Estimating biomass from body size of European spiders based on regression models

Abstract

There is a need for reliable and standardized methods to measure functional species traits. Body mass is just one dimension of body size, a most important morphological trait, because it is directly linked with metabolic rate and affected by environmental conditions. However, it is still not widely used due to the difficulties and constraints of direct measures. Weighing many (small) animals (i.e., arthropods) is laborious, time consuming and biased when using preserved material. Therefore, the applicability of general equations for estimating mass from body size of spiders was tested. We calculated linear regressions to estimate fresh and dry mass of spiders from different body measures (i.e., body length, carapace length and width) of 189 spiders sampled in southern Germany. We compared these regressions with each other and with equations from the literature and tested the impact of taxa, sex and habitat on the accuracy of biomass estimates using an independent test dataset of 166 spiders. All size-fresh mass regressions were highly significant with R2 values between 0.81 and 0.98. The slope of the ln-transformed body mass - body size relationship ranged between 2.51 and 2.95. The regressions including total body length always showed higher R2 values, i.e., they provide better predictions of body mass than carapace measures. The body length-dry mass regression was also highly significant and the mean ratio dry mass/fresh mass was 0.22. Taxon-, sex- or microhabitat-specific regressions did not produce better estimates than general regressions. Therefore, we strongly recommend the use of general regressions in the context of biomass estimation of assemblages and propose parameters from our regressions to be used for European spiders.