Adapting NCWAP's Tool for Other Regions

Repeatability and flexibility are desirable qualities of a good spatial decision support system (SDSS) (for more information, see the Introduction to Decision Support Systems section). Once a sound assessment framework has been conceptually designed, managers and planners dealing with the same, or similar, management issues can benefit if the modeling approach is easily modifiable and/or transferable to other study areas. Although some adjustments to the SDSS may be necessary to accommodate differences in landscape features, ecosystem processes, or management regimes at the new locale, the basic logic, science, and technology of an established SDSS can ideally be carried over. Having the ability to apply a uniform assessment structure to multiple locations (yet at a spatially relevant scale) can be instrumental for regional planning and management. But this does not imply uniform management action. Recall that decision support systems do not yield explicit courses of action for a management question; they can uniformly and efficiently process, summarize, or rank data, but the results need to be interpreted by the decision-maker, and often augmented with other knowledge and information, to form an appropriate management action.

Since salmonid recovery is a widespread concern in the Pacific Northwest, it is not surprising that resource managers and planners outside of Northern California are interested in the Ecosystem Management Decision Support (EMDS) system and the modeling approaches developed by the North Coast Watershed Assessment Program (NCWAP) for assessing suitable salmonid conditions (see the About the EMDS System section for more information). For example, the US Department of Agriculture Forestry Service is using the EMDS system in a similar capacity “to characterize the ecological condition of watersheds and aquatic ecosystems” as part of the Northwest Forest Plan (Reeves, et al. 2001). In San Mateo and Santa Cruz Counties of Central California, environmental conditions and target species for recovery differ somewhat from Northern California, but the limiting factors assessed by NCWAP’s knowledge base models are highly applicable. This is because expert opinion has played a critical role in developing the knowledge bases and identifying the key concerns, even though the science and metrics for some of the criteria are incomplete. For example, experts generally agree that large woody debris (LWD) is an important factor for salmonid survival and should be considered in an assessment, but there is uncertainty as to the best way to measure LWD and as to what levels or amounts of LWD are suitable and unsuitable (that is, threshold values) for salmonids. (LWD, it should be noted, is important in that it provides the stream channel complexity essential to salmonid rearing in this region, but in other regions, stream complexity can be provided by boulders, streambed rock outcrops, and so forth.)

It is also beneficial to apply the EMDS system and NCWAP’s knowledge bases in other parts of California where endangered or threatened salmonids run because it can help support regional planning for salmonid recovery. The National Oceanic and Atmospheric Administration National Marine Fisheries Service (NOAA Fisheries) is responsible for guiding recovery planning for Pacific Coast salmonids. They have organized their recovery planning efforts into a series of discrete geographic areas, or domains, with the intention to develop a recovery plan for each domain and the salmonid populations within them (NOAA Fisheries 2000). Four recovery domains are located either entirely or partially in the state of California, encompassing multiple counties and watersheds (see the NOAA Fisheries Recovery Planning section of this Web site for more information). Therefore, careful planning is needed to develop and coordinate comprehensive and focused recovery efforts. The EMDS system and NCWAP’s peer-reviewed knowledge bases can provide NOAA Fisheries with a basic yet flexible watershed assessment tool to support uniform decision-making in and across recovery domains.

However, there are several challenges to ready implementation of EMDS and NCWAP’s knowledge bases for assessing habitat suitability for salmonids outside of Northern California. Although the EMDS system is fairly simple to run with geographic information systems (GIS) software, NCWAP’s knowledge bases are complex and data-rich, as are most models that deal with ecosystem functions or processes. This is also the result of the rigorous peer review of NCWAP’s knowledge bases, where scientific experts weighed in heavily on the logic and criteria to improve the integrity of them. This is especially true for the Potential Sediment Production Model. It requires specific data sets on a watershed basis, such as locations of historical logging. Much of the data must also undergo substantial pre-processing, such as calculations of road density according to hillslope categories, before running the EMDS system.

Obtaining appropriate data sets for multiple watersheds of a basin can also be challenging. When attempting to apply the Potential Sedimentation Production Model in San Mateo or Santa Cruz Counties, sufficient, publicly available geological and hydrological data to assess sedimentation from natural processes could not be obtained for a watershed. Historical logging data were not readily available either. Together, these missing data sources make it impossible to effectively run the model in these counties as of June 2004. A further challenge is that if data are not available for all planning watersheds in a basin, it is not clear how many watersheds are needed for a valid assessment. Since NCWAP’s knowledge bases are presently designed to rank suitability among watersheds (as opposed to absolute measures of suitability, for which criteria are lacking for many parameters), clearly more than two watersheds are necessary. Additional testing and validation of the knowledge bases is necessary to determine the minimum number of watersheds needed to have confidence in model outputs. However, the wider the region over which NCWAP’s knowledge bases and the same data collection techniques are applied, the more useful it becomes as a tool to compare between watersheds. In that case, the dynamic ranges of the custom parameters of the knowledge bases get better defined, and one can better determine what constitutes good as opposed to bad conditions for salmonids (Rich Walker, personal communication 2004).

Specific criteria of the knowledge bases or data may also need to be modified for different locations, such as changes in the threshold values or in the weighting factors used for each criteria. For example, in Central California, stream temperature would still be important criteria in the Water Quality Model, but the threshold values (temperatures considered suitable for sustaining healthy salmonids) would probably be slightly different than in Northern California. Such data changes would need to be made. In addition, if one wanted to change the architecture of the knowledge base, such as the way in which parameters are combined at different levels of an evaluation, Netweaver™ software would be required to alter the knowledge base. For example, “Extensive Land Use” (defined in NCWAP’s Potential Sediment Production Model as livestock grazing) is not commonly found in most watersheds of San Mateo and Santa Cruz Counties, so one might want to alter this parameter and its relationships in the knowledge base under “Land Use Related Erosion.” Likewise, if there was a unique land use or environmental condition occurring in Central California that was not already reflected in the knowledge base, Netweaver would need be employed.

These challenges in implementing NCWAP’s knowledge bases in other regions should not deter users from exploring them. Even if the models are not run with the EMDS system, the logic and criteria proposed for the knowledge bases can be helpful in understanding limiting factors analysis for salmonids. Managers involved with recovery planning of Pacific Coast salmon can benefit from the steps NCWAP has taken to begin assessing the primary limiting factors of salmonids through a standardized, rule-based modeling system. NCWAP’s knowledge bases for the EMDS system should not be the only resource used to assess habitat suitability and support watershed-level planning and management, but it can be a valuable addition to a manager’s toolbox.

References

Reeves, G.H., D.B. Hohler, D.P. Larsen, D.E. Busch, K. Kratz, K. Reynolds, K.F. Stein, T. Atzet, P. Hays, and M. Tehan. 2001. "Aquatic and Riparian Effectiveness Monitoring Plan for the Northwest Forest Plan." USDA Forest Service, Pacific Northwest Research Station. Draft Watershed Monitoring Plan. View on-line document.

NOAA Fisheries. 2000. Recovery Planning for West Coast Salmon [Web page]. NOAA Fisheries [cited November 24, 2003]. View on-line source.