Partitioning macroscale and microscale ecological processes using covariate-driven non-stationary spatial models

Area of Science:

• Ecology
• Environmental Science
• Spatial Statistics

Background:

• Ecological inference often requires integrating data across spatial scales, leading to complex spatial dependence structures.
• Fully non-stationary models accurately represent these structures but are challenging for ecologists to estimate and interpret.
• Lake eutrophication is a pervasive environmental issue driven by complex spatial processes.

Purpose of the Study:

• To improve the interpretability of a recently developed non-stationary model for ecological applications.
• To apply the enhanced model to understand spatial processes driving lake eutrophication in different US regions.
• To visualize and explain the spatial dependence structure of environmental variables using ecologically relevant covariates.

Main Methods:

• Reformulated a non-stationary model by incorporating environmental predictors into the covariance function.
• Developed visual tools (ellipses) to represent spatial correlation range and directionality.
• Applied the model to analyze total phosphorus and chlorophyll a spatial structures in the Midwest and Northeast US.

Main Results:

• In the Midwest, forest cover influenced total phosphorus spatial homogeneity (macroscale processes), while forest cover and baseflow affected chlorophyll a spatial homogeneity (microscale processes).
• In the Northeast, urban land use and baseflow decreased phosphorus concentration homogeneity (microscale processes), but covariates did not strongly explain chlorophyll a spatial structure.
• The study demonstrated that spatial dependence structures shift across space and can be explained by covariates from different scales.

Conclusions:

• The enhanced non-stationary model facilitates the use and interpretation of complex spatial models in ecology.
• The findings provide novel insights into the scale-dependent spatial processes driving lake eutrophication.
• Understanding these spatial dynamics is crucial for managing complex environmental issues like eutrophication.