Mark is interested in the application of multi-resolution optical remote sensing data to quantify humid tropical forest cover loss via "smart sampling" and "wall-to-wall mapping". He finished his Ph.D. in Geographic Information Science & Engineering with Dr. Matthew Hansen in Dec 2010.
Research question: How can Landsat and MODIS be utilized to derive precise, spatially and temporally consistent, and policy relevant information on forest cover loss over the persistently cloudy humid tropics?
Project in a Nutshell:
# Spatially concentrated deforestation may be missed by a small random sample of Landsat data leading to imprecise estimates. Precise estimates can be archived by using either a larger sample size or a MODIS-targeted sampling design.
# Detailed mapping of humid tropical forest is complicated by persistent clouds. Epochal Landsat composites are valuable for gap-free forest cover mapping.
# Deforestation mapping over persistently cloudy areas is a frontier. Analyzing per Landsat pixel trajectories of all cloud-free observations over a decade provides robust and frequent deforestation maps.
Project frame: Mark's research project directly supports two NASA Land Cover Land Use Change grants: "Producing Composite Imagery and Forest Cover and Change Characterizations for the Humid Tropics" and “Establishing a global forest monitoring capability using multi-resolution and multi-temporal remotely sensed data sets” (Hansen - PI).
Publications: Mark Broich, Stephen V. Stehman, Matthew C. Hansen, Peter Potapov, Yosio E. Shimabukuro, A comparison of sampling designs for estimating deforestation from Landsat imagery: A case study of the Brazilian Legal Amazon, Remote Sensing of Environment, In Press, DOI: 10.1016/j.rse.2009.07.011.
Presentations: Mark presented initial results at the AGU Fall meeting 2007 in San Francisco (Poster), at the XIV Brazilian Remote Sensing Symposium (Fig. 1-2, Conference proceedings), at the Brazilian National Institute for Space Research (INPE) [Fig. 3-6], and at the Indonesian National Institute of Aeronautics and Space (LAPAN) [Fig. 7].
Capacity building: Mark also held a forest remote sensing workshop for members of the Indonesian SEKALA NGO network. He further contributed to Australia's Overseas Aid Program (AUSAID) International Forest Carbon Initiative by providing a capacity building training at the Indonesian National Institute of Aeronautics and Space (LAPAN) to strengthen remote sensing expertise for forest monitoring (Fig. 8). Both the workshop and the training focused on the use of epochal Landsat image composites (e.g. Fig. 4-6) to monitor the loss of Intact Forest Landscapes (IFL).
Previous experience: Mark’s previous remote sensing experience includes the investigation of Leaf Area Index patterns in the Kakamega Forest (Kenya) based on field and Landsat data. His work was part of a joint effort of an international biodiversity research project (BIOTA Africa) and the German Remote Sensing Data Center (DLR-DFD). Prior work also includes geological mapping based on multiple remotely sensed datasets during a graduate student exchange program with the School of Ocean and Earth Science and Technology at the University of Hawaii at Manoa. Mark holds a M.A. degree in Geography with minors in Geology and Environmental Engineering from the Technical University Aachen (RWTH-Aachen), Germany.