Generalized approach to landscape-wide seasonal forecasting with temperature-driven simulation models
1996
Regniere, J.
Determining and depicting the landscape-wide outcome of temperature-dependent ecological processes is important in the study of spatial patterns in insect ecology as well as in the efficient deployment of human and material resources for pest management. A generalized approach is proposed that consists of a temperature-driven simulation model that uses data from a limited number of strategically located points in the landscape as its input. Elevation, slope, and aspect are varied systematically for each point. Air temperature data are obtained as both real-time and climatic average data, matched with simulation points on the basis of distance, elevation, and climatic zones. Differences in latitude and elevation are compensated for with lapse rate transformations. Slope and aspect are combined into a single exposure expression, and overheating is calculated on the basis of incoming solar radiation. Landscape-wide projections of features in model output are obtained by algebraic transformation of a digital elevation model of the landscape, using a multiple regression equation relating the output feature to latitude, longitude, elevation, and exposure. The various sources of error involved in seasonal forecasting in insect ecology and management, as well as an outline of the procedure required to test the validity of landscape-wide projections, are discussed through a case study of the phenology of the spruce budworm, Choristoneura fumiferana (Clemens).
显示更多 [+] 显示较少 [-]