Understanding the way regional landscapes operate, evolve, and change is a key area of research for ecosystem science. It is also essential to support the “place‐based” management approach being advocated by the U.S. Environmental Protection Agency and other management agencies. We developed a spatially explicit, process‐based model of the 2352 km2 Patuxent River watershed in Maryland to integrate data and knowledge over several spatial, temporal, and complexity scales, and to serve as an aid to regional management. In particular, the model addresses the effects of both the magnitude and spatial patterns of human settlements and agricultural practices on hydrology, plant productivity, and nutrient cycling in the landscape. The spatial resolution is variable, with a maximum of 200 × 200 m to allow adequate depiction of the pattern of ecosystems and human settlement on the landscape. The temporal resolution is different for various components of the model, ranging from hourly time steps in the hydrologic sector to yearly time steps in the economic land‐use transition module. We used a modular, multiscale approach to calibrate and test the model. Model results show good agreement with data for several components of the model at several scales. A range of scenarios with the calibrated model shows the implications of past and alternative future land‐use patterns and policies. We analyzed 18 scenarios including: (1) historical land‐use in 1650, 1850, 1950, 1972, 1990, and 1997; (2) a “buildout” scenario based on fully developing all the land currently zoned for development; (3) four future development patterns based on an empirical economic land‐use conversion model; (4) agricultural “best management practices” that lower fertilizer application; (5) four “replacement” scenarios of land‐use change to analyze the relative contributions of agriculture and urban land uses; and (6) two “clustering” scenarios with significantly more and less clustered residential development than the current pattern. Results indicate the complex nature of the landscape response and the need for spatially explicit modeling.

We analyzed the relationship between resolution and predictability and found that while increasing resolution provides more descriptive information about the patterns in data, it also increases the difficulty of accurately modeling those patterns. There are limits to the predictability of natural phenomenon at particular resolutions, and "fractal-like" rules determine how both "data" and "model" predictability change with resolution.We analyzed land use data by resampling map data sets at several different spatial resolutions and measuring predictability at each. Spatial auto-predictability (Pa) is the reduction in uncertainty about the state of a pixel in a scene given knowledge of the state of adjacent pixels in that scene, and spatial cross-predictability (Pc) is the reduction in uncertainty about the state of a pixel in a scene given knowledge of the state of corresponding pixels in other scenes. Pa is a measure of the internal pattern in the data while Pc is a measure of the ability of some other "model" to represent that pattern. We found a strong linear relationship between the log of Pa and the log of resolution (measured as the number of pixels per square kilometer). This fractal-like characteristic of "self-similarity" with decreasing resolution implies that predictability may be best described using a unitless dimension that summarizes how it changes with resolution. While Pa generally increases with increasing resolution (because more information is being included), Pc generally falls or remains stable (because it is easier to model aggregate results than fine grain ones). Thus one can define an "optimal" resolution for a particular modeling problem that balances the benefit in terms of increasing data predictability (Pa) as one increases resolution, with the cost of decreasing model predictability (Pc)"

No abstract

No abstract

No abstract