Fine Resolution Probabilistic Land Cover Classification of Landscapes in the Southeastern United States

Abstract

Abstract:Land cover classification provides valuable information for prioritizing management and conservation operations across large landscapes. Current regional scale land cover geospatial products within the United States have a spatial resolution that is too coarse to provide the necessary information for operations at the local and project scales. This paper describes a methodology that uses recent advances in spatial analysis software to create a land cover classification over a large region in the southeastern United States at a fine (1 m) spatial resolution. This methodology used image texture metrics and principle components derived from National Agriculture Imagery Program (NAIP) aerial photographic imagery, visually classified locations, and a softmax neural network model. The model efficiently produced classification surfaces at 1 m resolution across roughly 11.6 million hectares (28.8 million acres) with less than 10% average error in modeled probability. The classification surfaces consist of probability estimates of 13 visually distinct classes for each 1 m cell across the study area. This methodology and the tools used in this study constitute a highly flexible fine resolution land cover classification that can be applied across large extents using standard computer hardware, common and open source software and publicly available imagery.

Using Forest Inventory Data with Landsat 8 Imagery to Map Longleaf Pine Forest Characteristics in Georgia, USA

“…In the final L1 and L2 raster surfaces, extrapolated cells were populated with NULL values using the "Extract Model Domain" tool (Table A3, Figures A3 and A4). While modeled relationships were a significant improvement over classical spatial aggregation techniques such as FIA plot summaries (i.e., Figure 4 standard deviation verses Figure 7 RMSE) and are an improvement over previous longleaf mapping projects that include striping [10,12], there was more variability in the relationships than we anticipated. Sources of variation are discussed in Section 4.…”

“…Strong relationships between forest characteristics measured in field plots and metrics derived from imagery were found in this investigation and this estimation approach significantly reduced error and improved spatial detail over classical aggregation techniques [12]. Despite clear benefits over traditional techniques of classifying forests and estimating forest characteristics using plot data alone, both studies [10,12] encountered problems with using NAIP imagery. Due to the variation in image acquisition dates and the color balancing technique used to mosaic NAIP images together, resulting raster surfaces had increased estimation error and displayed visual striping along flight paths.…”

“…This work leveraged new statistical methods and novel data structures and algorithms that were initially developed for mapping forest characteristics in the western USA [5][6][7][8]. Image texture metrics and principle components derived from National Agriculture Imagery Program (NAIP) aerial photographic imagery [9] were used with a softmax neural network model to produce probabilistic classification surfaces at 1 m resolution across 11.6 million ha in the southeast USA [10]. Species composition, basal area and stand tree density for pine and hardwood tree species, including longleaf pine, were then estimated at coarser resolution (78 m by 70 m) for the study area using random forest methods, Forest Inventory and Analysis (FIA) inventory plot data [11] and image spectral and texture metrics [12].…”

Mitigating the Impact of Field and Image Registration Errors through Spatial Aggregation

“…Ideas such as spatial contiguity, patch size, and patch juxtapositioning, and their relationships to processes and concepts such as forest management, land use planning, and sustainable forestry have in part fueled the desire to precisely and accurately define existing patterns at fine spatial detail, across broad extents [17][18][19]. Coupled with the availability of fine-grained remotely sensed data ( 5 m) and advancements in computer-based hardware and software [20], a fine-scaled depiction of the landscape can now be produced across broad extents relatively quickly, at a low cost [21][22][23]. At the same time, the fine-grain nature of these types of data provide unique opportunities to relate measured characteristics of the landscapes for small spatial extents (response variables) to remotely sensed data (predictor variables).…”

“…Many have capitalized on this point to develop mathematical, statistical, and spatial models that can be used to create surfaces depicting landscape variables of interest using geo-rectified field and remotely sensed data [18,[22][23][24]. Generally, this process can be described as: (1) registering both field and remotely sensed data to a known coordinate system, (2) using the spatial coordinates of the field and remotely sensed data to link measured values in the field to remotely sensed data, (3) building a model for the linked variable as a function of variables derived from the remotely sensed data, and (4) applying the model to remotely sensed surfaces to create a continuous surface of estimated characteristics.…”