The spatial and environmental distribution of the global eddy covariance tower network

Key findings from this study

• The study found that half of global eddy covariance towers have a nearest neighbor within 10 kilometers, indicating pronounced spatial clustering.
• The researchers demonstrate that tower locations systematically capture soils with 20% more silt, higher nitrogen content, and greater cation exchange capacity than global soil averages.
• The authors report that global flux syntheses should recognize this sampling bias toward fertile soils when interpreting ecosystem-atmosphere interaction estimates.

Overview

Eddy covariance towers provide critical measurements of ecosystem-atmosphere interactions globally. This study evaluates whether the spatial distribution and soil characteristics at 1233 eddy covariance tower sites represent the range of terrestrial soil conditions. The analysis combines tower location data with global soil databases to assess network coverage and soil property representation.

Methods and approach

The researchers compiled geospatial data for 1233 eddy covariance towers and compared soil properties at tower locations against global soil texture and nutrient distributions. They calculated nearest-neighbor distances between towers and compared soil characteristics including texture, nitrogen content, organic carbon, and cation exchange capacity at pixels with and without towers.

Results

Half of the global eddy covariance towers have a nearest neighbor within 10 kilometers, indicating substantial spatial clustering. Soil pixels with towers contain nearly 20% more silt and 8% less sand relative to the global soil texture distribution. Towers preferentially occupy more fertile soils, with soil nitrogen at 0.58 grams per kilogram compared to 0.38 grams per kilogram globally and organic carbon at 8.3 grams per kilogram versus 5.4 grams per kilogram globally. Upper soil layers at tower sites exhibit 10% greater cation exchange capacity than pixels without towers.

The spatial clustering of towers reflects both practical research constraints and the concentration of existing scientific infrastructure in accessible regions. Tower sites systematically capture higher-fertility soils than the terrestrial surface average, which influences the environmental conditions represented in global flux synthesis studies.

Implications

Global syntheses derived from eddy covariance networks inherently overrepresent fertile soil conditions and may not reflect ecosystem-atmosphere interactions across the full range of terrestrial soil variability. This bias affects estimates of carbon cycling, nutrient fluxes, and other ecosystem functions in less fertile regions. Research conclusions about ecosystem responses to climate forcing may have limited applicability to carbon-poor and nutrient-poor systems.

Expanding tower networks toward underrepresented regions with distinct soil properties would improve global representativeness of flux measurements. Strategic placement in regions with sandy soils, lower nutrient status, and higher spatial isolation from existing towers could better capture global soil diversity. Increased collaboration and funding for towers in underrepresented areas would enhance the validity of global syntheses.

Scope and limitations

This summary is based on the study abstract and available metadata. It does not include a full analysis of the complete paper, supplementary materials, or underlying datasets unless explicitly stated. Findings should be interpreted in the context of the original publication.