Assessing Agricultural System Vulnerability to Floods: A Hybrid Approach Using Emergy and a Landscape Fragmentation Index
Feng Wu,Yufan Sun,Zhongxiao Sun,Sihong Wu,Qian Zhang
DOI: https://doi.org/10.1016/j.ecolind.2017.10.050
IF: 6.9
2017-01-01
Ecological Indicators
Abstract:In recent decades, the frequencies and intensities of extreme weather events have increased in many parts of the world. Floods, as one of the main types of extreme weather event, have a major influence on agroecosystem productivity, and, in turn, on agricultural income and food security. Consequently, analyzing agricultural system vulnerability to floods plays a significant role in food production and agroecosystem health. In this study, we establish a three-layer indicator system to evaluate agricultural vulnerability at the county level for flood-prone regions in China. Specifically, in the first layer, we assess agricultural vulnerability to floods based on the constructs of exposure, sensitivity, and adaptability. Indicators in the second layer include precipitation, runoff, land use, and capital, and are measured to capture the primary constructs. Together, the indicators are used to calculate agricultural system vulnerability to floods. We then innovatively correct the assessment results of vulnerability with the aid of a landscape fragmentation index, given that landscape fragmentation is known to influence the vulnerability of agricultural systems. The results for agricultural vulnerability to floods demonstrate clear spatial variations at the county level in 1995, 2000, 2005, and 2010, and also show changes in the spatial distribution of vulnerability over time. In this regard, areas that are distributed near inland rivers, lakes, and the southern coastal areas, and those areas with dense river networks, have relatively high vulnerabilities. The assessment results also indicate that the maximum and average intensities of vulnerability have decreased over time, although the extent of vulnerable agricultural land has increased. Importantly, by comparing the results between selected county pairs, the assessment results corrected using landscape fragmentation index is verified to be more robust and objective than without correction.