@article{lin_sills_cheshire_2014, title={Targeting areas for Reducing Emissions from Deforestation and forest Degradation (REDD+) projects in Tanzania}, volume={24}, ISSN={["1872-9495"]}, DOI={10.1016/j.gloenvcha.2013.12.003}, abstractNote={Reducing Emissions from Deforestation and forest Degradation (REDD+) has gained momentum as a climate mitigation strategy that can be implemented at multiple scales. Sub-nationally, REDD+ projects that aim to capture carbon funding are implemented throughout tropical countries. A spatial targeting approach for optimal REDD+ project landscape is demonstrated using Tanzania as an example. This study used GIS-based Multi-criteria Decision Analysis to identify potential areas for REDD+ projects development incorporating different combinations of criteria. The first approach, efficient targeting, focuses on areas with high forest carbon content, high deforestation risk and low opportunity cost. The second approach, co-benefits targeting, aims at areas with high biodiversity and high poverty rate on top of criteria in efficient targeting. The resulting suitability maps displays areas of high, medium and low suitability for future REDD+ projects development based on the targeting approaches. Locations of current REDD+ projects in Tanzania were also overlaid with suitability map to visually inspect how they match up. This approach allows decision-makers to prioritize preferences for various site-selection criteria and make informed decisions about REDD+ projects locations.}, journal={GLOBAL ENVIRONMENTAL CHANGE-HUMAN AND POLICY DIMENSIONS}, author={Lin, Liwei and Sills, Erin and Cheshire, Heather}, year={2014}, month={Jan}, pages={277–286} }
 @article{carr_cheshire_hess_bailey_devine_2011, title={Assessing embedded geospatial student learning outcomes in forestry and natural resources curricula}, volume={109}, number={7}, journal={Journal of Forestry}, author={Carr, J. D. and Cheshire, H. M. and Hess, G. R. and Bailey, D. and Devine, H. A.}, year={2011}, pages={409–416} }
 @article{hester_nelson_cakir_khorram_cheshire_2010, title={High-resolution land cover change detection based on fuzzy uncertainty analysis and change reasoning}, volume={31}, ISSN={0143-1161 1366-5901}, url={http://dx.doi.org/10.1080/01431160902893493}, DOI={10.1080/01431160902893493}, abstractNote={Land cover change detection is an important research and application area for analysts of remote sensing data. The primary objective of the research described here was to develop a change detection method capable of accommodating spatial and classification uncertainty in generating an accurate map of land cover change using high resolution satellite imagery. As a secondary objective, this method was designed to facilitate the mapping of particular types and locations of change based on specific study goals. Urban land cover change pertinent to surface water quality in Raleigh, North Carolina, was assessed using land cover classifications derived from pan-sharpened, 0.61 m QuickBird images from 2002 and 2005. Post-classification map errors were evaluated using a fuzzy logic approach. First, a ‘change index’ representing a quantitative gradient along which land cover change is characterized by both certainty and relevance, was created. The result was a continuous representation of change, a product type that retains more information and flexibility than discrete maps of change. Finally, fuzzy logic and change reasoning results were integrated into a binary change/no change map that quantified the most certain, likely, and relevant change regions within the study area. A ‘from-to’ change map was developed from this binary map inserting the type of change identified in the raw post-classification map. A from-to change map had an overall accuracy of 78.9% (κ = 0.747) and effectively mapped land cover changes posing a threat to water quality, including increases in impervious surface. This work presents an efficient fuzzy framework for transforming map uncertainty into accurate and practical change analysis.}, number={2}, journal={International Journal of Remote Sensing}, publisher={Informa UK Limited}, author={Hester, D. B. and Nelson, S. A. C. and Cakir, H. I. and Khorram, S. and Cheshire, H.}, year={2010}, month={Jan}, pages={455–475} }
 @article{koch_cheshire_devine_2006, title={Landscape-scale prediction of hemlock woolly adelgid, Adelges tsugae (Homoptera : Adelgidae), infestation in the southern Appalachian Mountains}, volume={35}, ISSN={["1938-2936"]}, DOI={10.1603/0046-225X(2006)35[1313:LPOHWA]2.0.CO;2}, abstractNote={Abstract After causing substantial mortality in the northeastern and mid-Atlantic United States, the hemlock woolly adelgid, Adelges tsugae Annand (Homoptera: Adelgidae), has recently invaded the southern Appalachian region. Although general estimates of regional spread exist, the landscape-level dynamics of A. tsugae invasion are poorly understood—particularly factors predicting where the pest is likely to first infest a landscape. We examined first-year infestation locations from Great Smoky Mountains National Park and the Blue Ridge Parkway to identify possible factors. For 84 infested and 67 uninfested sites, we calculated values for a suite of variables using a geographic information system. After identifying significant variables, we applied four statistical techniques—discriminant analysis, k-nearest neighbor analysis, logistic regression, and decision trees—to derive classification functions separating the infested and uninfested groups. We used the resulting functions to generate maps of A. tsugae infestation risk in the Great Smoky Mountains. Three proximity variables (distance to the closest stream, trail, and road) appeared in all four classification functions, which performed well in terms of error rate. Discriminant analysis was the most accurate and efficient technique, but logistic regression best balanced accuracy, efficiency, and ease of use. Our results suggest that roads, major trails, and riparian corridors provide connectivity enabling long-distance dispersal of A. tsugae, probably by humans or birds. The derived classification functions can yield A. tsugae infestation risk maps for elsewhere in the southern Appalachian region, allowing forest managers to better target control efforts.}, number={5}, journal={ENVIRONMENTAL ENTOMOLOGY}, author={Koch, F. H. and Cheshire, H. M. and Devine, H. A.}, year={2006}, month={Oct}, pages={1313–1323} }
 @article{richards_apperson_ghosh_cheshire_zeichner_2006, title={Spatial analysis of Aedes albopictus (Diptera : Culicidae) oviposition in suburban neighborhoods of a piedmont community in North Carolina}, volume={43}, ISSN={["0022-2585"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-33750941447&partnerID=MN8TOARS}, DOI={10.1603/0022-2585(2006)43[976:SAOAAD]2.0.CO;2}, abstractNote={Abstract Temporal and spatial distribution of egg-laying by Aedes albopictus (Skuse) (Diptera: Culicidae) was investigated in suburban neighborhoods in Raleigh, NC, by using oviposition traps (ovitraps) at fixed sampling stations during the 2002 and 2003 mosquito seasons. Variations in the phenology of oviposition between the two mosquito seasons resulted from differences in the patterns and amounts of rainfall early in the season. Aerial images of each study neighborhood were digitized, and the proportions of specific types of land cover within buffer zones encompassing ovitraps were estimated. Retrospective analyses showed that in some neighborhoods, oviposition intensity was significantly associated with specific types of land cover. However, in general, it seemed that gravid Ae. albopictus searched throughout the landscape for water-filled containers in which to lay eggs. Peridomestic surveys were carried out concurrently with ovitrap collections to estimate production of Ae. albopictus pupae in discarded water-filled containers and the abundance of females in vegetation that made up the resting habitat. Results of linear regression analyses indicated that the mean standing crop of pupae (total and per container) per residence was not a significant predictor of mean egg densities in ovitraps. However, the mean standing crop of adult females was a significant but weak predictor variable, because the magnitude and sign of regression coefficients varied between neighborhoods. Linear spatial regression analyses revealed that oviposition intensity was not spatially dependent on pupal standing crop or the numbers of pupae-positive containers distributed peridomestically. However, a weak spatial dependence on the standing crop of adult females was found in some neighborhoods. Based on spherical variogram models, kriging was carried out to predict the spatial patterns of oviposition in suburban neighborhoods. Focal areas of high and low oviposition intensity were evident in most neighborhoods; however, the spatial patterns of oviposition changed between mosquito seasons. Kriging predictions were evaluated, using cross-validation, by systematically removing each data point from our data set and predicting the removed point by using the remaining points. The root mean square (standardized) error values of best fitting variogram models approximated 1, and plots of standardized PRESS residuals showed no distinct pattern for most neighborhoods, indicating that predictions of the spatial distribution of oviposition intensity were valid. Spherical variogram models are a satisfactory method for describing the spatial distribution of Ae. albopictus oviposition, and kriging can be a useful technique for predicting oviposition intensity at locations that have not been sampled.}, number={5}, journal={JOURNAL OF MEDICAL ENTOMOLOGY}, author={Richards, Stephanie L. and Apperson, Charles S. and Ghosh, Sujit K. and Cheshire, Heather M. and Zeichner, Brian C.}, year={2006}, month={Sep}, pages={976–989} }
 @article{flores_allen_cheshire_davis_fuentes_kelting_2006, title={Using multispectral satellite imagery to estimate leaf area and response to silvicultural treatments in loblolly pine stands}, volume={36}, ISSN={["0045-5067"]}, DOI={10.1139/X06-030}, abstractNote={ The relationship between leaf area index (LAI) of loblolly pine plantations and the broadband simple ratio (SR) vegetation index calculated from Landsat 7 Enhanced Thematic Mapper Plus (ETM+) data was examined. An equation was derived to estimate LAI from readily available Landsat 7 ETM+ data. The equation developed to predict LAI with Landsat 7 ETM+ data was tested with ground LAI measurements taken in 12 plots. The root mean square error of prediction was 0.29, an error of approximately 14% in prediction. The ability of Landsat 7 ETM+ data to consistently estimate SR over time was tested using two scenes acquired on different dates during the winter (December to early March). Comparison between the two images on a pixel-by-pixel basis showed that approximately 96% of the pixels had a difference of <0.5 units of SR (approximately 0.3 units of LAI). When the comparison was made on a stand-by-stand basis (average stand SR), a maximum difference of 0.2 units of SR (approximately 0.12 units of LAI) was found. These results suggest that stand LAI of loblolly pine plantations can be accurately estimated from readily available remote sensing data and provide an opportunity to apply the findings from ecophysiological studies in field plots to forest management decisions at an operational scale. }, number={6}, journal={CANADIAN JOURNAL OF FOREST RESEARCH-REVUE CANADIENNE DE RECHERCHE FORESTIERE}, author={Flores, FJ and Allen, HL and Cheshire, HM and Davis, JM and Fuentes, M and Kelting, D}, year={2006}, month={Jun}, pages={1587–1596} }
 @article{hess_cheshire_2002, title={Spatial information technologies: Integrating the tools with the curricula}, volume={100}, ISBN={0022-1201}, number={1}, journal={Journal of Forestry}, author={Hess, G. R. and Cheshire, H. M.}, year={2002}, pages={29} }
 @article{langley_cheshire_humes_2001, title={A comparison of single date and multitemporal satellite image classifications in a semi-arid grassland}, volume={49}, ISSN={["1095-922X"]}, DOI={10.1006/jare.2000.0771}, abstractNote={Landsat Thematic Mapper (TM) satellite data were used to produce maps depicting ranges of major vegetation types at the Jornada Experimental Range, New Mexico, U.S.A. Single date and multitemporal classification accuracies were compared using vegetation ground data as references. Single date image classifications were more accurate than multitemporal images for mapping land cover types in this region. Use of single date imagery generally involves less expenditure of time and costs related to data acquisition and processing. Multitemporal images have improved classification accuracies in some landscapes; however, single date images may provide a reliable method for mapping vegetation cover in semi-arid environments.}, number={2}, journal={JOURNAL OF ARID ENVIRONMENTS}, author={Langley, SK and Cheshire, HM and Humes, KS}, year={2001}, month={Oct}, pages={401–411} }