Does sampling resolution influence the relationship between plant community diversity and aboveground productivity at a northern prairie site? An investigation using ground-based radiometry

While recent studies generally support the idea that diverse plant communities function better than depauperate versions, there is less agreement as to whether this is caused by the number of species present, the identities of the species present, the number of functional groups that these species make up, or by which functional groups are represented. In this study, we use ground-based remote sensing and botanical field sampling to address these issues at a northern prairie site. We assess the degree to which observed diversity-productivity relationships are dependent on (a) the measure of diversity used (species richness vs. functional group richness), (b) the presence of particular species or functional groups, and (c) sampling resolution. To achieve this, we collected plant diversity and spectrally-derived productivity information over plots of comparable ground resolution (approx. 0.5 m) during the summer of 1998 from three upland grassland communities, then used a spatially nested sampling design, to scale each property and their relationships to 2.5 m, 10 m and 50 m sampling resolutions. Our results showed that the overall strengths and shapes of observed diversity-productivity relationships were dependent on all of the above factors. Overall richness-productivity relationships were found to be asymptotic at all observational scales. Multiple regressions indicated that the individual effects of species richness were generally stronger and more significant than functional group richness. The presence of particular species had significant effects on productivity at the plot (0.5 m) level, but not at coarser resolutions.