Final Report Summary: New York Aquatic GAP Pilot Project
Marcia S. Meixler and Mark B. Bain
New York Cooperative Fish and Wildlife Research Unit, Cornell University, Ithaca
A pilot GAP project for aquatic systems began in 1995 to define a methodology and determine the feasibility of predicting biodiversity distribution. Similar to gap analysis in terrestrial environments, gap analysis for aquatic systems uses remotely sensed data for habitat mapping, infers aquatic biodiversity distribution from habitat data, and provides large-scale information for targeting conservation measures. Our pilot project has been a low-level effort (e.g., a one-person project) for two years. We established methods for database development and GIS analyses using one river basin in western New York.
The original purposes of gap analysis (Scott et al. 1993) remained unchanged when applied to aquatic environments. However, the connected nature of aquatic habitats and the mobility of aquatic species complicate traditional gap analysis methods. Emphasis was placed on streams and rivers, as these waterbodies harbor a large majority of the freshwater biodiversity in the United States and are the focus of water quality assessments by management agencies. Methods were developed to reflect habitat status over a network of streams because aquatic species respond to cumulative effects due to the flowing nature of water in streams. A key habitat attribute influencing aquatic species is water quality, thus we developed a nonpoint-source pollution model that relies on land cover. This component of our aquatic GAP model integrates the terrestrial and aquatic parts of gap analysis. Finally, aquatic biodiversity conservation will likely focus on land management, not land acquisitions, since aquatic biodiversity is generally highest in large streams and rivers, making land acquisitions impractical. Again, the linking of terrestrial and aquatic GAP coverages is essential to address conservation issues.
The basic aquatic GAP model predicts relative levels of fish and macroinvertebrate diversity and identifies stream reaches having high biodiversity that are without management or protection. This was accomplished by classifying stream segments into habitat types using five attributes: stream size, habitat quality, water quality, stream gradient, and riparian forest cover. Stream segments were classified into one of eighteen habitat types for fish diversity predictions and one of eight habitat types for macroinvertebrate diversity predictions. Fish species and macroinvertebrate taxa were linked to habitat types using life history data. Maps and information on management and conservation areas were included in the GIS to locate unprotected stream segments with high diversity. As in other GAP projects, these are the "gaps" or areas where future conservation efforts should be focused or management practices altered.
Our aquatic GAP pilot was developed for the Allegheny River watershed of western New York. This region has a mix of forests (67%), agriculture (crop and dairy 28%), water and wetlands (3.4%), residential and urban areas (1.5%), and barren land (.1%). Aquatic habitats are largely comprised of small headwater streams (86% of the stream kilometers), with only 11% of the stream kilometers in large streams and small rivers and even fewer kilometers in large rivers. From our GAP model, 92% of the stream kilometers were modified or highly altered habitats leaving just 8% of the stream kilometers as potentially supporting the highest species diversity. The classification of stream segments into modified and highly altered categories was largely due to agricultural land use predominantly occurring in stream valleys and roads adjacent to streams. Approximately 79% of the stream kilometers were predicted to have good water quality using the nonpoint-source pollution model. When habitat status and water quality were combined, 913 (94%) of 980 stream segments were considered altered in some way. Thus, although degraded water quality was an important factor in the anticipated reduction of biodiversity for all sizes of streams, it was not nearly as dispersed or prevalent as physical habitat degradation. Overall high-quality stream segments were few and were well distributed across watersheds. These high-diversity habitats were the stream segments with high-quality water, intact channel habitat, a closed streamside canopy, and a high gradient. Good-quality streams were primarily headwaters (91%), with the remaining 9% comprised of large streams and small rivers.
The most diverse fish habitats were predicted to occur in large stream and small river segments with intact habitat quality and water quality suitable for life support. Only eight stream segments were identified in this class. Due to the large degree of human land use immediately adjacent to streams, there was an abundance of stream segments classified as modified and highly altered in habitat quality. This, in addition to the stream segments classified as degraded in water quality, greatly reduced the number of stream segments available for classification as highest in fish diversity. The most diverse macroinvertebrate habitats were predicted to occur in high-gradient, closed-canopy streams with good water quality (262 stream segments). Unlike in predictions for fish, the limiting factor in anticipated macroinvertebrate diversity was water quality degradation. Good water quality was predicted to be prevalent, especially in typically high-gradient headwaters, therefore there were many sites expected to have high macroinvertebrate diversity.
The goal of our pilot project was to demonstrate the feasibility and utility of the gap analysis methodology for predicting biodiversity distribution at the watershed scale. We illustrated this through the creation of a geographic information system model that classified stream habitats and related fish species and macroinvertebrate taxa to these habitats. The use of a land cover map in the prediction of water quality served as a link to gap analysis efforts in terrestrial systems. One major finding in this pilot project was the scarcity of stream segments with high predicted fish diversity, defined as large streams or small rivers with intact habitat and good water quality. GIS analyses showed that agricultural land use and roads were concentrated in midsized stream valleys where flat land along rivers was not vulnerable to destructive flooding. Although intuitive as an afterthought, the GAP model identified this pattern and clearly explained why so few quality midsized streams remain in what is largely a forested setting.
Another finding suggests that the existing conservation status may not actually afford significant protection. Of all stream segments, about half (48%) were under some form of protection by state parks, wildlife management areas, state-regulated wetlands, and water quality management classes A or AA. Despite the large percentage of streams in protected areas, the conservation-oriented management classes do not appear effective for aquatic biodiversity conservation. Many stream segments in protected areas were classified as poor quality. The protection of small land units did not substantially reduce the effects of runoff from agricultural lands or of alteration of streams along roads and farms. The utility of our GAP model in conservation planning was demonstrated by being practical to implement and capable of making predictions of the distribution of biodiversity and of management gaps.
Literature Cited
Scott, J.M., F. Davis, B. Csuti, R. Noss, B. Butterfield, C. Groves, H.Anderson, S. Caicco, F. D’Erchia, T.C. Edwards, Jr., J. Ulliman, and R.G. Wright. 1993. Gap Analysis: A geographic approach to protection of biological diversity. Wildlife Monographs 123:1-41.