top of page
Lu Liang (梁璐)
Assistant Professor
Characterizing the spatio-temporal pattern of forest disturbance and simulating future landscape scenarios
RESEARCH INTERESTS
Understanding disease ecology in a spatial framework
I integrated an array of cross-disciplinary methods, including GIS, remote sensing, phylogenetic analysis, multivariate analysis, ecological niche modeling and landscape genetics to 1) investigate the spatio-temporal pattern of avian flu spread and the underlying interactions with bird migration pathway and poultry transportation network at the global scale; 2) highlight environmental factors that explain the genetic divergence pattern of Oncomelania hupensis; and 3) simulate and predict the distribution of schistosomiasis with changing climate.
Some finished and ongoing projects include 30-m resolution global land cover mapping, time-series wetland and urban built-up area mapping in China, and forest disturbance mapping in the Southern Rocky Mountains. The classification tasks were ran on the supercomputer at Tsinghua with a parallelization scheme. Uncertainties assessment was conducted to monitor error propagation in hierarchical assessment, and to infer meaningful signals in landscape processes that contain noise.
An automatic remote sensing classification workflow was designed to depict the time-series forest disturbance pattern and the associated disturbance-succession pathways in the Southern Rocky Mountain Ecoregion over the past decade. The landscape-level drivers that explain the disturbance dynamic patterns were investigated and a spatial explicit model was then built to simulate future landscape patterns under various climate scenarios.
Just a sample of my work. To see more or discuss possible work >>
Such as literatures on mapping and infectious disease, historical climate event chronicles, volunteer-based bird watching, and schistosomiasis distribution atlas, to facilitate geospatial data (geodata) mining and knowledge discovery.
Multi-temporal land cover and land use (LULC) mapping at large-scale to depict dynamic landscape processes
Geoinformation retrieval from non-spatial unstructured data
bottom of page