The use of high and medium resolution imagery to detect agricultural land cover in Chinese Cities: A case study of Nanjing-2000 to 2015

Authors: Erik Breidinger*, Western Michigan University
Topics: China, Remote Sensing, Land Use and Land Cover Change
Keywords: Remote sensing, agriculture, China, land cover/land use
Session Type: Paper
Day: 4/5/2019
Start / End Time: 5:00 PM / 6:40 PM
Room: Roosevelt 4.5, Marriott, Exhibition Level
Presentation File: No File Uploaded


According to McGee and Ginsburg’s desakota hypothesis, rapidly growing Asian cities differ significantly from large Western cities in their land cover/land use (LC/LU) as they retain a significant portion of agricultural land and labor despite rapid urbanization. However, significant amounts of agricultural production within desakotas takes place in plastic sheeting and glass greenhouses, causing a unique problem when calculating LC/LU estimates via traditional remote sensing techniques using ArcGIS 10.6. While greenhouses appear equivalent to developed land spectrally, their purpose is entirely agricultural. This study provides an improved method of calculating greenhouse land-cover as agricultural land-use with minimal manual editing in LC/LU analysis using the Jiangning district of the city of Nanjing as the study area of the desakota hypothesis. Satellite images from Landsat 8 Operational Land Imager (OLI) and Landsats 4 and 5 Thematic Mapper (TM) of Nanjing at 30-meter pixel resolution are analyzed for the years 2000, 2010, and 2016, and also incorporating an additional 2-meter pixel resolution image from Wolrdviewer-2 for 2016 for testing accuracy. Additional ground truthing and training points were collected on-site in Nanjing in May 2018. Synthesizing high- and medium-resolution images into per-pixel, sub-pixel, and object-based classification techniques achieved a marginal increase in greenhouse areas incorporated into estimates of arable land

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login