Daniel Unger

Estimating number of trees, tree height and crown width using Lidar data

Estimating tree characteristics with field plots located in remote and inaccessible areas can be a costly and timely endeavor. Light Detection and Ranging (Lidar) remote sensing allowing for the estimation of the 3-dimensional structure of forest vegetation offers an alternative to traditional ground based forest measurements. This project assessed the utility of using Lidar data to estimate number of trees, tree height and crown width within Barksdale Air Force Base forest management area, Bossier City, Louisiana. Two programs, Lidar Data Filtering and Forest Studies (Tiffs) and Lidar Analyst were used to derive forest measurements, which were compared to field measurements. Based on Root Mean Square Error (RMSE), Lidar Analyst (3.81 trees) performed better than Tiffs (5.71 trees) at estimating average tree count per plot. Tiffs was better at deriving average tree height than Lidar Analyst with an RMSE of 19.08 feet to Lidar Analyst’s RMSE of 21.20 feet. Lidar Analyst, with a RMSE of 25.41 feet, was better in deriving average crown diameter over Tiffs RMSE of 30.54 feet. All linear correlation coefficients between average field measured tree height and Lidar derived average tree height were highly significant at the 0.01 probability level for both Tiffs and Lidar Analyst on hardwood, conifers and a combined hardwood-conifer comparison.

Fuel loading prediction models developed from aerial photographs of the Sangre de Cristo and Jemez mountains of New Mexico, USA

Fuel load prediction equations that made use of aerial photographs were developed for Mixed Conifer, Ponderosa Pine (Pinus ponderosa Dougl. ex Laws.) and Pinyon–Juniper (Pinus edulis Engelm.)–(Juniperus monosperma Engelm.) cover types from one-time measurements made in the Santa Fe watershed (SFWS) located in the Sangre de Cristo Mountains of northern New Mexico, and at Los Alamos National Laboratory (LANL) located in the Jemez Mountains of northern New Mexico. The results of the watershed data set were favorable and exhibited a high degree of relative accuracy. The results from the LANL data set did not share the same degree of accuracy, but rather exhibited a high degree of error. Use of these or similar prediction equations may be limited to certain regions and community types that exhibit similar regional characteristics such as terrain, soil, and weather conditions. Applied use of the prediction equations required less time than traditional fuel sampling performed onsite, but suffered from a loss of accuracy. It is strongly suggested that additional study of this method be undertaken to generate more accurate and reliable equations. Hopefully, more accurate equations may augment existing fuel sampling techniques and be put to practical use for fire planning purposes.

Quantifying Land Cover Change Due to Petroleum Exploration and Production in the Haynesville Shale Region Using Remote Sensing

The Haynesville Shale lies under areas of Louisiana and Texas and is one of the largest gas plays in the U.S. Encompassing approximately 2.9 million ha, this area has been subject to intensive exploration for oil and gas, while over 90% of it has traditionally been used for forestry and agriculture. In order to detect the landscape change in the past few decades, Landsat Thematic Mapper (TM) imagery for six years (1984, 1989, 1994, 2000, 2006, and 2011) was acquired. Unsupervised classifications were performed to classify each image into four cover types: agriculture, forest, well pad, and other. Change detection was then conducted between two classified maps of different years for a time series analysis. Finally, landscape metrics were calculated to assess landscape fragmentation. The overall classification accuracy ranged from 84.7% to 88.3%. The total amount of land cover change from 1984 to 2011 was 24%, with 0.9% of agricultural land and 0.4% of forest land changed to well pads. The results of Patch-Per-Unit area (PPU) index indicated that the well pad class was highly fragmented, while agriculture (4.4-8.6 per sq km) consistently showed a higher magnitude of fragmentation than forest (0.8-1.4 per sq km).

Comparing remotely sensed Pictometry® Web-based height estimates with in situ clinometer and laser range finder height estimates

Abstract Heights of 30 light poles were measured with a telescopic height pole. Clinometer and laser range finder in situ estimated light pole height was compared to Pictometry® estimated light pole height using hyperspatial 4-in. (10.2-cm) multispectral imagery within a Web-based interface. Average percent agreement between light pole height and clinometer and laser range finder estimated that light pole height ranged from 3.97% to 3.79% for clinometer and laser range finder estimated light pole height, respectively. Average percent agreement between light pole height and Pictometry® estimated light pole height at image magnification factors of 100%, 125%, 150%, 200%, and 300% magnification ranged from 1.77% to 2.39%. Root-mean-square error (RMSE) between light pole height and clinometer and laser range finder estimated that light pole height ranged from 0.22 to 0.20 m for clinometer and laser range finder estimated light pole height, respectively. RMSE between light pole height and Pictometry® estimated light pole height ranged from 0.10 to 0.14 m. An analysis of variance between absolute errors of light pole height estimate by different techniques indicated that Pictometry® was significantly more accurate than both clinometer and laser range finder light pole height estimates.

Short communication: A GIS tool for plant spatial pattern analysis

A GIS program, ArcPlantPattern, was developed with Visual Basic .NET and ArcObjects as an ArcGIS extension to assist the investigation of plant distribution patterns (species composition as occurrence probability and spacing as distances among species) and to design planting plan maps for patch planting. ArcPlantPattern is the first software of its kind. It can be used for arid and semiarid lands reclamation, burned area rehabilitation, or designing landscapes with a required plant community distribution. ArcPlantPattern may also be applicable to other spatial point pattern analysis, such as geology, geography and wildlife habitat.

Mapping oilfield brine-contaminated sites with mid-spatial resolution remotely sensed data

An environmental problem associated with petroleum production is the disposal of brine, which is produced during petroleum exploration and production. Oilfield brine, if improperly handled, transported, and disposed of, can pose a serious threat to surrounding water resources, arable lands, and plant communities. Although field checking of known oilfield brine-contaminated sites is relatively straightforward, the ability to detect and inventory brine-contaminated sites over remote and expansive areas can be time consuming and expensive. A more efficient and cost-effective method is needed to delineate brine-contaminated sites accurately. The chief aim of this project was to test a remote sensing method to map accurately and quantify contaminated oilfield brine sites in west Texas. Landsat ETM+ data of west Texas were obtained, de-correlated with a three-band dataset using principal component analysis (PCA), and classified into brine and non-brine locations using supervised classification with a maximum likelihood classification algorithm. Results show the Landsat ETM+ data is effective in quantifying previously unknown oilfield brine-contaminated areas larger than 2 acres in west Texas. Overall map accuracy was 91.67%, user’s accuracy was 87.50% for brine-contaminated sites, and the kappa statistic was 82.35%. Once contaminated brine sites have been mapped via remote sensing, the spatial location and quantity of the sites can make land reclamation and restoration decisions more timely and cost-effectively compared to traditional ground surveys.

Validating the geometric accuracy of high spatial resolution multispectral satellite data

Uses of high spatial resolution data obtained from satellite-based sensors include creating land cover maps, deriving large-scale quantitative assessments such as vegetation indices, and visually assessing an area for qualitative information only assessable from large-scale digital data. One of the more popular uses of high spatial resolution data is to use the image as a base map for on-screen digitizing spatially dependent vector products. Since most geographic information system (GIS) databases store a variety of current and historical data, the accuracy of any on-screen digitized product is dependent on the spatial accuracy of the reference data. Therefore, it is important to understand and validate the accuracy of data used to create spatially referenced product, even though the data come with high spatial resolution. One of the more popular and historical high spatial resolution data within most GIS labs is QuickBirds multispectral data at 2.44 × 2.44 m2. Although there are current sensors available with a higher spatial resolution, the sometimes prohibitive expense of obtaining high spatial resolution data necessitates the need to utilize and assess historic data. Since the QuickBird has been the mainstay of high spatial resolution data since 2001, understanding the geometric accuracy of the DigitalGlobes QuickBird user-defined panchromatic and multispectral image bundle product remains relevant. In this study, we assessed the positional accuracy of this product for its utility as an “off the shelf” base map for creating other spatially referenced products. The average Euclidean distance, RMSE (root mean square error), and RSME (root square mean error) between QuickBird-identified Universal Transverse Mercator (UTM) coordinates and coincident in situ GPS-collected UTM coordinates were calculated at 33 systematically selected locations throughout the city of Nacogdoches, Texas, USA. The average Euclidean distance, RMSE, and RSME between QuickBirds projected UTM coordinates and its corresponding GPS-collected UTM coordinates measured at 5.34 meters, 5.79 meters, and 4.05 meters, respectively. They were well within DigitalGlobes stated RMSE positional accuracy of 14.0 meters for a panchromatic and multispectral QuickBird image bundle.

Old Field Communities and Restoration Potential at Mammoth Cave National Park, USA

Abstract Old fields (at least 67 years since abandonment) within Mammoth Cave National Park, USA are dominated by coniferous species (Juniperus virginiana L. and Pinus virginiana) instead of the desired deciduous species (Carya glabra, Quercus alba, Q. muehlenbergii, Q. prinus, and Q. velutina) that dominate much of the rest of the park. Species composition above ground and in the seedbank of old fields and adjacent desired future condition areas, (identified by the United States National Park Service (NPS) as oak and hickory-dominated) were evaluated and compared. Species composition and dominance have shifted from oak species toward conifer-dominated stands due to previous land conversion to agriculture and the exclusion of fire. Management practices that can be implemented by the NPS to alter the condition of the old fields to achieve the desired future condition include thinning treatments and reintroduction of the historic fire regime.