Research

Our research focuses on understanding land use/land cover change, forest/agriculture ecosystem dynamics, and human-environment systems through the integrated use of geospatial information technologies (GIS, remote sensing, GPS) and spatial statistical methods.

Major Research Projects:

(1) Land use land cover change, climate change, and policy

An early paper (Jiang and Zhou 2006), funded by the National Natural Science Foundation of China, examined urban sprawl of Beijing using remote sensing, GIS, and fractal theory. This study finds that Beijing has experienced continuing decentralization, leapfrog development, and segregation of land use. Later, Jiang received a research grant from Office of International Affairs, Ohio State University, to revisit this question by employing a new method to estimate a compactness index of urban form. The study, reported in Jiang and Liu (2012), shows that the central city of Beijing is highly compact, confirming several findings by others, but also shows that the extent and magnitude of urban sprawl and compactness have significant impact on urban environment including especially urban heat island and local meteorology. Both studies relate to policy decisions for urban planning and land management.

In joint work with Wainwright, Liu, and Mercer, we conduct a thorough analysis of forest/agriculture use by Mayans in Southern Belize (Wainwright, Jiang and Liu 2013; Wainwright, Jiang, Mercer and Liu 2015; Jiang, Liu, Wainwright, and Mercer, in preparation). This pertains to the long dispute between Maya communities and the Government of Belize as to Maya's indigenous title to forest lands from which they produce a livelihood. A central element in this contention is that Maya communities destroy the forest through "slash and burn" farming. Encouraged by the Maya Leaders Alliance, we analyze forest change in Southern Belize using remote sensing and GIS. Our work reveals contrasting results: (i) For areas where customary Maya land use and tenure have been consistently practiced since the colonial era, forests are relatively stable; (ii) For areas where customary Maya land use practices are no longer the norm, significant deforestation is observed. This work suggests that the popular association of "slash and burn farming" with "deforestation", or the so-called "Maya deforestation", may not be true. A broader implication is that integration of political ecology (PE) and land change science (LCS) can lead to "policy adoptable" results.

(2) Spatial data quality

An important technical issue in land use land cover change research is spatial data quality. As land use maps are usually derived from remote sensing image classification, the question turns to the spatial quality of image classification. We develop methods for solving this problem, showing that class membership probabilities can be used to represent spatial data quality.

Methods on spatial classification quality can be grouped into into three types: regression-based, interpolation-based, and score-based. Among these, the score-based method is more often used because it has fewer limitations than the others. Under this approach, class membership probabilities (i.e., scores) output from a classification method are used to represent spatial data quality at the pixel level.

(3) Urban studies: multiple angles

Currently, we are working on three projects related to cities and urbanization (stay tuned for detailed results):

     (i)   Urban crime
     (ii)  Housing abandonment
     (iii) Urban climate

(4) Penology: Global warming on forest change

This is a new project in collaboration with the labs of Drs Lapenas and Buyantuev.

This project is partially funded by U.S. National Science Foundation (NSF) Major Research Instrument (MRI) grant. We obtained an Unmanned Aerial System (UAS) throught this grant.

Also check out our Phonocam installed at the Huyck Preserve and Biological Research Station.