Detection and analysis of changes in clear-cut harvest patterns using remote sensing and GIS

clear cut;clear-cut;clearcut;remote sensing;isodata;gis;forestry;geographic information systems

Bailey, Andrew Daniel ; Dr. Stacy A. C. Nelson, Committee Member,Dr. George R. Hess, Committee Member,Dr. Heather M. Cheshire, Committee Chair,Bailey, Andrew Daniel ; Dr. Stacy A. C. Nelson ; Committee Member ; Dr. George R. Hess ; Committee Member ; Dr. Heather M. Cheshire ; Committee Chair

University:North Carolina State University

关键词: clear cut;    clear-cut;    clearcut;    remote sensing;    isodata;    gis;    forestry;    geographic information systems;   

Others  :  https://repository.lib.ncsu.edu/bitstream/handle/1840.16/330/etd.pdf?sequence=1&isAllowed=y

美国|英语

来源: null

【 摘 要 】

During the last two decades of the 20th century, policies and practices affecting clear-cut harvesting in North Carolina have changed. Increased harvesting levels, voluntary limits on clear-cut size and location within the forest products industry, and the introduction of wood chip mills are potential sources of change in clear-cutting patterns across the landscape. An investigation of potential changes was conducted by mapping clear-cuts and using landscape metrics to quantify landscape pattern change.For a 13,000 square kilometer area in northwestern North Carolina, unsupervised image classification techniques were used to classify multiple Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper (ETM+) datasets collected in 1984 and 2000. General land use and land cover maps were produced in which clear-cut areas had been separated from agricultural and developed cleared areas. Clearcut area was slightly overestimated during this procedure to minimize the chance of misclassifying clear-cut areas. An accuracy assessment was completed using digital and hardcopy aerial photography, finding overall accuracy of 82% for 2000 and 76% for 1984 land use and land cover datasets. Superior data from the Landsat ETM+ sensor accounted for the higher accuracy of the 2000 dataset when compared the dataset compiled for 1984 using lower quality Landsat TM datasets. Potential methods for increasing the accuracy of future studies are discussed. Based on the results of this study, it is likely that further refinement of the detection technique could lead to greater clear-cut detection accuracy.Several metrics were calculated on the 1984 and 2000 datasets to estimate landscape change in clear-cut harvest patterns in response to changing practices and policies. Over the study time period, the amount of clear-cut area detected increased. This change is coincident with an increase in the number of wood chip mills in and around the study area. Most of the increase in area is attributable to a large increase in the number of small clear-cut patches, while large clear-cut patches occurred less frequently in 2000 than in 1984. This change is coincident with the development of policies designed to restrict the size of individual clear-cut patches. Also, more clear-cut patches were detected in mountainous areas in 2000 than in 1984. Suspected reasons for this change include improved transportation infrastructure into high elevation areas and changing tree species needs by wood products companies. Distances from clear-cut patch to chip mill increased during the time period, indicating that the arrival of chip mill technology at the beginning of the study time period has not resulted in clear-cuts that are located closer to mill sites. Measurements of forest patch shape showed no significant change between 1984 and 2000 though visual analysis of the classified land cover maps suggests that changes have occurred. Potentially more useful approaches to measure effects of policy and practice change on forest patch shape are discussed.