Integrating Information at State, Regional, and Local Scales: An Essential Step for Watershed Management, Restoration, and Monitoring

---

Watersheds are landscape mosaics; therefore, watershed structure andfunction is dependent on scale. Temporal and spatial scales influence theinferences we can make about landscape patterns and processes. Spatial scaleis the dimension of an object or process characterized by both grain andextent. Grain is defined as the finest level of spatial resolution possiblewith a given data set and extent is the size of a study area. The scaleat which watershed measurements are taken influences our ability to detectspatial patterns. Biotic and abiotic processes vary in their operating scale.For example, anadromous fish are affected by stream and ocean environments.In contrast, native minnows are influenced by processes that occur withina stream or tributary.

People's objectives set the working scale of a project, which in turninfluences the type of digital data that will be useful to meet the objectivesof the project. For example, the state water quality control board is interestedin hydrological data statewide. Though land owners may work at the locallevel to stabilize stream banks to prevent erosion, geographic informationsystems (GIS), a computer mapping and analysis tool, can help integratethe physical and biological information from various scales that is relevantto watershed management, restoration, and monitoring. Digital geographicdata used in assembling a GIS come in a variety of accuracies and scales,ranging from aerial photography at 1:200 to satellite imagery at 1:1,000,000.Clearly, there is a role for data at all scales. Currently, however, digitaldata is almost exclusively restricted to the global scale (1:80,000 - 1:1,000,000),leaving many local projects without large-scale, local data.

A Case Study: Parsons Creek, A Russian River Tributary

Parson's creek is a minor tributary of the Russian River, which runsthrough the University of California's Hopland Research and Extension Center.The creek becomes almost dry in summer and can exceed several hundred cubicfeet per second in winter. Historically, the creek supported a significantsteelhead population. However, today the population is greatly reduced.The creek is affected by grazing, roads, and development. Gravel removaland channel alteration was conducted in the 1950s and the creek was severelyaltered by a 1964 flood. Because of these human-induced and natural disturbances,the creek channel has eroded deeper and wider and the riparian vegetationhas been severely damaged. Continued grazing, browsing, and seasonal movementof bedload have prevented plant regeneration. This habitat alteration istypical of many tributaries of the Russian River.

Hopland Research and Extension Center Staff began the Parson's CreekProject to protect and restore habitat for anadramous fishes along a portionof the Creek. The long-term goal is to increase streambed stability andshade canopy throughout the watershed. This will require cooperation amongseveral landowners. We are using the Parson's Creek watershed as a casestudy of the use of GIS data for studying watersheds to assist the landownersin prioritizing and monitoring their efforts. We have identified data frommany sources, including hydrology, pasture boundaries, partial stream datafrom Global Positioning System (GPS), elevation, Klamath Bioregion vegetationclassification, and California Department of Forestry's hardwood pixel data.These layers provided the opportunity to correlate mapped vegetation typeswith vegetation data in the field. The classified satellite imagery is toosmall in scale to meet the objectives of the project. Large-scale aerialphotography, combined with ground-truthing, can provide better informationon riparian vegetation than small scale satellite imagery.

Hydrology Digital Line Graphs do not match well with the actual locationof Parson's Creek as measured using GPS. In some locations, the two streamlines are as much as 63 meters apart. This is a problem of scale, becausethe GPS data has a rated accuracy of approximately 3 to 5 meters, whereasthe hydrology data was digitized for maps intended to be reproduced at a1:100,000 scale. This discrepancy illustrates the inherent difficultiesof integrating small, linear features such as streams and riparian areasin a GIS. In fact, we found that walking the stream with a GPS unit is themost accurate way to map the stream channel.

Conclusions

GIS provide a format for integrating data that has been mapped at differentscales and varying resolutions. However, numerous circumstances still preventavailable digital data from being useful. These circumstances include technicaldifficulties inherent in integrating data from different sources and extrapolatingfrom data with a given grain and extent to another area with different dimensions.The solution to these problems often requires establishing uniform standardsfor future digital data sets and extensive documentation of the meta-data.Adopting this policy will help with some of the technical difficulties ofintegrating data, but this will not solve the entire problem.

Economic, political, and bureaucratic problems prevent access to andintegration of local, regional, and statewide efforts. Theoretically, large-scaleand local GIS efforts could be linked to region-wide coverages, therebypermitting the extrapolation of information from among small watershed projects.Ideally, small- scale data should be used to prioritize local watershedconservation and restoration efforts. These projects should then have accessto large-scale information. Access to both scales of data can make a differencein watershed planning, project implementation, and monitoring.


prepared and edited by Richard B. Standiford and Pamela Tinnin