Search

Spectral remote sensing has the potential to improve scouting and management of soybean aphid (Aphis glycines Matsumura), which can cause yield losses of over 40% in the North Central Region of the United States. We used linear support vector machines (SVMs) to determine 1) whether hyperspectral samples could be classified into treat/no-treat classes based on the economic threshold (250 aphids per plant) and 2) how many wavelengths or features are needed to generate an accurate model without overfitting the data. A range of aphid infestation levels on soybean was created using caged field plots in 2013, 2014, 2017, and 2018 in Minnesota and in 2017 and 2018 in Iowa. Hyperspectral measurements of soybean canopies in each plot were recorded with a spectroradiometer. SVM training and testing were performed using 15 combinations of normalized canopy reflectance at wavelengths of 720, 750, 780, and 1,010 nm. Pairwise Bonferroni-adjusted t-tests of Cohen’s kappa values showed four wavelength combinations were optimal, namely model 1 (780 nm), model 2 (780 and 1,010 nm), model 3 (780, 1,010, and 720 nm), and model 4 (780, 1,010, 720, and 750 nm). Model 2 showed the best overall performance, with an accuracy of 89.4%, a sensitivity of 81.2%, and a specificity of 91.6%. The findings from this experiment provide the first documentation of successful classification of remotely sensed spectral data of soybean aphid-induced stress into threshold-based classes.

Spatial variability is a prominent feature of various geographic phenomena such as climatic zones, USDA plant hardiness zones, and terrestrial habitat types (e.g., forest, grasslands, wetlands, and deserts). However, current deep learning methods follow a spatial-one-size-fits-all (OSFA) approach to train single deep neural network models that do not account for spatial variability. Quantification of spatial variability can be challenging due to the influence of many geophysical factors. In preliminary work, we proposed a spatial variability aware neural network (SVANN-I, formerly called SVANN ) approach where weights are a function of location but the neural network architecture is location independent. In this work, we explore a more flexible SVANN-E approach where neural network architecture varies across geographic locations. In addition, we provide a taxonomy of SVANN types and a physics inspired interpretation model. Experiments with aerial imagery based wetland mapping show that SVANN-I outperforms OSFA and SVANN-E performs the best of all.

Soybean aphid, Aphis glycines Matsumura (Hemiptera: Aphididae), is a common pest of soybean, Glycine max (L.) Merrill (Fabales: Fabaceae), in North America requiring frequent scouting as part of an integrated pest management plan. Current scouting methods are time consuming and provide incomplete coverage of soybean. Unmanned aerial vehicles (UAVs) are capable of collecting high-resolution imagery that offer more detailed coverage in agricultural fields than traditional scouting methods. Recently, it was documented that changes to the spectral reflectance of soybean canopies caused by aphid-induced stress could be detected from ground-based sensors; however, it remained unknown whether these changes could also be detected from UAV-based sensors. Small-plot trials were conducted in 2017 and 2018 where cages were used to manipulate aphid populations. Additional open-field trials were conducted in 2018 where insecticides were used to create a gradient of aphid pressure. Whole-plant soybean aphid densities were recorded along with UAV-based multispectral imagery. Simple linear regressions were used to determine whether UAV-based multispectral reflectance was associated with aphid populations. Our findings indicate that near-infrared reflectance decreased with increasing soybean aphid populations in caged trials when cumulative aphid days surpassed the economic injury level, and in open-field trials when soybean aphid populations were above the economic threshold. These findings provide the first documentation of soybean aphid-induced stress being detected from UAV-based multispectral imagery and advance the use of UAVs for remote scouting of soybean aphid and other field crop pests.

Class ambiguity refers to the phenomenon whereby similar features correspond to different classes at different locations. Given heterogeneous geographic data with class ambiguity, the spatial ensemble learning (SEL) problem aims to find a decomposition of the geographic area into disjoint zones such that class ambiguity is minimized and a local classifier can be learned in each zone. The problem is important for applications such as land cover mapping from heterogeneous earth observation data with spectral confusion. However, the problem is challenging due to its high computational cost. Related work in ensemble learning either assumes an identical sample distribution (e.g., bagging, boosting, random forest) or decomposes multi-modular input data in the feature vector space (e.g., mixture of experts, multimodal ensemble) and thus cannot effectively minimize class ambiguity. In contrast, we propose a spatial ensemble framework that explicitly partitions input data in geographic space. Our approach first preprocesses data into homogeneous spatial patches and uses a greedy heuristic to allocate pairs of patches with high class ambiguity into different zones. We further extend our spatial ensemble learning framework with spatial dependency between nearby zones based on the spatial autocorrelation effect. Both theoretical analysis and experimental evaluations on two real world wetland mapping datasets show the feasibility of the proposed approach.

Building footprints are among the most predominant features in urban areas, and provide valuable information for urban planning, solar energy suitability analysis, etc. We aim to automatically and ...

Outbreaks of blastomycosis, caused by the fungus Blastomyces dermatitidis, occur in areas of endemicity in the United States and Canada, but the geographic range of blastomycosis is expanding. Previous studies inferred the location of B. dermatitidis through epidemiologic data associated with outbreaks because culture of B. dermatitidis from the environment is often unsuccessful. In this study, we used a culture-independent, PCR-based method to identify B. dermatitidis DNA in environmental samples using the BAD1 promoter region. We tested 250 environmental samples collected in Minnesota, either associated with blastomycosis outbreaks or environmental samples collected from regions of high and low endemicity to determine the basal prevalence of B. dermatitidis in the environment. We identified a fifth BAD1 promoter haplotype of B. dermatitidis prevalent in Minnesota. Ecological niche analysis identified latitude, longitude, elevation, and site classification as environmental parameters associated with the presence of B. dermatitidis. Using this analysis, a random forest model predicted the presence of B. dermatitidis in basal environmental samples with 75% accuracy. These data support the use of culture-independent, PCR-based environmental sampling to track spread into new regions and to characterize the unknown B. dermatitidis environmental niche. IMPORTANCE Upon inhalation of spores from the fungus Blastomyces dermatitidis from the environment, humans and animals can develop the disease blastomycosis. Based on disease epidemiology, B. dermatitidis is known to be endemic in the United States and Canada around the Great Lakes and in the Ohio and Mississippi River Valleys but is starting to emerge in other areas. B. dermatitidis is extremely difficult to culture from the environment, so little is known about the environmental reservoirs for this pathogen. We used a culture-independent PCR-based assay to identify the presence of B. dermatitidis DNA in soil samples from Minnesota. By combining molecular data with ecological niche modeling, we were able to predict the presence of B. dermatitidis in environmental samples with 75% accuracy and to define characteristics of the B. dermatitidis environmental niche. Importantly, we showed the effectiveness of using a PCR-based assay to identify B. dermatitidis in environmental samples.

Wetland managers, citizens and government leaders are observing rapid changes in coastal wetlands and associated habitats around the Great Lakes Basin due to human activity and climate variability. SAR and optical satellite sensors offer cost effective management tools that can be used to monitor wetlands over time, covering large areas like the Great Lakes and providing information to those making management and policy decisions. In this paper we describe ongoing efforts to monitor dynamic changes in wetland vegetation, surface water extent, and water level change. Included are assessments of simulated Radarsat Constellation Mission data to determine feasibility of continued monitoring into the future. Results show that integration of data from multiple sensors is most effective for monitoring coastal wetlands in the Great Lakes region. While products developed using methods described in this article provide valuable management tools, more effort is needed to reach the goal of establishing a dynamic, near-real-time, remote sensing-based monitoring program for the basin.

The ability to predict spatially explicit nitrogen uptake (NUP) in maize (Zea mays L) during the early development stages provides clear value for making in-season nitrogen fertilizer applications that can improve NUP efficiency and reduce the risk of nitrogen loss to the environment. Aerial hyperspectral imaging is an attractive agronomic research tool for its ability to capture spectral data over relatively large areas, enabling its use for predicting NUP at the field scale. The overarching goal of this work was to use supervised learning regression algorithms—Lasso, support vector regression (SVR), random forest, and partial least squares regression (PLSR)—to predict early season (i.e., V6–V14) maize NUP at three experimental sites in Minnesota using high-resolution hyperspectral imagery. In addition to the spectral features offered by hyperspectral imaging, the 10th percentile Modified Chlorophyll Absorption Ratio Index Improved (MCARI2) was made available to the learning models as an auxiliary feature to assess its ability to improve NUP prediction accuracy. The trained models demonstrated robustness by maintaining satisfactory prediction accuracy across locations, pixel sizes, development stages, and a broad range of NUP values (4.8 to 182 kg ha-1). Using the four most informative spectral features in addition to the auxiliary feature, the mean absolute error (MAE) of Lasso, SVR, and PLSR models (9.4, 9.7, and 9.5 kg ha-1, respectively) was lower than that of random forest (11.2 kg ha-1). The relative MAE for the Lasso, SVR, PLSR, and random forest models was 16.5%, 17.0%, 16.6%, and 19.6%, respectively. The inclusion of the auxiliary feature not only improved overall prediction accuracy by 1.6 kg ha-1 (14%) across all models, but it also reduced the number of input features required to reach optimal performance. The variance of predicted NUP increased as the measured NUP increased (MAE of the Lasso model increased from 4.0 to 12.1 kg ha-1 for measured NUP less than 25 kg ha-1 and greater than 100 kg ha-1, respectively). The most influential spectral features were oftentimes adjacent to each other (i.e., within approximately 6 nm), indicating the importance of both spectral precision and derivative spectra around key wavelengths for explaining NUP. Finally, several challenges and opportunities are discussed regarding the use of these results in the context of improving nitrogen fertilizer management.

Ash trees (Fraxinus spp.) are a prominent species in Minnesota forests, with an estimated 1.1 billion trees in the state, totaling approximately 8% of all trees. Ash trees are threatened by the invasive emerald ash borer (Agrilus planipennis Fairmaire), which typically results in close to 100% tree mortality within one to five years of infestation. A detailed, wall-to-wall map of ash presence is highly desirable for forest management and monitoring applications. We used Google Earth Engine to compile Landsat time series analysis, which provided unique information on phenologic patterns across the landscape to identify ash species. Topographic position information derived from lidar was added to improve spatial maps of ash abundance. These input data were combined to produce a classification map and identify the abundance of ash forests that exist in the state of Minnesota. Overall, 12,524 km2 of forestland was predicted to have greater than 10% probability of ash species present. The overall accuracy of the composite ash presence/absence map was 64% for all ash species and 72% for black ash, and classification accuracy increased with the length of the time series. Average height derived from lidar was the best model predictor for ash basal area (R2 = 0.40), which, on average, was estimated as 16.1 m2 ha−1. Information produced from this map will be useful for natural resource managers and planners in developing forest management strategies which account for the spatial distribution of ash on the landscape. The approach used in this analysis is easily transferable and broadly scalable to other regions threatened with forest health problems such as invasive insects.

Imagery from the Landsat Program has been used frequently as a source of auxiliary data for modeling land cover, as well as a variety of attributes associated with tree cover. With ready access to all scenes in the archive since 2008 due to the USGS Landsat Data Policy, new approaches to deriving such auxiliary data from dense Landsat time series are required. Several methods have previously been developed for use with finer temporal resolution imagery (e.g. AVHRR and MODIS), including image compositing and harmonic regression using Fourier series. The manuscript presents a study, using Minnesota, USA during the years 2009–2013 as the study area and timeframe. The study examined the relative predictive power of land cover models, in particular those related to tree cover, using predictor variables based solely on composite imagery versus those using estimated harmonic regression coefficients. The study used two common non-parametric modeling approaches (i.e. k-nearest neighbors and random forests) for fitting classification and regression models of multiple attributes measured on USFS Forest Inventory and Analysis plots using all available Landsat imagery for the study area and timeframe. The estimated Fourier coefficients developed by harmonic regression of tasseled cap transformation time series data were shown to be correlated with land cover, including tree cover. Regression models using estimated Fourier coefficients as predictor variables showed a two- to threefold increase in explained variance for a small set of continuous response variables, relative to comparable models using monthly image composites. Similarly, the overall accuracies of classification models using the estimated Fourier coefficients were approximately 10–20 percentage points higher than the models using the image composites, with corresponding individual class accuracies between six and 45 percentage points higher.

Invasive plant species are an increasing worldwide threat both ecologically and financially. Knowing the location of these invasive plant infestations is the first step in their control. Surveying for invasive Phragmites australis is particularly challenging due to limited accessibility in wetland environments. Unoccupied aircraft systems (UAS) are a popular choice for invasive species management due to their ability to survey challenging environments and their high spatial and temporal resolution. This study tested the utility of three-band (i.e., red, green, and blue; RGB) UAS imagery for mapping Phragmites in the St. Louis River Estuary in Minnesota, U.S.A. and Saginaw Bay in Michigan, U.S.A. Iterative object-based image analysis techniques were used to identify two classes, Phragmites and Not Phragmites. Additionally, the effectiveness of canopy height models (CHMs) created from two data types, UAS imagery and commercial satellite stereo retrievals, and the RADARSAT-2 horizontal-horizontal (HH) polarization were tested for Phragmites identification. The highest overall classification accuracy of 90% was achieved when pairing the UAS imagery with a UAS-derived CHM. Producer’s accuracy for the Phragmites class ranged from 3 to 76%, and the user’s accuracies were above 90%. The Not Phragmites class had user’s and producer’s accuracies above 88%. Inclusion of the RADARSAT-2 HH polarization caused a slight reduction in classification accuracy. Commercial satellite stereo retrievals increased commission errors due to decreased spatial resolution and vertical accuracy. The lowest classification accuracy was seen when using only the RGB UAS imagery. UAS are promising for Phragmites identification, but the imagery should be used in conjunction with a CHM.

The objective of this project was to validate the efficacy of a miniature multispectral, single-sensor camera for detecting stress in deciduous juvenile tree foliage in a controlled environment. To that end, deciduous liners (one year old, nursery-grown transplants) representing five tree species (Celtis occidentalis L., Gleditsia triacanthos forma inermis Schneid, Gymnocladus dioicus (L.) K. Koch, Quercus rubra L., and Ulmus americana L.) were containerized and subsequently subjected to two treatments representing common landscape site stresses: drought and elevated soil pH. To minimize any atmospheric effects on reflectance values, the trees were placed in enclosed, artificially lit chambers during the detection process. The normalized difference vegetation index (NDVI) was utilized as the metric for analysis. Results showed the single imager sensor derived NDVI values were reliable indicators of tree stress within the species groups. Results also revealed that the derived values were not reliable indicators of tree stress across the species groups.

Tree management is becoming a big issue in a variety of societal domains. In recent years, historic wildfires and blackouts caused by failures in tree management have increased in both quantity and severity, resulting in many deaths and financial loses in the tens of billions of dollars. Many communities are also suffering from massive tree loss (e.g., in the millions) that affects the health and well-being of citizens. These problems are likely to worsen due to climate change, aging infrastructure and population growth. Tree management needs a revolution to deal with these urgent problems. This opens up new challenges and opportunities for the spatial community. This paper presents some of the open research problems from the perspectives of individual tree mapping and characterization as well as decision making and in-field intervention.

Given remote sensing datasets in a spatial domain, we aim to detect geospatial objects with minimum bounding rectangles (i.e., angle-aware) leveraging deep learning frameworks. Geospatial objects (e.g., buildings, vehicles, farms) provide meaningful information for a variety of societal applications, including urban planning, census, sustainable development, security surveillance, agricultural management, etc. The detection of these objects are challenging because their directions are often heavily mixed and not parallel to the orthogonal directions of an image frame due to topography, planning, etc. In addition, there is very limited training data with angle information for most types of objects. In related work, state-of-the-art deep learning frameworks detect objects using orthogonal bounding rectangles (i.e., sides are parallel to the sides of an input image), so they cannot identify the directions of objects and generate loose rectangular bounds on objects. We propose an Unsupervised Augmentation (UA) framework to detect geospatial objects with general minimum bounding rectangles (i.e., with angles). The UA framework contains two schemes, namely a ROtation-Vector (ROV) based scheme and a context-based scheme. The schemes completely avoid the need for: (1) additional ground-truth data with annotated angles; (2) restructuring of existing network architectures; and (3) re-training. Experimental results show that the UA framework can well approximate the angles of objects and generate much tighter bounding boxes on objects.

Tree inventories are important datasets for many societal applications (e.g., urban planning). However, tree inventories still remain unavailable in most urban areas. We aim to automate tree identification at individual levels in urban areas at a large scale using remote sensing datasets. The problem is challenging due to the complexity of the landscape in urban scenarios and the lack of ground truth data. In related work, tree identification algorithms have mainly focused on controlled forest regions where the landscape is mostly homogeneous with trees, making the methods difficult to generalize to urban environments. We propose a TIMBER framework to find individual trees in complex urban environments and a Core Object REduction (CORE) algorithm to improve the computational efficiency of TIMBER. Experiments show that TIMBER can efficiently detect urban trees with high accuracy.

Long-term, large-scale underground coal mining activities in Huainan, China, have damaged the stability of the overlying rocks, resulting in large waterlogged areas. The water quality in these areas is compromised by agricultural pollutants and discharge of untreated waste. In this study, chlorophyll-a (Chl-a) concentrations were estimated in two types of waterlogged area, semi-closed (the Panji area [Pj]) and closed (the Guqiao area [Gq]), using information from water samples and in situ hyperspectral data collected in May and November 2013. The absorption coefficients of phytoplankton (aph(λ)), colored dissolved organic matter (aCDOM(λ)), and nonpigmented suspended matter (ad(λ)) were determined. Nonpigmented components (ad(λ) and aCDOM(λ)) had a significant influence on the waterlogged areas, more so in Pj than in Gq because of its connection to the Ni River.The NIR-red, three-band, and normalized difference chlorophyll index (NDCI) models were calibrated and validated to estimate Chl-a concentrations ...

Given learning samples from a raster data set, spatial decision tree learning aims to find a decision tree classifier that minimizes classification errors as well as salt-and-pepper noise. The problem has important societal applications such as land cover classification for natural resource management. However, the problem is challenging due to the fact that learning samples show spatial autocorrelation in class labels, instead of being independently identically distributed. Related work relies on local tests (i.e., testing feature information of a location) and cannot adequately model the spatial autocorrelation effect, resulting in salt-and-pepper noise. In contrast, we recently proposed a focal-test-based spatial decision tree (FTSDT), in which the tree traversal direction of a sample is based on both local and focal (neighborhood) information. Preliminary results showed that FTSDT reduces classification errors and salt-and-pepper noise. This paper extends our recent work by introducing a new focal test approach with adaptive neighborhoods that avoids over-smoothing in wedge-shaped areas. We also conduct computational refinement on the FTSDT training algorithm by reusing focal values across candidate thresholds. Theoretical analysis shows that the refined training algorithm is correct and more scalable. Experiment results on real world data sets show that new FTSDT with adaptive neighborhoods improves classification accuracy, and that our computational refinement significantly reduces training time.

PLS models had the best potential to predict leaf N concentration.MTCI was the best spectral index to be used for variable rate nitrogen prescriptions.Applying the NSI formula to spectral data made it insensitive to external factors such as cultivar.Across N rates, spectral data with high variability often overestimates the N stress level.Canopy-scale spectral data can distinguish between N treatments better than tissue samples. To use remotely sensed spectral data for determining rates and timing of variable rate nitrogen (N) applications at a commercial scale, the most reliable indicators of crop N status must be determined. This study evaluated the ability of hyperspectral remote sensing to predict N stress in potatoes (Solanum tuberosum) during two growing seasons (2010 and 2011). Spectral data were evaluated using ground based measurements of leaf N concentration. Two canopy-scale hyperspectral images were acquired with an AISA-Eagle hyperspectral camera in both years. The experiment included five N treatments with varying rates and timing of N fertilizer and two potato cultivars, Russet Burbank (RB) and Alpine Russet (AR). Partial Least Squares regression (PLS) models resulted in the best prediction of leaf N concentration (r2=0.79, Root Mean Square Error of Cross Validation (RMSECV)=14% across dates for RB; r2=0.77, RMSECV=13% across dates for AR). Applying the Nitrogen Sufficiency Index (NSI) formula to spectral indices/models made them mostly insensitive to the effects of cultivar. The most promising technique for determining N stress in potato based on spectral indices was found to be the MERIS Terrestrial Chlorophyll Index (MTCI) due to a combination of relatively high r2 values, lower RMSECVs, and high accuracy assessment. Pairwise comparison tests from the means separation showed that spectral indices/models from the imagery resulted in more statistically significant groupings of crop stress levels for the spectra than leaf N concentration because canopy-scale spectral data are affected by both tissue N concentration and biomass. The results of this study suggest that upon proper sensor calibration, canopy-scale spectral data may be the most sensitive tool available to detect N status of a potato crop.

Comprehensive wetland inventories are an essential tool for wetland management, but developing and maintaining an inventory is expensive and technically challenging. Funding for these efforts has also been problematic. Here we describe a large-area application of a semi-automated process used to update a wetland inventory for east-central Minnesota. The original inventory for this area was the product of a labor-intensive, manual photo-interpretation process. The present application incorporated high resolution, multi-spectral imagery from multiple seasons; high resolution elevation data derived from lidar; satellite radar imagery; and other GIS data. Map production combined image segmentation and random forest classification along with aerial photo-interpretation. More than 1000 validation data points were acquired using both independent photo-interpretation as well as field reconnaissance. Overall accuracy for wetland identification was 90 % compared to field data and 93 % compared to photo-interpretation data. Overall accuracy for wetland type was 72 and 78 % compared to field and photo-interpretation data, respectively. By automating the most time consuming part of the image interpretations, initial delineation of boundaries and identification of broad wetland classes, we were able to allow the image interpreters to focus their efforts on the more difficult components, such as the assignment of detailed wetland classes and modifiers.