Fish & wildlife Subbasin planning Guides/forms

Out of subbasin survival factors in Ecosystem Diagnosis and Treatment

Mobrand Biometrics, Inc.   |   October 9, 2003

Related link: Oregon-specific Out-of-subbasin effects guide

Introduction

Many subbasin planners have elected to use Ecosystem Diagnosis and Treatment (EDT) as a primary assessment tool for aquatic habitats.  The EDT assessment of aquatic habitat is based on construction of life history trajectories that begin and end with spawning at particular points within a subbasin at specific times of the year.  EDT estimates survival and capacity of a focal species (e.g., spring chinook salmon) within a defined study area (e.g., a subbasin) based on habitat characteristics and combines this with predefined survival rates outside the study area.  These survival rates are an indication of the stresses placed on migrating fish as they navigate their way through the Columbia and Snake mainstems, the estuary, and the ocean.  Collectively, these stresses have been termed ?Out of Subbasin Effects? or OOSE. 

The Council's Technical Guide for Subbasin Planners identifies the need to consider out of subbasin effects for each focal species.  The Independent Science Review Panel (ISRP) has indicated that out of subbasin effects will be one of the elements that will be considered in its review of subbasin plans.  As a contribution to the need to supply subbasin planners with a set of assumptions regarding the out of subbasin effects, we are providing here the assumptions that are currently incorporated in the Ecosystem Diagnosis and Treatment model that is being used by subbasin planners. 

Methods

The out of subbasin assumptions currently used in EDT were developed as part of the Council's Multi-species Framework Project.  Calculations behind the results provided here were documented in the final project report to the Council from Mobrand Biometrics and in Marcot and others (2002).  The Framework assumptions were intended to capture conditions prevailing in the region around the year 2000.  The current out of subbasin assumptions in EDT are based on passage and hydrologic modeling done by the Council and the National Marine Fisheries Service.

The OOSE are defined for this memo as the total survival rate of juvenile fish from the mouth of the subbasin to their return to the subbasin as adults.  OOSE accounts for survival conditions through the hydroelectric system, the Columbia River below Bonneville Dam, the estuary, the ocean and any harvest occurring outside the subbasin.  To be specific, OOSE = Survival through the hydro system X survival in the lower Columbia River X survival through the estuary X survival in the ocean X overall harvest rate.  For subbasins below Bonneville Dam the first term is omitted.  This definition of the OOSE makes it equivalent to the smolt to adult survival rate or SAR that has been used in other modeling efforts.  SAR is typically depicted as a number between zero (no survival) and one (100% survival).  

The SAR is specific for a species and is related to the position of the subbasin within the Columbia Basin and especially relative to its position within the hydroelectric system.  In other words, because the SAR is affected by survival through the hydroelectric system (see equation above), the SAR is affected by the number of dams that fish must traverse to get to and from the subbasin.  As a result, we see SARs generally decline going upstream in the Columbia River.

Because the out of subbasin assumptions reduce to the SARs that result from the model, we have represented the combined effect of all current OOSE assumptions in EDT as the SARs for spring and fall chinook salmon projected from various strategic points in the Columbia Basin (Table 1).  In most cases these points are major dams, as these are the points where (1) significant changes in SAR can be anticipated, and (2) data are most readily available. The three exceptions are (1) the lower Columbia, which begins at the mouth of the Columbia River, (2) the Hanford Reach of the Columbia, which begins at the confluence with the Snake, and (3) the Lower Snake, which begins at this same point.  These SARs include all considerations for dam passage, survival below Bonneville Dam, survival through the Columbia estuary and the ocean and assumed harvest outside the subbasin.  The hope is that by focusing on the SARs (which incorporate all forms of out of subbasin mortality and can be related to empirical survival estimates), the region can avoid becoming embroiled in debates over details of individual survival components as part of the subbasin planning process.  This is consistent with direction provided by the Council in previous reports on the Out of Subbasin Effects issue. 

Results

The results in Table 1 are provided to clarify the assumptions that are available to subbasin planners regarding the SARs in EDT.  SAR has been estimated from empirical data in a few subbasins in the PATH process and elsewhere.  We have compared the estimated SARs in EDT to available empirical estimates of SARs and find them generally in agreement.  However, if managers and planners feel that other SAR assumptions are more appropriate for subbasin planning, the assumptions in EDT can be modified. 

Expl. = exploitation rate, approximately equivalent to harvest rate

Table 1.  Smolt to adult survival rates (SAR) for spring and fall chinook currently used in the Ecosystem Diagnosis and Treatment model.

The results in Table 1 approximate the survival rates that would be applied to spring and fall chinook entering the Columbia River or Snake River at the points in the table.  For example, spring chinook entering the Snake River at the head of Lower Granite Pool would be subject to an SAR of 0.9 percent in EDT.  This SAR incorporates an assumed harvest on spring chinook of 6.8 percent.  The SAR for the Lower Columbia represents survival of fish entering just below Bonneville Dam.  The total SAR that is actually applied to each population may vary slightly from these rates.  For example, if the subbasin enters at the midpoint of a reservoir, the population will not receive the mortality associated with the entire pool but will receive a mortality rate adjusted for the travel speed through the shorter distance.  The SARs for fall chinook represent survival of actively migrating juveniles.  Because fall chinook also include a component of fish that rear for some period within the mainstem Columbia and Snake rivers, total survival of fall chinook from each point may differ from the results in Table 1.

The SARs in Table 1 represent survival under ?typical? conditions in the Columbia River and the ocean.  EDT characterizes a single habitat configuration, i.e. annual time series data is not used.  Empirical estimates of SAR that have been reported in the PATH process and elsewhere vary widely between years reflecting environmental variation including regime shifts in ocean survival conditions.  However, the EDT assessment is intended to characterize the potential of current habitat in a subbasin with respect to a focal species and does not include environmental variability.

Relationship to Subbasin Planning

Table 2 relates various subbasins to the measurement points (dams) in Table 1 to give planners a general indication of the survival rate that would be applied to their subbasins if analyzed in EDT.  It should be stressed that these are approximate survival levels and that the actual survival rates that any group of fish experiences in the model may be slightly different than these.  The differences arise from two factors: First, subbasins enter the Columbia at various points along a reservoir and may experience different survivals.  This is because survival is roughly equivalent to distance traveled.  For example, the survival applied to fish from the John Day River, which enters just above John Day Dam and the Umatilla River, which enters at the head of John Day Reservoir, will be somewhat different reflecting the length of the reservoir.  Secondly, the out of subbasin survival is dependent on population specific factors, especially the time of entry into the Columbia River.  Depending on how the planners describe their populations and conditions in their subbasins, fish may enter the river through the model across slightly different weeks and experience slightly different conditions in the model that would be reflected in the actual survival rates. Hence, the values in Table 1 provide a general indication of survival associated with the subbasins in Table 2, but not an exact value that is applied.

Table 2. Subbasins associated with each SAR point in Table 1.

Cautions

Any effort to characterize out of subbasin survival is fraught with risk.  Data are typically not available at strategic locations.  This is especially so with the estuary and ocean.  The lack of data is accentuated by the wide range of natural variability over time.  Estimates of anthropogenic effects and variability associated with anthropogenic effects are often fraught with controversy.  This is particularly so given the wide variety of interests with a stake in these measurements and the tendency of these interests to interpret applicable data differently.  As suggested in Marcot and others (2002), how one interprets out of subbasin survival data often depends on one's ?world view,? reflecting a host of social and scientific factors.

For these reasons, it is important to remember that the purpose of subbasin planning is to analyze conditions in each subbasin and not to resolve the long standing technical and policy debates associated with survival in the mainstem rivers and ocean.  The primary need in subbasin planning is for out of subbasin assumptions to provide a realistic context for consideration of subbasin conditions. The values in Table 1 should be reviewed for their utility to provide that context, and suggested changes should have the purpose of improving our description of the context for subbasin planning.  

The information provided in Table 1 is presented to provide subbasin planners with an indication of the numbers that are embedded in the EDT model.  These numbers may also be helpful in describing out of subbasin effects in the subbasin plan narrative.  In any such narrative it is important to indicate that these numbers are approximations and to note that, while potentially useful for subbasin planning purposes, it would be folly to use these numbers to draw definite conclusions regarding specific impacts or to make decisions on specific restoration projects.

Attachment 1:  Dam survival assumed as part of the SAR in EDT.

The tables below from Marcot and others (2002) provide the schedule of survival rates at each dam for each month of the year for spring and fall chinook salmon.  In EDT, fish leave the subbasin and enter the mainstem across a range of months.  They move down at travel speeds related to flow encountering daily survival rates in the reservoirs.  Fish are then passed through a dam where they encounter the survival rates in the tables below.  A portion of the fish may be transported downstream.  The dam survival rates below were calculated using the National Marine Fisheries Service SimPass model with conditions specified in the Biological Opinion prevailing in 2000.  Other mainstem passage survival assumptions are described in Marcot and others (2002).

Yearlings chinook dam survival rates currently used in EDT

Subyearlings chinook dam survival assumptions used in EDT

Reference cited

Marcot, B. G., W. E. McConnaha, P. H. Whitney, T. A. O'Neil, P. J. Paquet, L. E. Mobrand, G. R. Blair, L. C. Lestelle, K. M. Malone, and K. I. Jenkins. 2002. A multi-species framework approach to the Columbia River Basin. Northwest Power Planning Council, Portland, OR www.edthome.org/framework

^ top