Intelligent Agricultural Management Considering N$_2$O Emission and Climate Variability with Uncertainties

TL;DR

The paper tackles the challenge of climate variability and soil N$_2$O emissions in farming by formulating agricultural management as a partially observable Markov decision process and solving it with a recurrent deep Q-network embedded in Gym-DSSAT. It introduces both deterministic and probabilistic ML N$_2$O emission predictors, integrates them into a crop simulator, and uses a stochastic weather generator to test policy robustness under warming and drought. The study demonstrates that N$_2$O-aware RL policies can balance yield, fertilizer and irrigation use, and leaching/emission penalties, with optimal policies showing superior adaptability to temperature and precipitation changes. This work advances sustainable agriculture by combining POMDP-based RL, probabilistic emission forecasting, and climate-variability testing to derive resilient management strategies with practical implications for GHG mitigation and resource efficiency.

Abstract

This study examines how artificial intelligence (AI), especially Reinforcement Learning (RL), can be used in farming to boost crop yields, fine-tune nitrogen use and watering, and reduce nitrate runoff and greenhouse gases, focusing on Nitrous Oxide (N$_2$O) emissions from soil. Facing climate change and limited agricultural knowledge, we use Partially Observable Markov Decision Processes (POMDPs) with a crop simulator to model AI agents' interactions with farming environments. We apply deep Q-learning with Recurrent Neural Network (RNN)-based Q networks for training agents on optimal actions. Also, we develop Machine Learning (ML) models to predict N$_2$O emissions, integrating these predictions into the simulator. Our research tackles uncertainties in N$_2$O emission estimates with a probabilistic ML approach and climate variability through a stochastic weather model, offering a range of emission outcomes to improve forecast reliability and decision-making. By incorporating climate change effects, we enhance agents' climate adaptability, aiming for resilient agricultural practices. Results show these agents can align crop productivity with environmental concerns by penalizing N$_2$O emissions, adapting effectively to climate shifts like warmer temperatures and less rain. This strategy improves farm management under climate change, highlighting AI's role in sustainable agriculture.

Figures (11)

• Figure 1: GRU-based Q-network architecture.
• Figure 2: The predictive N2O daily flux (g/ha) compared to true values by using a deterministic ML model.
• Figure 3: The predictive N2O daily flux (g/ha) compared to true values by using a probabilistic ML model.
• Figure 4: Fertilization strategies across three cases.
• Figure 5: Irrigation strategies across three cases.