Author
|
Submitted to: Journal of Plant Ecology
Publication Type: Peer Reviewed Journal Publication Acceptance Date: 4/1/2005 Publication Date: 12/20/2006 Citation: Goslee, S.C. 2006. Behavior of Vegetation Sampling Methods in the Presence of Spatial Autocorrelation. Journal Of Plant Ecology. 187(2):203-212. Interpretive Summary: Plant species in natural ecosystems are rarely distributed uniformly. The interaction between spatial autocorrelation, total plant cover, and sampling method can have important implications for research conclusions. Simulation results show that transects are inefficient at sampling highly clumped distributions, but that randomly located quadrats are much better. Multiscale sampling schemes are very effective at locating rare species and estimating total species richness, but are more resource-intensive to sample. Technical Abstract: Spatial autocorrelation in vegetation has been discussed extensively, but little is known about how various plant sampling methods perform when confronted with varying levels of autocorrelation. Simulated species distributions with varying levels of total abundance and spatial autocorrelation were sampled using four methods: strip transect, randomly located quadrats, the non-nested multiscale modified Whittaker plot and the nested multiscale North Carolina Vegetation Survey plot. Cover and frequency estimates varied widely within and between methods, especially in the presence of high spatial autocorrelation and for species with moderate abundances. Transect sampling showed the highest variability. Transect and random methods were likely to miss rare species entirely, even with large numbers of quadrats. North Carolina Vegetation Survey plots were the most accurate because cover estimates were made on a larger area than used by the other three methods. Total species richness calculated using semilog species-area curves was overestimated by transect and random sampling. Both multiscale methods overestimated species richness when autocorrelation was low, and overestimated it when autocorrelation was high. There was no clear advantage of nested or non-nested methods for any of the measures assessed. For all methods, cover and especially frequency estimates were highly variable, and depended on both the degree of autocorrelation and the sampling method used; care must be taken in comparing results between studies. |
