We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Environmental factors related to biophysical suitability and agronomic effects of biodegradable mulch applications: Benchmarking key variables using machine learning
Summary
Researchers used machine learning models to analyze how environmental conditions like temperature, precipitation, soil pH, and organic carbon affect whether biodegradable plastic mulches — an alternative to conventional plastic that can shed microplastics — effectively support crop growth and break down at appropriate rates. Results identified optimal conditions, including soil pH between 6.2–7.8 and monthly temperatures above 3.4°C, as benchmark thresholds for successful biodegradable mulch performance.
The agricultural sector faces unprecedented challenges in sustaining food production amidst rising population and environmental change. These environmental changes include droughts, rising temperatures, soil erosion, and weed invasion. Prior research has explored the potential of biodegradable mulch to help address these challenges while also reducing microplastic pollution associated with plastic mulch. Despite numerous research efforts on biodegradable mulch, there is limited evidence on how environmental factors influence the effectiveness of biodegradable mulch across diverse sites. This study uses machine learning models to examine how biophysical environmental conditions relate to the agronomic impacts and degradation rates of biodegradable mulch. The results of Random Forest, Support Vector Machine, and Decision Tree models confirm that precipitation and temperature are relevant in predicting the effects of biodegradable mulches, with hot and arid climate conditions associated with positive mulch effects. Soil attributes, including texture, organic carbon, and pH levels, are also identified as key variables. Notably, Decision Tree models indicate that maintaining soil pH levels between 6.2–7.8 and ensuring minimum monthly temperatures exceed 3.4 °C are identified as benchmark values for achieving positive mulch effects on agronomic performance. Meanwhile, monthly precipitation above 78 mm is associated with high degradation rates that exceed regulatory standards and reduce the effectiveness of mulch application. Despite variations in overall accuracy, the Random Forest and Decision Tree models demonstrated robustness in their potential reliability in classifying mulch effect outcomes. These results serve as a useful guide to identifying potential suitable sites for biodegradable mulch application and suggest further product development is needed to meet the needs of diverse environmental contexts in efforts to scale up adoption.