Chad Babcock - Remote Sensing and Geostatistics
Climate change is one of the largest challenges society faces today. Changing climates will affect the spectrum of managed and natural ecosystem services upon which we depend, e.g., food production/safety, forest systems, and water quality/quantity. Degradation and loss of these sustaining ecosystem services can ultimately affect regional economic viability and public health. As the reality of climate change becomes increasingly apparent, environmental scientists, managers, and policy makers need information to define strategies to manage and create ecosystems that can adapt to new climatic conditions.
My research focuses on the development of spatial temporal maps of several meteorologic and hydrologic drought indicators, e.g., Palmer Drought Severity Index, Standardized Precipitation Index and others. Using coupled historic weather station and regional climate model data, I will develop a series of models to predict monthly drought indexes from present to the year 2070 for the contiguous United States. The maps created can assist policy makers and land managers in the development of approaches to adjust to the earth's ever-changing climate.
Mark Devisser - Spatial Modeling
My general interests lie in biogeography, spatial ecology, and dynamic modeling. During my undergraduate career at Portland State University I focused on ecology, biogeography, and geospatial technologies. Since matriculation at Michigan State University, I continued ecological research and have expanded my understanding and use of geographic information science (GIScience) and remote sensing in my course work, masters thesis, and current research endeavors under the expert tutelage of Dr. Joseph P. Messina.
At present my research focuses on tsetse fly distributions in Kenya. Tsetse are the primary vector for African trypanosomiasis, a neglected tropical disease that affects both humans and livestock throughout Sub-Saharan Africa. Several factors can lead to an expansion or contraction of tsetse spatio-temporal distributions; including the introduction of non-native plant species. One such plant species theorized to impact the amount of suitable tsetse land cover is Prosopis juliflora. My research aims to analyze the suitability and impact of Prosopis juliflora with regards to tsetse habitat in Kenya using synthetic aperture radar (SAR) data, LANDSAT data, and in situ field observations. The Tsetse Ecological Distribution (TED) Model, which was developed as part of my masters thesis, will then be used to predict the effect of Prosopis juliflora invasion on tsetse distributions.
Helen Enander - Medical Geography
My interests lie in the areas of Geographic Information Science, spatial analysis, and modeling. I work as a GIS Analyst for Michigan Natural Features Inventory in the area of biodiversity 
conservation. A current research project involves developing methods for using NEXRAD weather radar to quantify bird migration concentration areas in relation to Michigan's shoreline. These dataare useful for identifying those areas of potential high risk in the development of wind energy resources. Other work has centered on questions of how to represent and understand processes and systems given the spatial information that is available. Results include models to identify andpredict high-quality biodiversity areas, and prioritize areas for conservation.
I have also worked on applications of health geography: analyzing adverse reproductive outcomes, refining approaches to determining hospital bed need in Michigan, and looking at racial differences in access to inpatient care. In the area of physical geography I've contributed to two seminar class projects looking first at scale issues in the soil landscape, and secondly, creating a new physiographic map of Michigan (publication in progress).
Shaun Langley - GIS and Remote Sensing
Broadly, my dissertation research addresses the adaptation of new technologies for research. Specifically, interested in the development of database systems for the management of large volumes of spatial data and how these systems can be used to simplify the dissemination of research and collaboration with other institutions. Finally, I am interested in developing software for the next generation of mobile devices, for use in a research context, that greatly enhances the quality and scope of data being collected.
Siam Lawawirojwong - Remote Sensing
Improving the accuracy and reducing the time required for image classification and change detection have long been the goal of a number of remote sensing research. In order to attain this goal, many approaches of optimizations were developed. In light of many limitations and accuracy for the improvement of image classification, the main objective of this research is to develop Artificial Neural Network (ANN) to enhance the accuracy of the satellite image classification and automated change detection for agricultural crop monitoring. Specifically, ANN produces agricultural crop maps of Southeast Asia applying MODIS data. In addition to spectral values, this research will utilize the phenological information as the training dataset for the learning process of ANN which makes this method obviously distinguish from currently used methods. Consequently, the result produced by ANN classification based on phenological models is compared to the result produced by currently used method in terms of accuracy. Finally, this study proposes ANN classification based on the phenological information which is able to extract texture information for image classification. This research shows that the enhanced ANN algorithm is very accurate and significant alternative for regional scale agricultural crop classification and change detection.
Xue (Michelle) Li - GIS and LUCC
With improving understanding of human-environment systems as well as advancing computing technology, modeling of complex system dynamics, such as Land Use and Cover Change (LUCC), is becoming increasingly better defined and sophisticated. However, a fundamental problem within these modeling processes, the relatively low quality of input data, such as land cover map, seldom receives enough attention. Even if the model accurately represents the mechanism of the system, errors and uncertainties within the input data will propagate and probably produce drastic effects on the simulation results. Due to the limitation of current technology, the quality of certain data set is difficult to improve at present. An alternative approach is to incorporate our understanding of errors and uncertainty into the process of complex system modeling, in order to acquire more knowledge on their effect and thus develop appropriate confidence intervals of the modeling result.
This research will looking at potential LUCC in Urumqi area, China, for the near-term (2021-2030). The main objectives of this research are: 1) measuring the propagation pattern of uncertainty from input data to the result, identifying confidence intervals as well as confidence regions accordingly within the study area; 2) discuss some most likely future scenarios of LUCC for Urumqi based on the acquired knowledge of uncertainty. A well defined and widely used LUCC model, dyna-CLUE, will be employed as the core approach of future LUCC simulation. An uncertainty model based on geostatistical technology will be embedded, and Monte Carlo simulation will be applied to test the effect of input data uncertainty on the result of future LUCC scenarios. The outcome of this study will improve the reliability of knowledge we draw from traditional LUCC models. Some socio-environmental adaptation recommendations will then be proposed accordingly to assist policy-making.
Glen O'Neil - GIS
As the most important natural resource in the Great Lakes Basin, preserving and protecting water quality is essential to the health of the region's ecosystem and economy. Pollutant loading from surface runoff to streams and lakes is arguably the single greatest threat to water quality in the region. Rainfall runoff carries pesticides, fertilizers, and manure from farm lands into nearby tributaries, feeding the growth of oxygen-depleting algae and subsequent dead zones where aquatic life is virtually absent (as seen in Lake Erie). The increase in surface imperviousness from urbanization reduces filtration and increases the flashiness of river flows in storm events. These stronger currents erode stream banks, adding to the sun-blocking sediments that starve fish 
populations of the nutrients they need to survive. Conservation practices, such as grass filter strips, conservation tillage, wetlands, and low impact development (LID) can mitigate these threats. However, in order to realize the maximum improvement in water quality these practices must be targeted on the locations that cause the majority of the problem.
My research is focused on developing and evaluating models that optimize the locations of water quality related conservation practices. I am interested in evaluating spatial uncertainty in watershed-scale models of sediment and nutrient loading. I am also interested in measuring uncertainty at field scale models of surface hydrology. For my Master's thesis I developed a method for quantifying error in DEM-derived flow-direction rasters. This effort required the development of methods to extract stream features from LiDAR DEMs. I would like to extend this uncertainty analysis by refining these stream-feature extraction methods and measuring how flow-direction error changes with scale. Better understanding the uncertainty in these water-driven models will help inform conservation technicians and planners as they decide which models to employ, and will ultimately lead to refined models that can better target the installation of conservation practices.
Nick Perdue - Dasymetric Mapping Techniques of Urban Spaces
My research is focused on understanding population dynamics, movements, and living conditions in urban spaces. Current population densities of cities, a measure of the expierenced crowdedness or spaciousness of populations, is done by calculating population per ground area. In the urban environment, density should be measured with consideration of the vertical landscape of residential 
buildings and how that changes traditional population density measurements. A vertical measurement of density is a more accurate measure to represent how people live in buildings within a city. Chicago is the city of focus in this research. Chicago has a large population in a relatively small geographic space with neighborhoods that vary greatly in terms of size and population making it ideal to develop and test this vertical population density method. A dasymetric approach to population mapping is used in the study- the number of people in each census space are allocated to buildings with a defined vertical space and residential designation. With this method an index of personal space is created from which socioeconomic variables can be correlated with a numerical value of living space and living condition. This method redefines the way in which we look at the spatial distribution of the urban population. I am also interested in the cartographic representations of urban population, the limitations and advantages of various methods that are used to represent density and distributions.
Chaun Qin - GIS and Remote Sensing
My research is focused on the question of how the phenology has changed in East Africa due to global climate change. The consequences of global climate change, such as rising atmospheric CO2 level, increasing temperature, and changes in precipitation have major impacts on the ecosystems, especially affecting plants. At the ecosystem level, the species composition, diversity, and geographical locations are all being impacted. At the species level, pla
nt phenology, the seasonal life cycle events of plants, is among the first responses of plants to climate change. Plant phenology is crucial to survival and reproduction of plants. However, the life stages of plants can be regulated by seasonal climate changes, including variations in temperature, precipitation, and photoperiod. Investigating how the phenology is involved with the seasonal climatic changes not only offers evidence of climate change happening now but also helps in assessment of the potential impact on plants in future. Currently, most related studies were conducted in mid-latitude areas. Research specifically done in East Africa is very necessary as East Africa is already threatened by climatic changes and it is among the most vulnerable regions because of the lack of economic, development, and institutional capacity. More importantly, previous research has not make efforts to investigate the drivers of these phenological changes. Thus, the overall objective of this study is to investigate the phenological changes in East Africa using remotely sensed data and to explore the causes of these kinds of phenological changes. Three research questions were addressed in the research:
1) What are the spatial variations of phenological variables over the study area?
2) How has the phenology changed at individual pixel from 1982 to 2006? What are the spatial variations of these phenological changes over East Africa?
3) What are the possible drivers of the phenological changes at some hotspots, considering the climatic and anthropogenic factors?
Tanita Suepa - Remote Sensing, GIS, and Cartography
The temporal and spatial characteristics are considered major advantages of remote sensing approaches to investigate biophysical characteristics. Phenological properties, such as leaf unfolding, first bloom, and leaf fall, which are influenced by the environment and climate, have become the emerging indicators of landscape and global environmental changes. Southeast Asia (SEA) is one of the world's most vulnerable regions affected by the impacts of monsoon climate changes and this region has a complex annual cycle with several growing seasons during a year. It is critical to provide strong scientific evidence that would be the potential information to support the study of these significant changes. Strong relationships between remote sensing products and vegetation phenology enable the vegetation analysis of the process and the pattern of environmental changes in SEA through time and space. 
In this research, MODIS (16-day EVI, 250 km.) from 2000-2009 is applied to extract phenological parameters to study annual variations. The EVI profile and phenological parameters are analyzed to represent regional vegetation dynamics and long term distribution in light of climate and land cover change. Climate factors are investigated with phenological patterns. Additionally, map animations are created to visualize geographical distribution and spatio-temporal patterns of the changes.
The result of this research provides information on the spatial distribution and the temporal trends of vegetation changes, possibly due to human activities and climate change. The findings also indicate interannual variations of phenological parameters, which correspond to land cover and climate variability. The exploration of vegetation time-series, spatial-temporal patterns, and trends of vegetation phenology in SEA will be beneficial to the perception of the global climate change system.
Ying Tang - GIS
My research interest lies in land use and land cover change, climate change, urbanization and carbon storage. China has experienced rapid population and economy growth in the past decades. Much research concerning this growth has been conducted using remote sensing data. However, data sources are limited and expensive in China. New approaches to generate quantitative measuring methods in a data-sparse environment are significant. My current work is trying to estimate gross primary product (GPP) from the 1970s to present with light use efficiency methods using public-accessible Landsat data, MODIS EVI products, MODIS land use land cover products. Final results will be compared to MODIS GPP which doesn't take urban area GPP into account. Urban expansion will also be quantified using remote sensing data. The relationship of urbanization and its carbon consequence will be reported. I'm also interested in accuracy assessment based on GIS. One of my previous projects was concerning soil erosion accuracy assessment with RUSLE model based on GIS. In this study, ASTER, SRTM and DLG data was resampled and used to calculate soil erosion. Differences of the results gained from three types of DEM data were examined. I'm currently a second-year master student. Upon graduation, I'll continue for a PhD degree in geography with my primary interest in global change.