Machine learning approach to estimate soil matric potential in the plant root zone based on remote sensing data

Abstract

There is an increasing interest in using the Internet of Things (IoT) in the agriculture sector to acquire soil- and crop-related parameters that provide helpful information to manage farms more efficiently. One example of this technology is using IoT soil moisture sensors for scheduling irrigation. Soil moisture sensors are usually deployed in nodes. A more significant number of sensors/nodes is recommended in larger fields, such as those found in broadacre agriculture, to better account for soil heterogeneity. However, this comes at a higher and often limiting cost for farmers (purchase, labour costs from installation and removal, and maintenance). Methodologies that enable maintaining the monitoring capability/intensity with a reduced number of in-field sensors would be valuable for the sector and of great interest. In this study, sensor data analysis conducted across two irrigation seasons in three cotton fields from two cotton-growing areas of Australia, identified a relationship between soil matric potential and cumulative satellite-derived crop evapotranspiration (ETcn) between irrigation events. A second-degree function represents this relationship, which is affected by the crop development stage, rainfall, irrigation events and the transition between saturated and non-saturated soil. Two machine learning models [a Dense Multilayer Perceptron (DMP) and Support Vector Regression (SVR) algorithms] were studied to explore these second-degree function properties and assess whether the models were capable of learning the pattern of the soil matric potential-ETcn relation to estimate soil moisture from satellite-derived ETc measurements. The algorithms performance evaluation in predicting soil matric potential applied the k-fold method in each farm individually and combining data from all fields and seasons. The latter approach made it possible to avoid the influence of farm consultants’ decisions regarding when to irrigate the crop in the training process. Both algorithms accurately estimated soil matric potential for individual (up to 90% of predicted values within ±10 kPa) and combined datasets (73% of predicted values within ±10 kPa). The technique presented here can accurately monitor soil matric potential in the root zone of cotton plants with reduced in-field sensor equipment and offers promising applications for its use in irrigation-decision systems.

Remote sensing to predict soil moisture tension in water saving rice systems of temperate South-Eastern Australia

“…Such forecasting has been demonstrated for use in cotton production by Refs. 36 and 45 and offers potential to be incorporated into an automated irrigation system for water saving rice production. Although generally not an issue in semi-arid regions, cloud cover may limit the accuracy of predictions, with increased frequency of imagery preferable.…”

“…34,35 Further, in broadacre surface irrigation, growers of crops such as wheat and cotton use cumulative ETc as calculated by remote sensed decision support tools, such as IrriSAT and EEFLUX, to schedule irrigation. 30,31 However, this is not currently possible in rice as ETc irrigation thresholds for water saving rice are not known, nor is the relationship between ETc and SMT in water saving rice systems despite 36 reporting a relationship between in-field sensed SMT and satellite derived ETc between irrigation events in cotton. SMT decline during non-ponded periods of rice cultivation is hypothesized to be correlated to the cumulative ETc between irrigation events.…”

Cotton aphid infestation monitoring using Sentinel-2 MSI imagery coupled with derivative of ratio spectroscopy and random forest algorithm

“…In addition to feature extraction, the factors that affect the accuracy of pest classification are also related to the selection of classification methods. At present, machine learning algorithms have been applied in many fields and achieved good results ( Lu et al., 2022 ; Maia et al., 2022 ). In this study, five machine learning algorithms were selected for comparison of classification accuracy, and the best classification performance was achieved by RF, demonstrating the good performance of the algorithm in terms of pest classification potential.…”

Back to the future for drought tolerance

“…This can be achieved through methods such as remote sensing or the utilization of soil moisture probes. While existing techniques for soil moisture measurement primarily serve large‐scale hydrological and geoscience research (Liu et al ., 2022) or farming decision‐making (Maia et al ., 2022), developing more suitable approaches tailored for plant breeding is essential. The EM38, an electromagnetic induction instrument, offers a noninvasive and rapid approach for measuring soil moisture at multiple soil depths and soil electrical conductivity (Phathutshedzo‐Eugene et al ., 2023), making it promising for incorporation in plant breeding research trials.…”