Replicating IoT Sensors for Soil Moisture Monitoring in Maize Farms: A Study in Zimbabwe

This study builds on previous research conducted in Zimbabwe from to , focusing on the deployment and adoption of Internet of Things (IoT) sensors for monitoring soil moisture levels in maize farms. A replication study design was employed, mirroring the original methodology. Sensors were installed across selected maize fields, with data collected over a period to evaluate sensor performance and its influence on soil moisture levels. The sensors demonstrated an accuracy rate of 95% in measuring soil moisture accurately within ±2%, indicating high precision in their functionality. The replication study confirms the effectiveness of IoT sensors for monitoring soil moisture, aligning with previous research findings but providing a contemporary validation and additional data points. Given the positive outcomes, there is a recommendation to integrate these sensors into agricultural management practices as a standard tool for enhancing crop yields and sustainability. Model estimation used $\hat{\theta}=argmin_{\theta}\sum_i\ell(y_i,f_\theta(x_i))+\lambda\lVert\theta\rVert_2^2$, with performance evaluated using out-of-sample error.