Advocates for the Anthropocene accept that humans are significantly altering the environment¹. This research addresses the inverse question of whether and how environmental disturbances influence human behavior². This research further presents empirical evidence of feedback impacts of human adaptation responses on the social-environmental systems³. 

Climatic shocks and stresses often force individuals to make changes to their environment to sustain life and livelihoods. Many of these adaptation actions result in land changes, such as shifting from rice cultivation to aquaculture. Once aggregated, these land changes can have unexpected long-term consequences, such as increasing soil and groundwater salinity, affecting the well-being of people and places. Drivers and outcomes of land changes are especially critical to understand because such changes are largely irreversible in the limited resources and timeframes useful for human societies⁴. 

It is particularly urgent to study these land changes in coastal areas, which house over 40 percent of the human population and unique ecosystems but are transforming rapidly due to environmental disturbances as well as human adaptation responses⁵. It is a particularly pressing issue for regions with weak land governance, such as parts of the global South, where aggregated loosely regulated individual actions have the potential to lock places in unsustainable development pathways or poverty traps⁶. 

Objectives

This study assesses how environmental shocks and stresses contribute to human-induced land changes along the coast of India. In particular, the study examines the impact of historical cyclonic storm surges and coastal salinity on land changes to aquaculture from 1985 to 2022. Additionally, we assess the changes in soil salinity before and after the land change, to understand the feedback impact of land change on salinity. 

Methodology

This research combines spatial data and quantitative statistical methods. First, temporal land cover changes to aquaculture are identified along the Indian coast using Landsat data. Second, spatial extents of coastal floods for the period 1985-2022 are derived from two sources: Global Flood Database based on MODIS data for the 2000-2022 period⁸, and constructed using Dartmouth Flood Observatory⁹ data and Landsat satellite images for the period 1985-2000. Third, the extent of soil salinity is assessed for the starting period (1985) and ending period (2022) using Landsat data¹⁰. Finally, socioeconomic variables from the Census of India, such as poverty rate and average household size, are included in the model to control for other external drivers. 

Using a pixel of 30mx30m as the unit of analysis, this study assesses the responsiveness of land-use change to environmental drivers. A spatial regression analysis set up using a difference-in-difference framework permits a comparison between roughly similar places that differ in their environmental shock and stress conditions and land change trajectories over time and space.

Findings

The study finds a strong association between existing salinity conditions and aquaculture activity. This suggests a potential increase in aquaculture activity in the future, given salinity is expected to increase following projections of climate-change-induced sea-level rise^11,12. The analysis also finds that a place affected by past coastal flooding and storm surges has a higher likelihood of getting converted into aquaculture, as compared to those that are not impacted. Moreover, the study finds a neighborhood amplification effect, where areas adjacent to those directly impacted by storms in the past also tend to get converted into aquaculture. A second-order feedback outcome is explained by the strong correlation between aquaculture land change and increased salinity outcomes over time.