one variable needed to be looked at and in an integrated manner. Steven Courtney of SEI clarified, and NMFS concurred, that the agency is looking for "a finite list of a small number of factors to be evaluated and a recognition that they interact" ¾ that they are not entirely independent of each other. Quite an involved discussion ensued on NMFS request for more information on what is happening in the estuary periphery where the fish tend to reside. The panel felt that since the Corps model made the case that not much would happen in the channel, "physical reasoning" dictates that even less would occur in the periphery. Casillas replied that NMFS wants a "justification" that the Corps model "adequately resolves what's going on in the periphery." The panel pressed the point that the model more than adequately makes the case, and that therefore, nothing remains to be refined with regard to the periphery. Casillas raised the need for the agency to satisfy the public, which set off a heated debate about what drives this process ¾ public opinion or science. Steven Courtney of SEI reminded the panel and the audience that the charge of the panel is to determine its level of confidence in the science and not to concern itself with how the public will react. The panel wanted to put to rest the debate over the Corps salinity model. Assuming the model structure is sound, the panel felt it adequately addresses the effects of salinity intrusion in both the channel and in the periphery. Further, the model is a valuable indicator of other effects; one can infer from the model, for example, that velocities, depth, and temperature would similarly not be greatly altered by the project. The panel agreed, however, that the model alone does not necessarily translate into salmon habitat attributes. For that reason, some panelists felt that more specific information is needed on the periphery. The panel was clear, however, that NMFS needs to define a finite set of parameters of concern for the periphery. One panel member raised the point that acceptance of the adequacy of the model is not universal; that outstanding issues remain concerning numerical mixing of algorithms, potential problems with different bathymetry data, and limited horizontal and vertical resolution. The panel expressed interest in learning more about the area above the head of the salt wedge, the narrow part of the freshwater tidal estuary. There, alteration of the channel might be significant in terms of affecting velocities and water level, especially in the shallow-water periphery areas. In view of the above discussion, Steven Courtney of SEI asked the panel whether it would be useful to have a broader range of conditions modeled, and if so, what would they be? The panel felt that the model should be run for extremely low-flow conditions to address a worst-case scenario, like a drought or an energy shortage. Courtney summarized that the panel gave a qualified 'yes' to looking at some broader conditions, but not every possible condition. "In fact, a very specified condition ¾ that is, low flow." Alternative Approach for Modeling The panel was greatly impressed by Baptista's model and presentation. Most of the discussion centered around the utility of modeling for low-flow conditions, given that Baptista's model shows that historically low-flow habitat opportunities have not been altered very much. Much concern was expressed, however, that the Columbia is already at low-flow conditions, so conceivably, things could be worse by August-September. Discussion of Modeling Outputs for Biological Parameters NMFS argued the need for more spatial resolution and greater emphasis on peripheral areas, not just on Cathlamet Bay (an important transition zone). The panel expressed frustration, however, that further spatial resolution is pointless if it can't be related to salmon habitat. The precision of physical models does not match up well with the imprecision of biology. While the panel felt that the Corps' salinity intrusion model is adequate in and of itself and serves as a useful indicator, there was general consensus that other physical factors (e.g., temperature, velocity, and depth) with known ecological importance could be modeled with some degree of reliability. If no appreciable physical changes are discerned then, as with the salinity model results, one could infer that there would be little to no impact on salmon habitat. The panel felt that it would be useful to compare the WES model with Baptista's model, using the same inputs. If no appreciable difference is discerned between the two models, NMFS stated that it would be comfortable moving forward with just one of the models. A long discussion ensued on filling out the matrix parameters, beginning with a discussion of temperature. The Corps argued that the project will most likely allow colder water to penetrate farther into the estuary with salinity intrusion in the main channel, with the effect of lowering temperature. Therefore, the Corps felt that modeling for temperature is pointless. NMFS, however, felt that it is still important to look at temperature; if it is not too difficult to look at it, then NMFS would prefer that temperature be modeled. The same was said of modeling for velocity. The Corps then expressed concern about putting absolute criteria on the various parameters; that that would yield precision without accuracy. The Corps also expressed concern about the cost and time required to run these new parameters. After some private caucusing with the project managers, Steven Courtney tasked the modelers with collaborating on developing a statement of feasibility prior to the May meeting. The project managers would then develop a proposal about what further modeling they are willing and able to support.