Improved allometric proxies for eelgrass conservation

Current anthropogenic influences threaten the permanence of eelgrass, a relevant macrophyte that brings about important ecological benefits including nursery for waterfowl and fish species, shoreline stabilization, nutrient recycling and carbon sequestration. Eelgrass restoration normally involves transplanted plots and monitoring success requires noninvasive assessments of standing stock and productivity. Allometric scaling of eelgrass leaf biomass and length can provide proxies for these assessments, but accuracy of allometric projections is mainly resultant of uncertainty propagation of parameters, so for the sake of suitability it is very important ensuring the most accurate estimates. The traditional approach for producing estimates of allometric parameters considers a linear regression model involving logarithms of the original response and explanatory variables along with a normally distributed additive error. The suitability of this method has been questioned on the ground of biased results raising nonlinear regression as a necessary amendment. Here we demonstrate that this controversy can be surpassed allowing for a logistic error structure and heteroscedasticity in the traditional method. The present arrangement delivered parameter estimates from raw data surpassing inconveniences of the traditional fitting procedure. Moreover, associated allometric proxies for average leaf biomass in shoots entailed similar reproducibilities than produced by using nonlinear regression and quality controlled data. In achieving suitable accuracy levels, the present approach required a sample of raw data of about 9% of involved in prior estimations based on quality controlled data. The improvements associated to the present approach grant highly consistent non-destructive assessments for the sake of eelgrass conservation.