Fluxbots: A Method for Building, Deploying, Collecting and Analyzing Data From an Array of Inexpensive, Autonomous Soil Carbon Flux Chambers

Abstract Soil carbon flux rates are a crucial metric of carbon cycling that contribute to calculating an ecosystem's carbon budget, and thus whether it is a source or sink of atmospheric carbon dioxide. However, soil carbon flux datasets are frequently low‐resolution across either space or time, limiting our abilities to identify small‐scale ecological contexts that influence soil carbon dynamics. Existing datasets are distributed unevenly, with some soil carbon‐rich regions (like tropical grasslands) significantly understudied. We developed an autonomous, inexpensive, do‐it‐yourself (DIY) soil carbon flux chamber (a “fluxbot”) and data processing software. We deployed a distributed array of 12 fluxbots in a long‐term experiment in a central Kenyan savanna where it has been logistically impossible to collect high‐resolution soil carbon flux data. With this array we collected over 10,000 individual flux estimates over almost two months, spanning the end of a dry season and the start of a wet season. With our successful deployment in situ, we demonstrate the potential for low‐cost, autonomous, DIY sensors in improving resolution of soil carbon flux datasets (particularly in under‐studied or logistically challenging systems). If implemented widely, such an improvement in data collection capacities could improve our understanding of ecological and climatic drivers of soil carbon flux dynamics on the local to global scale.