Data, data everywhere: detecting spatial patterns in fine-scale ecological information collected across a continent
Abstract
Context Fine-scale ecological data collected across broad regions are becoming increasingly available. Appropriate geographic analyses of these data can help identify locations of ecological concern. Objectives We present one such approach, spatial association of scalable hexagons (SASH), which
identifies locations where ecological phenomena occur at greater or lower frequencies than expected by chance. This approach is based on a sampling frame optimized for spatial neighborhood analysis, adjustable to the appropriate spatial resolution, and applicable to multiple data types. Methods We divided portions of the United States into scalable equal-area hexagonal cells and, using three types of data (field surveys, aerial surveys, satellite imagery), identified geographic clusters of forested areas having high and low values for (1) invasive plant diversity and cover, (2) mountain pine
beetle-induced tree mortality, and (3) wildland forest fire occurrences. Results Using the SASH approach, we detected statistically significant patterns of plant invasion, bark beetle-induced tree mortality, and fire occurrence density that will be useful for understanding macroscale patterns and processes associated with each forest health threat, for assessing its ecological and economic impacts, and for identifying areas where specific management activities may be needed. Conclusions The presented method is a ‘‘big data’’ analysis tool with potential application for macrosystems ecology studies that require rigorous testing of hypotheses within a spatial framework. This method is a standard component of annual national reports on forest health status and trends across the United States and can be applied easily to other regions and datasets.