Main content area

Modeling regional variation in net primary production of pinyon–juniper ecosystems

Huang, Cho-ying, Asner, Gregory P., Barger, Nichole N.
Ecological modelling 2012 v.227 pp. 82-92
arid lands, carbon, ecosystems, grazing, land tenure, models, plateaus, primary productivity, topography, vegetation cover, woodlands, Colorado
Spatial dynamics of carbon fluxes in dryland montane ecosystems are complicated and may be influenced by topographic conditions and land tenure. Here we employ a modified version of the Carnegie Ames Stanford Approach (CASA) ecosystem model to estimate annual net primary production (NPP) at a fine spatial resolution (30m) in pinyon–juniper (P–J) woodlands of the Colorado Plateau. NPP estimated by CASA was generally comparable to validation data from a statistical NPP model and field observations. We then compared modeled NPP results with spatial layers of topography and managed grazing to assess the influences of these factors on NPP. At the regional scale, there was a positive correlation between elevation and NPP (r²=0.20, p<0.0001), mainly due to an orographic effect, but slope and slope facing-derived dryness indices failed to explain modeled variation in NPP. Topographic analyses based on six terrain aspect classes showed that cooler and wetter north-facing slopes yielded higher NPP than did south-facing slopes. A multiple regression consisting of three numerical topographical attributes (elevation, slope and a dryness index) yielded the highest predictability for CASA NPP (adjusted r²=0.24, p<0.0001). Modeled NPP of a grazed site was significantly higher than that of an ungrazed site. Combining the results from this study and previous research efforts suggests that grazing may be correlated with higher woody vegetation cover, which elevates NPP in P–J woodlands of the Colorado Plateau. The findings reveal that the spatial pattern of NPP is complex, and can be strongly affected by topographic and/or anthropogenic factors even in relatively remote areas of this dryland region.