Multi-Scale and Multi-Sensor Analysis of Urban Cluster Development and Agricultural Land Loss in China and India

Over the next two decades, the combined urban population in China and India will grow by more than 700 million. China’s urban population is expected to increase by 400 million and India’s urban population will nearly double from today’s 350 million. Put into a global context, by 2030, nearly one-third of the world’s urban inhabitants will live in either China or India. The primary goal of this NASA LCLUC project was to quantify and understand the growth of urban clusters and the loss of agricultural land in these two rapidly urbanizing countries. I will highlight two key outputs from the project: the development of a methodology to identify the growth of urban clusters and the development of an algorithm to identify the loss of agricultural land. Both methods use time series MODIS and DMSP OLS data to assess land change.

In this study, we combined a hierarchical classification approach with econometric time series analysis to examine the loss of agricultural land in India using time series night time lights data and MODIS NDVI images for the period June, 2000 to May, 2011 (Pandey and Seto, 2015). Our methodology to identify the loss of agricultural land loss was in three stages: a) generating an urban/non-urban mask; b) identifying agricultural pixels that had been converted to urban; and c) identifying the timing of the land conversions (Fig 1).

Figure 1. Remote sensing methodology flowchart.

We then compared the loss of agricultural land as estimated by MODIS-NTL analysis with state-level agricultural land use statistics from the India Ministry of Agriculture. Our analysis uncovers some key trends in agricultural land loss. First, during 2001 – 2010, India lost 0.7 million hectares (roughly five times the size of Delhi) of its agricultural land to urban expansion. Second, agricultural land loss is occurring around smaller cities more than around bigger cities. Third, the northeastern states experienced the least amount of agricultural land loss compared to other states (Fig 2).

Our findings also show that there are geographic patterns to agricultural land loss: agricultural land loss is concentrated in a few districts and states, mainly those which have a larger number of operational or approved SEZs, with high rates of economic growth (Fig 3), and with higher agricultural land suitability compared to other states. Read more about the project

Figure 2. Total Area of Agricultural Land lost to Urban Growth in India (2001-2010).  Figure 3. Relationship between agricultural land loss and state level GDP growth.