Reducing Sampling Uncertainty in Aeolian Research to Improve Change Detection

• Published in: Journal of geophysical research. Earth surface Vol. 124; no. 6; pp. 1366 - 1377
• Main Authors: Webb, Nicholas P., Chappell, Adrian, Edwards, Brandon L., McCord, Sarah E., Van Zee, Justin W., Cooper, Bradley F., Courtright, Ericha M., Duniway, Michael C., Sharratt, Brenton, Tedela, Negussie, Toledo, David
• Format: Journal Article
• Language: English
• Published: 01.06.2019
• Subjects: aeolian; change detection; dust; land cover; land use; wind erosion
• ISSN: 2169-9003; 2169-9011
• DOI: 10.1029/2019JF005042

Measurements of aeolian sediment transport support our understanding of mineral dust impacts on Earth and human systems and assessments of aeolian process sensitivities to global environmental change. However, sample design principles are often overlooked in aeolian research. Here we use high‐density field measurements of sediment mass flux across land use and land cover types to examine sample size and power effects on detecting change in aeolian transport. Temporal variances were 1.6 to 10.1 times the magnitude of spatial variances in aeolian transport for six study sites. Differences in transport were detectable for >67% of comparisons among sites using ~27 samples. Failure to detect change with smaller sample sizes suggests that aeolian transport measurements and monitoring are much more uncertain than recognized. We show how small and selective sampling, common in aeolian research, gives the false impression that differences in aeolian transport can be detected, potentially undermining inferences about process and impacting reproducibility of aeolian research. Plain Language Summary Aeolian sediment transport, including wind erosion and dust emission, impacts agricultural production and food security, nutrient cycling, water resources, and climate. Measuring aeolian sediment transport is therefore important for developing an understanding of its impacts on Earth systems and society. However, little consideration has been given to how many samples are needed to measure aeolian transport and detect its change across space and through time. We investigate how sample size, design, and decisions about the precision of change detection affect aeolian transport monitoring. Using field measurements, we show that traditional approaches in aeolian research with small sample sizes and selective placement of equipment are often unable to detect change and support robust inferences about aeolian processes. Unless large numbers of samples are used, uncertainty in field measurements can be so large that it undermines our understanding of how and why aeolian sediment transport rates change across space and through time. Key Points Uncertainties in monitoring limit statistical inferences and our understanding of aeolian process responses to environmental change Small sample sizes lack power to detect change in aeolian transport rates across space and through time Sampling that adequately estimates spatial and temporal variances is needed to monitor aeolian transport with confidence to detect change