Toward a Standardized Method for Quantifying Ecosystem Hot Spots and Hot Moments.

Ecosystem "hot spots" and "hot moments"—respectively, places and times of disproportionately high biogeochemical activity—are an important and often invoked concept in ecosystem science. Despite the popularity of the concept, there is no standard approach to quantifying hot spots and hot moments, hindering progress in understanding the phenomenon. For example, lack of a standard quantitative approach hinders advances arising from synthesis across datasets and studies potentially representing different processes, ecosystem types, places, and times. We present an approach to quantifying hot spots and hot moments based on the skewness and kurtosis of data distributions. Our approach explicitly tests for the presence of hot spots and/or hot moments, as well as identifies observations regarded as hot spots and/or hot moments. We apply our method to three case studies representing different ecosystem contexts and focal variables: soil pore space CO2 concentrations in a humid temperate watershed; dissolved oxygen saturation in a hypereutrophic, shallow lake; and seagrass metabolism on the coast of a tropical island. A rarefaction-based sensitivity analysis showed how detection of HSHM using our methods can be sensitive to changes in spatial and temporal sampling regimes, depending on the behavior of the system. To facilitate adoption of our approach, we provide the R package "hotspomoments" to implement the method. Adoption of a standard quantitative approach to hot spots and hot moments would advance ecosystem science. [ABSTRACT FROM AUTHOR]