Date of Graduation

5-2016

Document Type

Thesis

Degree Name

Master of Arts in Geography (MA)

Degree Level

Graduate

Department

Geosciences

Advisor

Thomas R. Paradise

Committee Member

Fiona M. Davidson

Second Committee Member

Stephen K. Boss

Third Committee Member

Lawton L. Nalley

Fourth Committee Member

Kenneth L. Kvamme

Abstract

Droughts have affected more people than any other natural disaster in the last century, causing billions in economic damages and millions of deaths. As the Sea Surface Temperatures (SST) have heated in the Indian Ocean, drought patterns across South Asia have changed; the Indian monsoon has become more volatile and less predictable. In this study, monthly self-calibrated Palmer Drought Severity Index (sc-PDSI) data for the time period between 1950 and 2009 were interpolated to India’s districts; then the data were analyzed for changes in frequency and severity. The data were further evaluated using Anselin’s Local Moran’s I Statistic to elicit the spatial autocorrelation of droughts by decade from 1950 to 2009. This analysis showed the concentration of droughts in certain regions of India relative to others for each decade. Droughts in the 2000-2009 time period were particularly concentrated and severe over the Gangetic Plains, one of the primary regions for agricultural production.

After analyzing drought patterns in India, data for rice and wheat production between 1998 and 2009 were collected for each district. Then, an Ordinary Least Squares regression was used to analyze the metric tonnes (MT) of production per hectare per district and the total hectares planted in each district with annual sc-PDSI data and locational fixed effects by state for both rice and wheat. The MT per hectare measured each districts production efficiency during drought, while the hectares planted was used to proxy producer decision-making during a drought year. Droughts, as estimated by sc-PDSI, were significant at the P < 0.01 level for all four regressions: 1) MT per hectare of rice; 2) hectares planted to rice; 3) MT per hectare of wheat; and 4) hectares planted to wheat. The results suggest that during drought years, rice production is reduced by 0.04 MT per hectare (~2%), but wheat production increases by 0.02 MT per hectare (~3%). Conversely, districts planted an average of 2,210 more hectares of rice during a drought year and 1,133 less hectares of wheat. This may indicate that farmers are taking on more risk during drought years in the hope that price inflations for rice will be more profitable.

Furthermore, this study used data from the Indian Census of 2011 to explore the occurrence of socio-economic and cultural disparities in relation to drought and agricultural production. Statistics on labor, scheduled castes and tribes, and illiteracy were used as proxies to identify districts where impoverished communities reside. Correspondingly, among global change topics, there is growing literature around the concept of “double exposure”, which suggests that certain communities are vulnerable to a multiplicity of social and environmental factors. Several steps were taken to identify and map the districts in India that could be doubly exposed to both droughts and socio-economic or cultural status in agrarian areas. Overall, this study found changes in drought patterns, their relationship to rice and wheat production, and provided a platform for future data collection in the context of regional climate change and agrarian vulnerability.

Share

COinS