The tree-ring interpolation model (TRIM) and its application to Pinus monophylla chronologies in the Great Basin of North America

摘要

Tree-ring records are routinely used in applied environmental science as indicators of long-term changes at annual-to-seasonal time steps. Since natural resource managers, modellers and researchers benefit from the use of spatially interpolated products over large areas, there is a need to convert point tree-ring chronologies into gridded fields. Our tree-ring interpolation model (TRIM) estimates an index of tree radial growth at unsampled locations based on available tree-ring chronologies and site-specific topographic variables. TRIM follows an approach similar to the parameter-elevation regressions on independent slopes model (PRISM) for interpolating climate data, as it is based on simple linear regression between tree-ring data and elevation. Topographic features are defined starting from a digital elevation model with 30-m grid cells, and model output is currently available at 4-km grid spacing. Tree-ring chronologies used for predicting each grid cell value are selected according to linear distance (<200 km), slope aspect (in the same 90°-class) and steepness (within 2"). Additional model requirements can define elevation boundaries for the prediction and impose limits on the estimated values. We present here the details of the model and its application to Pinus monophyila Torr. & Frem. in the Great Basin of North America. In addition, we show how TRIM output can easily be compared with PRISM climate data and derived drought indices. Such tree-ring field reconstructions allow for a spatial visualization of ecohydroclimatic heterogeneity over time, which is increasingly being requested for watershed-scale management of natural and water resources.