National Agriculture Imagery Program Research Articles

A Comparison of Methods for Determining Forest Composition from High-Spatial-Resolution Remotely Sensed Imagery

Remotely sensed imagery has been used to support forest ecology and management for decades. In modern times, the propagation of high-spatial-resolution image analysis techniques and automated workflows have further strengthened this synergy, leading to the inquiry into more complex, local-scale, ecosystem characteristics. To appropriately inform decisions in forestry ecology and management, the most reliable and efficient methods should be adopted. For this reason, our research compares visual interpretation to digital (automated) processing for forest plot composition and individual tree identification. During this investigation, we qualitatively and quantitatively evaluated the process of classifying species groups within complex, mixed-species forests in New England. This analysis included a comparison of three high-resolution remotely sensed imagery sources: Google Earth, National Agriculture Imagery Program (NAIP) imagery, and unmanned aerial system (UAS) imagery. We discovered that, although the level of detail afforded by the UAS imagery spatial resolution (3.02 cm average pixel size) improved the visual interpretation results (7.87–9.59%), the highest thematic accuracy was still only 54.44% for the generalized composition groups. Our qualitative analysis of the uncertainty for visually interpreting different composition classes revealed the persistence of mislabeled hardwood compositions (including an early successional class) and an inability to consistently differentiate between ‘pure’ and ‘mixed’ stands. The results of digitally classifying the same forest compositions produced a higher level of accuracy for both detecting individual trees (93.9%) and labeling them (59.62–70.48%) using machine learning algorithms including classification and regression trees, random forest, and support vector machines. These results indicate that digital, automated, classification produced an increase in overall accuracy of 16.04% over visual interpretation for generalized forest composition classes. Other studies, which incorporate multitemporal, multispectral, or data fusion approaches provide evidence for further widening this gap. Further refinement of the methods for individual tree detection, delineation, and classification should be developed for structurally and compositionally complex forests to supplement the critical deficiency in local-scale forest information around the world.

Changes in the vegetation of New Orleans from the disturbance impacts of Hurricane Katrina in 2005

The catastrophic flooding of New Orleans following landfall of Hurricane Katrina in August of 2005 left behind a unique urban environment for investigating disaster impacts on vegetation disturbance and recovery. The goal of this study was to analyze the changes in urban tree canopy cover (both live standing and downed) and the extent of vegetation as shrubs or herbaceous cover on blocks located within the most flood-impacted neighborhoods in 2005 of New Orleans, down to the scale of individual property lots. Remote sensing aircraft images from the U. S. Geological Survey's Digital Orthophoto Quadrangles (DOQs) imagery and the National Agriculture Imagery Program (NAIP) were used to generate high-resolution (1-m) vegetation cover maps for 2004, 2005 (post-Katrina), and 2019 for deeply flooded neighborhoods from 2005. Segmentation and object-based classification applied to all color infrared (CIR) imagery was used to generate and compare high-resolution maps of large tree canopies, shrubs, and herbaceous cover within property lots. Results showed that the Lower Ninth Ward and the Seventh Ward neighborhoods suffered the highest damage levels to canopy cover from the impacts of Hurricane Katrina, losing around 15% in overall tree cover, and sustaining severe damaged on as much as 75% of the neighborhoods' pre-2005 tree cover. Nevertheless, the Lower Ninth Ward has had the highest rate of regrowth of vegetation cover (as of 2019), mainly owing to unmaintained tall shrub cover that in 2019 commonly exceeded 2 m in height on many of the neighborhood's blocks. On a city-wide scale, New Orleans's urban forest is significantly more fragmented today into small tree canopies than it was in 2004.

AN OPEN-SOURCE CANOPY CLASSIFICATION SYSTEM USING MACHINE-LEARNING TECHNIQUES WITHIN A PYTHON FRAMEWORK

Abstract. Studying deforestation has been an important topic in forestry research. Especially, canopy classification using remotely sensed data plays an essential role in monitoring tree canopy on a large scale. As remote sensing technologies advance, the quality and resolution of satellite imagery have significantly improved. Oftentimes, leveraging high-resolution imagery such as the National Agriculture Imagery Program (NAIP) imagery requires proprietary software. However, the lack of insight into the inner workings of such software and the inability of modifying its code lead many researchers towards open-source solutions. In this research, we introduce CanoClass, an open-source cross-platform canopy classification system written in Python. CanoClass utilizes the Random Forest and Extra Trees algorithms provided by scikit-learn to classify canopy using remote sensing imagery. Based on our benchmark tests, this new canopy classification system was 283 % to 464 % faster than commercial Feature Analyst, but it produced comparable results with a similarity of 87.56 % to 87.62 %.

Mapping Center Pivot Irrigation Fields in South Carolina with Google Earth Engine and the National Agricultural Imagery Program

Aerial images taken during the growing seasons of 2009, 2011, 2013, 2015, and 2017 were visually inspected for evidence of irrigation. Center pivot irrigation was identified by the characteristic shape of the spans and the curved tracks left by the wheels. The author manually delineated a polygon over each agricultural area where signs of irrigation infrastructure were observed. The result is a map of 2,689 polygons covering 146,662 acres in South Carolina. Compared with the United States Department of Agriculture 2017 Census of Agriculture, the sampling results account for over 69% of total irrigated area and over 98% of area irrigated solely by center pivots. Most center pivots covered from 25 to 75 acres, while the largest center pivot extended over 300 acres. These results are an important contribution to the quantification of water use in South Carolina.

Automated detection of individual juniper tree location and forest cover changes using Google Earth Engine

Tree detection is the first step in the appraisal of a forest, especially when the focus is monitoring the growth of tree canopy. The acquisition of annual very high-resolution aerial images by the National Agriculture Imagery Program (NAIP) and their accessibility through Google Earth Engine (GEE) supports the delineation of tree canopies and change over time in a cost and time-effective manner. The objectives of this study are to develop an automated method to detect the crowns of individual western Juniper (Juniperus occidentalis) trees and to assess the change of forest cover from multispectral 1-meter resolution NAIP images collected from 2009 to 2016 in Oregon, USA. The Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), and Ratio Vegetation Index (RVI) were calculated from the NAIP images, in addition to the red-green-blue-near infrared bands. To identify the most suitable approach for individual tree crown identification, we created two training datasets: one considering yearly images separately and one merging all images, irrespective of the year. We segmented individual tree crowns using a random forest algorithm implemented in GEE and seven rasters, namely the reflectance of four spectral bands as recorded by the NAIP images (i.e., the red-green-blue-near infrared) and three calculated indices (i.e., NDVI, NDWI, and RVI). We compared the estimated location of the trees, computed as the centroid of the crown, with the visually identified treetops, which were considered as validation locations. We found that tree location errors were smaller when years were analyzed individually than by merging the years. Measurements of completeness (74%), correctness (94%), and mean accuracy detection (82 %) show promising performance of the random forest algorithm in crown delineation, considering that only four original input bands were used for crown segmentation. The change in the calculated crown area for western juniper follows a sinusoidal curve, with a decrease from 2011 to 2012 and an increase from 2012 to 2014. The proposed approach has the potential to estimate individual tree locations and forest cover area dynamics at broad spatial scales using regularly collected airborne imagery with easy-to-implement methods.

Automated detection of individual Juniper tree location and forest cover changes using Google Earth Engine

Tree detection is the first step in the appraisal of a forest, especially when the focus is monitoring the growth of tree canopy. The acquisition of annual very high-resolution aerial images by the National Agriculture Imagery Program (NAIP) and their accessibility through Google Earth Engine (GEE) supports the delineation of tree canopies and change over time in a cost and time-effective manner. The objectives of this study are to develop an automated method to detect the crowns of individual western Juniper (Juniperus occidentalis) trees and to assess the change of forest cover from multispectral 1-meter resolution NAIP images collected from 2009 to 2016, in Oregon, USA. The Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), and Ratio Vegetation Index (RVI), were calculated from the NAIP images, in addition to the red-green-blue-near infrared bands. To identify the most suitable approach for individual tree crown identification, we created two training datasets: one considering yearly images separately and one merging all images, irrespective of the year. We segmented individual tree crowns using a random forest algorithm implemented in GEE and seven rasters, namely the reflectance of four spectral bands as recorded by the NAIP images (i.e., the red-green-blue-near infrared) and three calculated indices (i.e., NDVI, NDWI, and RVI). We compared the estimated location of the trees, computed as the centroid of the crown, with the visually identified treetops, which were considered as validation locations. We found that tree location errors were smaller when years were analyzed individually than by merging the years. Measurements of completeness (74%), correctness (94%), and mean accuracy detection (82 %) show promising performance of the random forest algorithm in crown delineation, considering that only four original input bands were used for crown segmentation. The change in the calculated crown area for western juniper follows a sinusoidal curve, with a decrease from 2011 to 2012 and an increase from 2012 to 2014. The proposed approach has the potential to estimate individual tree locations and forest cover area dynamics at broad spatial scales using regularly collected airborne imagery with easy-to-implement methods.

Mapping essential urban land use categories with open big data: Results for five metropolitan areas in the United States of America

Urban land-use maps outlining the distribution, pattern, and composition of various land use types are critically important for urban planning, environmental management, disaster control, health protection, and biodiversity conservation. Recent advances in remote sensing and social sensing data and methods have shown great potentials in mapping urban land use categories, but they are still constrained by mixed land uses, limited predictors, non-localized models, and often relatively low accuracies. To inform these issues, we proposed a robust and cost-effective framework for mapping urban land use categories using openly available multi-source geospatial “big data”. With street blocks generated from OpenStreetMap (OSM) data as the minimum classification unit, we integrated an expansive set of multi-scale spatially explicit information on land surface, vertical height, socio-economic attributes, social media, demography, and topography. We further proposed to apply the automatic ensemble learning that leverages a bunch of machine learning algorithms in deriving optimal urban land use classification maps. Results of block-level urban land use classification in five metropolitan areas of the United States found the overall accuracies of major-class (Level-I) and minor-class (Level-II) classification could be high as 91% and 86%, respectively. A multi-model comparison revealed that for urban land use classification with high-dimensional features, the multi-layer stacking ensemble models achieved better performance than base models such as random forest, extremely randomized trees, LightGBM, CatBoost, and neural networks. We found without very-high-resolution National Agriculture Imagery Program imagery, the classification results derived from Sentinel-1, Sentinel-2, and other open big data based features could achieve plausible overall accuracies of Level-I and Level-II classification at 88% and 81%, respectively. We also found that model transferability depended highly on the heterogeneity in characteristics of different regions. The methods and findings in this study systematically elucidate the role of data sources, classification methods, and feature transferability in block-level land use classifications, which have important implications for mapping multi-scale essential urban land use categories.

The tree cover and temperature disparity in US urbanized areas: Quantifying the association with income across 5,723 communities.

Urban tree cover provides benefits to human health and well-being, but previous studies suggest that tree cover is often inequitably distributed. Here, we use National Agriculture Imagery Program digital ortho photographs to survey the tree cover inequality for Census blocks in US large urbanized areas, home to 167 million people across 5,723 municipalities and other Census-designated places. We compared tree cover to summer land surface temperature, as measured using Landsat imagery. In 92% of the urbanized areas surveyed, low-income blocks have less tree cover than high-income blocks. On average, low-income blocks have 15.2% less tree cover and are 1.5⁰C hotter than high-income blocks. The greatest difference between low- and high-income blocks was found in urbanized areas in the Northeast of the United States, where low-income blocks in some urbanized areas have 30% less tree cover and are 4.0⁰C hotter. Even after controlling for population density and built-up intensity, the positive association between income and tree cover is significant, as is the positive association between proportion non-Hispanic white and tree cover. We estimate, after controlling for population density, that low-income blocks have 62 million fewer trees than high-income blocks, equal to a compensatory value of $56 billion ($1,349/person). An investment in tree planting and natural regeneration of $17.6 billion would be needed to close the tree cover disparity, benefitting 42 million people in low-income blocks.

Impacts of the San Francisco Bay Area shelter-in-place during the COVID-19 pandemic on urban heat fluxes

The purpose of this study was to make quantitative connections between changes in social and economic activities in northern California urban areas and related Earth system environmental responses to the COVID-19 pandemic in 2020. We tested the hypothesis that the absence of worker activities during Shelter-in-Place in the San Francisco Bay Area detectably altered the infrared heat flux from parking lots, highways, and large building rooftops, caused primarily by quantitative changes in the reflective properties in these different classes of urban surfaces. The Landsat satellite's thermal infrared (TIR) sensor imagery for surface temperature (ST) was quantified for all the large urban features in the Bay Area that have flat (impervious) surfaces, such parking lots, wide roadways, and rooftops. These large impervious surface features in the five-county Bay Area were first delineated and classified using sub-meter aerial imagery from the National Agriculture Imagery Program (NAIP). We then compared Landsat ST data acquired on (or near) the same dates from the three previous years (2017–2019) for all these contiguous impervious surfaces. Results showed that all the large parking lots, roadway corridors, and industrial/commercial rooftops across the entire Bay Area urban landscape were detected by Landsat ST time series as significantly cooler (by 5o C to 8o C) during the unprecedented Shelter-in-Place period of mid-March to late-May of 2020, compared to same months of the three previous years. The explanation for this region-wide cooling pattern in 2020 that was best supported by both remote sensing and ground-based data sets was that relatively low atmospheric aerosol lower (PM2.5) concentrations from mid-March to late May of 2020 resulted in weaker temperature inversions over the Bay Area, higher diurnal surface mixing, and lowered urban surface temperatures, compared to the three previous years.

High-resolution forest carbon mapping for climate mitigation baselines over the RGGI region, USA

Large-scale airborne lidar data collections can be used to generate high-resolution forest aboveground biomass maps at the state level and beyond as demonstrated in early phases of NASA’s Carbon Monitoring System program. While products like aboveground biomass maps derived from these leaf-off lidar datasets each can meet state- or substate-level measurement requirements individually, combining them over multiple jurisdictions does not guarantee the consistency required in forest carbon planning, trading and reporting schemes. In this study, we refine a multi-state level forest carbon monitoring framework that addresses these spatial inconsistencies caused by variability in data quality and modeling techniques. This work is built upon our long term efforts to link airborne lidar, National Agricultural Imagery Program imagery and USDA Forest Service Forest Inventory and Analysis plot measurements for high-resolution forest aboveground biomass mapping. Compared with machine learning algorithms (r 2 = 0.38, bias = −2.3, RMSE = 45.2 Mg ha−1), the use of a linear model is not only able to maintain a good prediction accuracy of aboveground biomass density (r 2 = 0.32, bias = 4.0, RMSE = 49.4 Mg ha−1) but largely mitigates problems related to variability in data quality. Our latest effort has led to the generation of a consistent 30 m pixel forest aboveground carbon map covering 11 states in the Regional Greenhouse Gas Initiative region of the USA. Such an approach can directly contribute to the formation of a cohesive forest carbon accounting system at national and even international levels, especially via future integrations with NASA’s spaceborne lidar missions.

Examining Landscape-Scale Fuel and Terrain Controls of Wildfire Spread Rates Using Repetitive Airborne Thermal Infrared (ATIR) Imagery

The objectives of this study are to evaluate landscape-scale fuel and terrain controls on fire rate of spread (ROS) estimates derived from repetitive airborne thermal infrared (ATIR) imagery sequences collected during the 2017 Thomas and Detwiler extreme wildfire events in California. Environmental covariate data were derived from prefire National Agriculture Imagery Program (NAIP) orthoimagery and USGS digital elevation models (DEMs). Active fronts and spread vectors of the expanding fires were delineated from ATIR imagery. Then, statistical relationships between fire spread rates and landscape covariates were analyzed using bivariate and multivariate regression. Directional slope is found to be the most statistically significant covariate with ROS for the five fire imagery sequences that were analyzed and its relationship with ROS is best characterized as an exponential growth function (adj. R2 max = 0.548, min = 0.075). Imaged-derived fuel covariates alone are statistically weak predictors of ROS (adj. R2 max = 0.363, min = 0.002) but, when included in multivariate models, increased ROS predictability and variance explanation (+14%) compared to models with directional slope alone.

Digital mapping of structural conservation practices in the Midwest U.S. croplands: Implementation and preliminary analysis

The application of best management practices is a long-term conservation effort in Midwest U.S. croplands, and many farmers have adopted structural conservation practices (SCPs) to reduce soil erosion and surface water runoff, such as terraces and grassed waterways. Despite that, the geographic distribution of these practices is barely known in the region, and mapping initiatives are required to develop timely and spatially explicit inventories of SCP areas. This study presents the first mapping of SCPs in the agricultural areas over 12 Midwest U.S. states. Semantic segmentation model (adapted U-Net) and National Agriculture Imagery Program 2018–2019 data were used to map the SCP areas at 2-m spatial resolution (490.2 billion pixels). In general, mapping results achieved 78.2% overall accuracy across 20 counties. Our results indicate that 52% of SCP areas are distributed over Iowa (26%), Illinois (15%), and Nebraska (11%). In contrast, the states with the lowest SCP areas are Michigan and North Dakota, with less than 4% of SCP areas. Since the SCP extent is also dependent on the number of cropland areas per state, the percentage of SCP per cropland area was calculated. Specifically, the average percentage of SCP area per cropland is ~1.19%, ranging from 0.8 (e.g., North Dakota and south Minnesota) to 5.5% (e.g., northeast Kansas and southwest Iowa). Interestingly, results also illustrate that regions with high soil erosion rates present the largest percentage of SCP areas in croplands, indicating conservation efforts by farmers. While this preliminary analysis shows some limitations in the mapping quality (mislabel, non-accurate location or discontinuity of SCP areas), the framework has a potential for operational conservation monitoring. The development of such mapping has positive implications for conservation programs, and this geospatial inventory is easily accessible information for the large-area evaluation of conservation practices across Midwest U.S. croplands.

Heat-Moderating Effects of Bus Stop Shelters and Tree Shade on Public Transport Ridership.

Rising temperatures threaten the resilience of public transit systems. We determined whether bus stop shelters and tree canopy surrounding bus stops moderated the effect of warm season temperatures on ridership in Austin, Texas, and whether shelters and trees were equitably distributed. For bus stops (n = 2271) of Capital Metropolitan Transportation Authority, boardings per bus were measured 1 April–30 September 2019. Air temperature data originated from the Camp Mabry weather station. Tree canopy was calculated by classification of high-resolution aerial imagery from the National Agriculture Imagery Program. Data on race, ethnicity, poverty level, median age, and bus commuters within census tracts of bus stops originated from the 2014–2018 American Community Survey. Using multilevel negative binomial regression models, we found that shelters did not moderate the effect of high temperatures on ridership (p > 0.05). During high temperatures, each one-percent increase in tree canopy was associated with a lesser decrease (1.6%) in ridership compared to if there were no trees (1.7%) (p < 0.001). In general, shelters and trees were equitably distributed. Insignificant or modest effects of shelters and trees on ridership during high temperatures may be attributed to the transit dependency of riders. For climate change adaptation, we recommend tree planting at bus stops to protect from ridership losses and unhealthy exposure to extreme heat.

A customized framework for regional classification of conifers using automated feature extraction

Pinyon and juniper expansion into sagebrush ecosystems is one of the major challenges facing land managers in the Great Basin. Effective pinyon and juniper treatment requires maps that accurately and precisely depict tree location and degree of woodland development so managers can target restoration efforts for early stages of pinyon and juniper expansion. However, available remotely sensed layers that cover a regional spatial extent lack the spatial resolution or accuracy to meet this need. Accuracy can be improved using object-based image analysis methods such as automated feature extraction, which has proven successful in accurately classifying land cover at the site-level but to date has yet to be applied to regional extents due to time and computational limitations. Using Feature Analyst™, we implement our framework with 1-m2 reference imagery provided by National Agricultural Imagery Program to classify conifers across Nevada and northeastern California. Our resulting binary conifer map has an overall accuracy of 86%. We discuss the advantages to accuracy and precision our framework provides compared to other classification methods.● This framework allows automated feature extraction for large quantities of data and very high spatial resolution imagery● It leverages supervised learning● It results in high accuracy maps for regional spatial extents

Vegetation mapping of No Name Key, Florida using lidar and multispectral remote sensing

ABSTRACT LIght Detection And Ranging (lidar) data have been widely used in the areas of ecological studies due to lidar’s ability to provide information on the vertical structure of vegetation in wildlife habitats. The overall objective of this project was to map the vegetation on No Name Key, Florida where endangered wildlife species reside using publicly available remote sensing data such as lidar data and high resolution aerial images (including National Agricultural Imagery Program (NAIP) images). The methods involved the use of 4 different classification algorithms (Support Vector Machine, Random Forest, Maximum Likelihood, and Mahalanobis Distance), different classification settings (default and custom settings), and normalization (original and normalized input bands) on 2 different input stacked images, NAIP image alone and NAIP combined with lidar data. A majority filter was applied to each classification output before performing the accuracy assessment. The result of performing the image classifications showed the following: across all inputs and classification algorithms, the highest overall accuracy (OA) and kappa coefficient (к) were achieved by the Random Forest classification on the NAIP-lidar stacked image with an applied majority filter and original input bands. The result also showed the airborne data combined with the lidar data resulted in a higher classification accuracy than the airborne data alone; and when normalization was applied to all the input bands or layers, the classification accuracies were not increased in most cases compared to when original bands were used in the classification. In most cases, the application of the majority filter increased the accuracy of the classification results as opposed to when no majority filter was applied. Instead of using default values, when the new parameter values (i.e., custom settings) for the penalty parameter (C) and the gamma (ɤ) were used, the accuracies of the Support Vector Machine (SVM) classified images were slightly increased compared to using the default values for the 2 parameters. With increased availability of public lidar data, combining the aerial image with lidar data would enhance the accuracy of vegetation mapping, which would lead to more effective and accurate wildlife management.

Classification Strategies for Unbalanced Binary Maps: Finding Ponderosa Pine (Pinus ponderosa) in the Willamette Valley

Forest species classifications are becoming increasingly automated as advances are made in machine learning. Complex algorithms can reach high accuracies, but are not always suitable for small-scale classifications, which may benefit from simpler conventional methods. The goal of this classification was to identify contiguous stands of ponderosa pine (Pinus ponderosa Douglas ex Lawson) against a mix of forest and non-forest background in the southern Willamette Valley, Oregon. The study area is approximately 816,600 ha, considerably larger than most study areas used for presenting techniques for tree species classification. To achieve the objective, we used two classification procedures, one parametric and one non-parametric. For the parametric method, we selected the maximum likelihood (ML) algorithm, whereas for the non-parametric method we chose the random forest (RF) algorithm. To identify ponderosa pine, we used 1 m spatial resolution red-green-blue-infrared (RGBI) aerial images supplied by the U.S. National Agriculture Imagery Program (NAIP) and 1 m spatial resolution canopy height models (CHMs) provided by the Oregon Department of Geology and Mineral Industries (DOGAMI). We tested four data variations for each method: Aerial imagery, CHM-masked aerial imagery, aerial imagery with an additional CHM band, and CHM-masked aerial imagery with a CHM band. The parametric classifications of aerial imagery alone reached an average kappa coefficient of 0.29, which increased to 0.51 when masked with CHM data. The incorporation of CHM data as a fifth band resulted in a similar improvement in kappa (0.47), but the most effective parametric method was the incorporation of CHM data as both a fifth band and a post-classification mask, resulting in a kappa coefficient of 0.89. The non-parametric classification of aerial imagery achieved a mean validation kappa coefficient of 0.85 collectively and 0.90 individually, which only increased by approximately 0.01 or less when the CHM masks were applied. The addition of the CHM band increased the kappa value to 0.91 for both individual and collective tile classifications. The highest kappa of all methods was achieved through five-band non-parametric classification with the addition of the CHM band (0.94) for both collective and individual classifications. Our results suggest that parametric methods, when enhanced with a CHM mask, could be suitable for large-area, small-scale classifications based on RGBI imagery, but a non-parametric classification of fused spectral and height data will generally achieve the highest accuracy for large, unbalanced datasets.

Assessing restoration potential for beaver (Castor canadensis) in the semiarid foothills of the Southern Rockies, USA

AbstractWe used the Beaver Restoration Assessment Tool (BRAT) developed at Utah State University to develop spatially explicit estimates of maximum beaver‐carrying capacity in a 160 km2 watershed in the foothills of the Southern Rockies. The watershed does not currently have beaver but has extensive evidence of past beaver occupation. BRAT uses input data on stream flow, topography, and vegetation. We compared BRAT results using three different types of vegetation inputs: the default LANDFIRE data at 30‐m spatial resolution and pixel‐based and object‐based image analysis (OBIA) with 1‐m resolution National Agriculture Imagery Program imagery. OBIA produced the most accurate results relative to ground‐based vegetation mapping. Changes in vegetation input data resulted in substantial changes in BRAT estimates of beaver‐carrying capacity. Using historic LANDFIRE vegetation data and field evidence of past beaver activity, contemporary beaver‐carrying capacity is much lower than historical capacity throughout the watershed. These changes are especially pronounced in low‐order stream segments, suggesting that beaver reintroduction in this watershed could be facilitated by measures to restore riparian vegetation along low‐order streams. This case study demonstrates the value of using BRAT as a tool prior to beaver reintroduction and stream restoration.

Evaluating the integrity of forested riparian buffers over a large area using LiDAR data and Google Earth Engine

Spatial and temporal changes in land cover have direct impacts on the hydrological cycle and stream quality. Techniques for accurately and efficiently mapping these changes are evolving quickly, and it is important to evaluate how useful these techniques are to address the environmental impact of land cover on riparian buffer areas. The objectives of this study were to: (1) determine the classes and distribution of land cover in the riparian areas of streams; (2) examine the discrepancies within the existing land cover data from National Land Cover Database (NLCD) using high-resolution imagery of the National Agriculture Imagery Program (NAIP) and a LiDAR canopy height model; and (3) develop a technique using LiDAR data to help characterize riparian buffers over large spatial extents. One-meter canopy height models were constructed in a high-throughput computing environment. The machine learning algorithm Support Vector Machine (SVM) was trained to perform supervised land cover classification at a 1-m resolution on the Google Earth Engine (GEE) platform using NAIP imagery and LiDAR-derived canopy height models. This integrated approach to land cover classification provided a substantial improvement in the resolution and accuracy of classifications with F1 Score of each land cover classification ranging from 64.88 to 95.32%. The resulting 1-m land cover map is a highly detailed representation of land cover in the study area. Forests (evergreen and deciduous) and wetlands are by far the dominant land cover classes in riparian zones of the Lower Savannah River Basin, followed by cultivated crops and pasture/hay. Stress from urbanization in the riparian zones appears to be localized. This study demonstrates a method to create accurate high-resolution riparian buffer maps which can be used to improve water management and provide future prospects for improving buffer zones monitoring to assess stream health.

Distance effects of gas field infrastructure on pygmy rabbits in southwestern Wyoming

AbstractAs domestic energy development activity continues in the western United States, wildlife conservation planning in affected regions is increasingly important. The geologic basins where oil and gas energy exploration is occurring are primarily sagebrush steppe rangelands. Sagebrush steppe habitats may support more than 20 vertebrate species of conservation concern, and for many of these species, information is lacking on the effects of gas energy development. In earlier work, we demonstrated a negative relationship among development density of gas field infrastructure and pygmy rabbits (Brachylagus idahoensis). We now examine the spatial relationship among gas field infrastructure, pygmy rabbits, and their habitat on four major gas fields in southwest Wyoming. Using data collected from 120 plots over three years (2011–2013) and 2012 National Agriculture Imagery Program (NAIP) imagery, we evaluated (1) whether well pads are more likely to be located in areas of pygmy rabbit habitat, (2) whether the presence and abundance of pygmy rabbits are related to distance from infrastructure, and, if so, (3) how much of the total surface area on a gas field is affected. Well pads on three gas fields occurred in higher quality pygmy rabbit habitat than did a set of randomly generated points, and the abundance and probability of pygmy rabbits being present were lower within approximately 0.5–1.5 km of the nearest road and 2 km of well pads and utilities. Buffering a digital layer of roads and well pads on one gas field revealed that nearly 82% of the (4417 km2) surface area was within 1 km of infrastructure, and over 95% of the gas field surface area was within 2 km. This need not be the case on future gas fields. Directional and horizontal well drilling technologies now make it possible for gas to be recovered from a greater area per well pad, enabling future gas field developments that require fewer well pads, roads, and pipeline corridors. Such changes would enable increased well pad spacing and provide the opportunity to locate gas field infrastructure in areas of poor quality wildlife habitat, avoid high priority habitat, and conserve a greater amount of on‐field wildlife habitat overall.

A Simplified Framework for High-Resolution Urban Vegetation Classification with Optical Imagery in the Los Angeles Megacity

High spatial resolution maps of Los Angeles, California are needed to capture the heterogeneity of urban land cover while spanning the regional domain used in carbon and water cycle models. We present a simplified framework for developing a high spatial resolution map of urban vegetation cover in the Southern California Air Basin (SoCAB) with publicly available satellite imagery. This method uses Sentinel-2 (10–60 × 10–60 m) and National Agriculture Imagery Program (NAIP) (0.6 × 0.6 m) optical imagery to classify urban and non-urban areas of impervious surface, tree, grass, shrub, bare soil/non-photosynthetic vegetation, and water. Our approach was designed for Los Angeles, a geographically complex megacity characterized by diverse Mediterranean land cover and a mix of high-rise buildings and topographic features that produce strong shadow effects. We show that a combined NAIP and Sentinel-2 classification reduces misclassified shadow pixels and resolves spatially heterogeneous vegetation gradients across urban and non-urban regions in SoCAB at 0.6–10 m resolution with 85% overall accuracy and 88% weighted overall accuracy. Results from this study will enable the long-term monitoring of land cover change associated with urbanization and quantification of biospheric contributions to carbon and water cycling in cities.