Modelling Climate Variability and Agricultural Productivity in the Garhwal Himalaya Using Granger Causality and ARIMA Approaches

Climate variability is considered one of the major challenges facing agricultural productivity in ecologically fragile mountain regions such as the Garhwal Himalaya. A study on this topic examines whether temperature and rainfall are Granger-causing each other, considering crop yields; it applies a Granger Causality Test and ARIMA-based time series modelling. A time series dataset of climate, from 1951 to 2023, including maximum temperature (T-max), minimum temperature (T-min), and rainfall, was collected to determine casual relationships and further predict future trends. In addition, the results indicate that T-max has a significant influence on the pattern of rainfall; hence, temperature changes cause a change in precipitation, whereas the latter is not found to be causally related to the former. Lastly, T-max and T-min were strongly interdependent, meaning that the fluctuation in temperature played a central role in influencing agricultural outcomes. The ARIMA model reveals a minor cooling trend in T-max from 2024 to 2030; thus, this might benefit heat-sensitive crops like wheat and potatoes but possibly be a concern for warm temperature crops like rice and mango. Such outcomes, therefore, demand the significance of climate-responsive agricultural planning, especially in rain-fed farming systems that are vulnerable to irregularity in precipitation. With the implementation of predictive climate models and adaptive farming strategies, policymakers and farmers can develop more resilient agricultural systems. The study, therefore, calls for enhanced water resource management, climate-smart crop selection, and policy intervention to mitigate climate risks in the Garhwal Himalaya. Future research studies should include extra climatic and agronomic variables by using high-level machine learning models to better enhance the forecast accuracy and resilience of agriculture with ongoing climate change.