Application of low-cost soil moisture sensors for irrigation management in Brassica oleracea var. acephala cultivation

Abstract

Water scarcity demands efficient irrigation, yet conventional technologies remain inaccessible to low-income farmers. Low-cost, easy-to-use sensors represent a promising alternative. This study develop and evaluate a soil moisture monitoring system using low-cost capacitive sensors (SKU: CE09640) integrated with an Arduino microcontroller, featuring RGB LED alerts to indicate soil water status. Field validation was conducted in kale (Brassica oleracea var. acephala) cultivation, comparing three irrigation management (IM) strategies: crop evapotranspiration (ETc), tensiometers (IMT), and capacitive sensors (IMC). Sensors were calibrated in the laboratory using the gravimetric method. The experiment followed a randomized block design in a 4 × 3 factorial scheme, with four irrigation levels (50 %, 75 %, 100 %, and 125 % of the irrigation depth estimated by each method). Number of commercial (NCL) and non-commercial leaves (NNCL), fresh weight (FMML), number of mature leaves (NML), and water use productivity (WUP) were evaluated. Calibration results showed a strong inverse correlation (R < -0.964), high fit (R2 > 0.95), and acceptable accuracy (RMSE < 0.05) in predicting soil moisture compared to gravimetric method. Field validation revealed no significant interaction between factors and similar performance among IM strategies for most variables. ETc-based IM showed the highest WUP (12.42 g l -1), followed by IMC (10.95 g l -1), and IMT (9.82 g l -1), indicating that the SKU: CE09640 sensors enables greater water savings than conventional devices like tensiometers. The findings support the use of SKU: CE09640 sensors is a viable, accessible solution for IM, promoting efficient water use and benefiting smallholder farmers with limited educational backgrounds or technical training.