CONICET | Buscador de Institutos y Recursos Humanos

Monitoring ofaboveground net primary production (ANPP) is critical for effectivemanagement of rangeland ecosystems but is problematic due to the vastextent of rangelands globally, and the high costs of ground-basedmeasurements. Remote sensing of absorbed photosynthetically activeradiation (APAR) can be used to predict ANPP, potentially offering analternative means of quantifying ANPP at both high temporal andspatial resolution across broad spatial extents. The relationshipbetween ANPP and APAR has often been quantified based on eitherspatial variation across a broad region or temporal variation at alocation over time, but rarely both. Here we assess: (i) if therelationship between ANPP and APAR is consistent when evaluatedacross time and space; (ii) potential factors driving differencesbetween temporal versus spatial models, and (iii) the magnitude ofpotential errors relating to space for time transformations inquantifying productivity. Using two complimentary ANPP datasets andremotely sensed data derived from MODIS and a Landsat/MODIS fusiondata product, we find that slopes of spatial models are generallygreater than slopes of temporal models. The abundance of plantspecies with different structural attributes, specifically theabundance of C4 shortgrasses with prostrate canopies versus taller,more productive C3 species with more vertically complex canopies,tended to vary more dramatically in space than over time. Thisdifference in spatial versus temporal variation in these key plantfunctional groups appears to be the primary driver of differences inslopes among regression models. While the individual models revealedstrong relationships between ANPP to APAR, the use of temporal modelsto predict variation in space (or vice versa) can increase error inremotely sensed predictions of ANPP.p { margin-bottom: 0.25cm; line-height: 115%; }