What the study found
The study found that the global eddy covariance tower network is unevenly distributed across space and soil conditions. Among 1,233 towers, half had a nearest neighbor within 10 km, and tower locations were associated with soil pixels that were more fertile than the global average.
Why the authors say this matters
The authors conclude that global syntheses of eddy covariance tower data should take into account that the network captures more fertile soils than the terrestrial surface on average. They also suggest that improving global representativeness will require collaborations and investment in underrepresented regions.
What the researchers tested
The researchers examined 1,233 global eddy covariance towers and compared their spatial spacing and surrounding soil characteristics with global soil database pixels. They looked at nearest-neighbor distances between towers and compared tower-associated soil texture and chemistry with pixels without towers.
What worked and what didn't
Tower-associated soil pixels had nearly 20% more silt and 8% less sand than the global soil texture distribution. They also had more soil nitrogen (0.58 g/kg vs. 0.38 g/kg), more organic carbon (8.3 g/kg vs. 5.4 g/kg), and 10% greater cation exchange capacity in upper layers than pixels without towers.
What to keep in mind
The abstract does not describe specific study limitations beyond noting gaps in network representativeness. The findings are based on the towers and soil data examined here, so the summary does not provide details about causes or effects beyond those comparisons.
- The study examined 1,233 global eddy covariance towers.
- Half of the towers had a nearest neighbor within 10 km.
- Tower locations were linked to soils with more silt, less sand, more nitrogen, and more organic carbon than average.
- Upper soil layers at tower pixels had 10% greater cation exchange capacity than pixels without towers.
- The authors say global tower syntheses should account for the network’s bias toward more fertile soils.
