Skip to main content
ARS Home » Plains Area » Fort Collins, Colorado » Center for Agricultural Resources Research » Water Management and Systems Research » Research » Publications at this Location » Publication #390953
Research Project: Improving the Sustainability of Irrigated Farming Systems in Semi-Arid Regions

Location: Water Management and Systems Research

Title: Optimizing soil moisture sensor depth for irrigation management using universal multiple linear regression

Author
item CLUTTER, MELISSA - Fort Lewis College
item DeJonge, Kendall

Submitted to: Journal of the ASABE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/3/2022
Publication Date: 7/1/2022
Citation: Clutter, M., DeJonge, K.C. 2022. Optimizing soil moisture sensor depth for irrigation management using universal multiple linear regression. Journal of the ASABE. 65(4):739-749. https://doi.org/10.13031/ja.15044.
DOI: https://doi.org/10.13031/ja.15044

Interpretive Summary: Soil moisture sensors are valuable tools in irrigation management, but due to soil spatial variability the ability for point-based sensors to characterize the entire soil water profile is challenging. At the Limited Irrigation Research Farm near Greeley, Colorado, a machine learning method was used with seven years of existing soil water data to optimize the number and placement of soil water sensors to characterize the entire soil water deficit (SWD) in the root zone. We found that, at our site, a single sensor buried at 30cm could characterize the SWD and be used for irrigation scheduling. This method could simplify and improve sensor-based irrigation scheduling methods for farmers.

Technical Abstract: Soil moisture sensors are valuable tools in irrigation management, but due to soil spatial variability the ability for point-based sensors to characterize the entire soil water profile is challenging. Universal Multiple Linear Regression (uMLR) is a simple machine learning tool that analyzes the relationship between all possible subsets of observations and the prediction of interest. This study used uMLR with soil water content data (via neutron probe and time domain reflectometry) taken from maize field experiments at the Limited Irrigation Research Farm near Greeley, Colorado, to optimize the number and placement of soil water sensors to characterize the entire soil water deficit (SWD) in the root zone. We evaluated seven field seasons with 12 treatments and 4 replicates of each, where volumetric soil water content (VWC) was typically taken 1-3 times weekly, at 15cm with TDR and deeper depths with neutron probe in 30cm increments down to 200cm, and VWC values were aggregated into SWD values. UMLR analysis was used to quantify accuracy of using single depth or combinations of two depths of VWC measurement to represent SWD. Results indicate that the best single sensor depth is at 30cm, and the best two-depth combination is at 0-15cm and 60cm. A continuous VWC sensor was used at 30cm and validated results with acceptable accuracy. This study suggests that in the sandy loam site evaluated and with varying irrigation levels, soil moisture sensor placement near the soil surface can be used to characterize SWD. Future studies should explore the transferability of these results to other locations.