Water and Energy Dynamics in a Disturbed Ecosystem

The resilience of ecosystems, especially wetlands, plays a vital role in maintaining biodiversity, regulating climate, and supporting the livelihoods of communities. However, these ecosystems are increasingly threatened by both human activities and natural disturbances. To effectively manage and conserve these environments, it is imperative to comprehend the dynamics of various proxies that can indicate the state of an ecosystem, which may exhibit linear, non-linear, or abrupt changes over time. Although there has been considerable research focused on the quantitative analysis of wetland ecosystem dynamics prior to a state change—such as the transition from wetland to bare land—there remains a gap in understanding the alterations in water and energy balance dynamics within wetlands. In this study, we investigated the temporal patterns of key components of the energy balance system, including latent heat flux, sensible heat flux, and soil heat flux, alongside water dynamics as primary indicators of wetland change. We quantified water and energy dynamics using data from the Moderate Resolution Imaging Spectroradiometer (MODIS) and the National Centers for Environmental Prediction (NCEP). By employing the Brock, Dechert, and Scheinkman (BDS) test along with non-parametric drift-diffusion jump (NDDJ) model approaches, we analyzed the temporal dynamics ofwater and energy dynamics at Kobi Wetland, a site protected under the Ramsar Convention. KobiWetland has experienced significant reductions in water levels, largely attributed to drought conditions and unsustainable agricultural practices in the area. The findings revealed a significant increase in non-linear dynamics in both water area and energy dynamics over the study time period. The BDS test and NDDJ analysis showed that sensible heat flux showed the strongest trend (Kendall’s τ = 0.48), with significant nonlinearity, compared with other indicators that can signal a loss of resilience and abrupt change. In contrast, the water area indicator (i.e., Modified Normalized Difference Water Index) has a weak trend (τ = 0.14) with minimal nonlinearity, while latent and soil heat flux exhibit moderate trends (Kendall’s τ = 0.34 and 0.38) and nonlinearity. This research contributes to the drought monitoring and predicting of water and energy dynamics changes in wetland ecosystems that can serve as signals of resilience loss.