Fluxes, Feedbacks, and Memory: Untangling Chamoli’s Seasonal Land–Atmosphere Coupling

Chamoli district, located in the Garhwal Himalaya of Uttarakhand, functions as a critical ecological buffer connecting mountain environments with downstream river systems. Its complex terrain, diverse biota, and glacier-fed rivers play an essential role in sustaining regional water resources and enhancing climate resilience. Despite its importance, studies exploring the land–atmosphere coupling processes in this climate-resilient region remain scarce.

In this work, we employ an information-theoretic approach to examine seasonal land–atmosphere interaction networks using key variables: precipitation (P), temperature (T), latent heat flux (LH), sensible heat flux (SH), wind speed (WS), incoming shortwave radiation (SWL), and relative humidity (Q). The analysis is conducted for four seasons: pre-monsoon (MAM), monsoon (JJAS), post-monsoon (ON), and winter (DJF). The derived networks distinguish between two types of links: instantaneous (real-time) and lagged (memory-controlled). Entropy-based diagnostics indicate that MAM and JJAS exhibit the highest dynamical variability, DJF represents the most quiescent period, and ON behaves as a transitional regime for Chamoli. Wind speed exerts a dominant real-time control on precipitation and also shows delayed influences at higher altitudes. In general, real-time coupling is strongest during the monsoon season, whereas comparatively enhanced memory-driven relationships mark winter.

The pre-COVID and post-COVID periods are compared to assess changes in information flow; we find that entropy deviation decreased around 2019, then increased after 2021. These findings refine our understanding of land–atmosphere dynamics over Chamoli and provide a reference state for evaluating future changes arising from natural climate variability and anthropogenic forcing.

Keywords—Land-atmospheric interaction; information-centric method; real-time interaction; entropy.