Matlab Software for Supervised Habitat Mapping of Freshwater Systems Using Image Processing

We present a software package for the supervised classification of images useful for cover-type mapping of freshwater habitat (e.g., water surface, gravel bars, vegetation). The software allows the user to select a representative subset of pixels within a specific area of interest in the image that the user has identified as a cover-type habitat of interest. We developed a graphical user interface (GUI) that allows the user to select single pixels using a dot, line, or group of pixels within a defined polygon that appears to the user to have a spectral similarity. Histogram plots for each band of the selected ground-truth subset aid the user in determining whether to accept or reject it as input data for the classification processes. A statistical model, or classifier, is then built using this pixel subset to assign every pixel in the image to a best-fit group based on reflectance or spectral similarity. Ideally, a classifier incorporates both spectral and spatial information. In our software, we implement quadratic discriminant analysis (QDA) for spectral classification and choose three spatial methods—mode filtering, probability label relaxation, and Markov random fields—to incorporate spatial context after computation of the spectral type. This multi-step interactive process makes the software quantitatively robust, broadly applicable, and easily usable for cover-type mapping of rivers, their floodplains, wetlands often components of these functionally linked freshwater systems. Indeed, this supervised classification approach is helpful for a wide range of cover-type mapping applications in freshwater systems but also estuarine and coastal systems as well. However, it can also aid many other applications, specifically for automatic and quantitative extraction of pixels that represent the water surface area of rivers and floodplains.

Semi-automated land cover mapping using an ensemble of support vector machines with moderate resolution imagery integrated into a custom decision support tool

Land cover type is a fundamental remote sensing-derived variable for terrain analysis and environmental mapping applications. The currently available products are produced only for a single season or a specific year. Some of these products have a coarse resolution and quickly become outdated, as land cover type can undergo significant change over a short time period. In order to enable on-demand generation of timely and accurate land cover type products, we developed a sensor-agnostic framework leveraging pre-trained machine learning models. We also generated land cover models for Sentinel-2 (20m) and Landsat 8 imagery (30m) using either a single date of imagery or two dates of imagery for mapping land cover type. The two-date model includes 11 land cover type classes, whereas the single-date model contains 6 classes. The models’ overall accuracies were 84% (Sentinel-2 single date), 82% (Sentinel-2 two date), and 86% (Landsat 8 two date) across the continental United States. The three different models were built into an ArcGIS Pro Python toolbox to enable a semi-automated workflow for end users to generate their own land cover type maps on demand. The toolboxes were built using parallel processing and image-splitting techniques to enable faster computation and for use on less-powerful machines.

Urban Heat and Urban Cool Islands of Different Land Use Types in Cuiabá, Brazil

Urban heat islands (UHIs), urban cool islands (UCIs), and their varying effects due to land use/land cover types and the local climate were investigated from 2014 to 2015 in three urban zones located in Cuiabá city, Brazil, during hot-humid, and hot-dry periods. All the urban zones were analysed for land use/land cover type, local climate, and rate of warming and cooling based on the difference in air temperature (ΔT) between the urban zones and the rural zone located outside the urban perimeter. The annual UHI effect in all the urban zones exhibited varying intensities during the day, with the highest daytime intensity recorded after the sunrise. The duration of UHI effect varied with land use/land cover type; a consequence of high built-up density, verticalization, waterproof surface, and other peculiarities of urban areas. In the urban zones with high built-up density, the duration of UHI effect was observed for up to 24 h, while in the urban areas with low built-up density, the maximum duration of UHI effect was 8 h. On an average, during the daytime, the urban zone with approximately 70% of vegetation cover and water bodies recorded a UCI value of approximately –8 °C, whereas the urban zones with approximately 80% waterproof surface and bare land recorded a UCI value close to +2 °C during the hot-dry and hot-humid periods. The results indicate that land use and land cover types directly influence UHI intensity.

Assessment of Dynamic Land System in Nilgiri Biosphere Reserve Using MODIS Derived Temporal Data Sets during 2001 to 2018

The Nilgiri Biosphere Reserve (NBR) is one of the largest protected ecologically sensitive areas in India. This study examined the land use/land cover (LULC) changes in NBR for past 18 years from 2001 to 2018 to figure out the LULC changed within a protected area using datasets in 2001, 2010, and 2018 with the help of pertinent geospatial techniques. MODIS Land Cover Type Product (MCD12Q1) accuracy was quantitatively analyzed based on ground truth data and Google Earth imagery. Validation of data were assessed using and overall 635 locations for its accuracy assessment. The obtained kappa coefficient of 0.75, denotes the classification has moderate accuracy. The results showed that in the past 18 years, woody savannas and grasslands were reduced by 299.47 sq.km and 155.32 sq.km respectively. The areas of croplands and cropland/natural vegetation mosaics were also increased by 34.84 sq.km and 54.41 sq.km, respectively. These results showed anthropogenic influences through agricultural practices within the NBR buffer zones. The mixed forests were increased by 266.01 sq.km. One of the significant changes was seen in closed shrublands, which were absent in 2018, that covered 1.50 sq.km in 2001. In addition, A gradual decrease in the area were noticed in woody savannas. From the outcomes, it is recommended that the LULC classes that cover minimal area may be unstable, so measures should be taken for their conservation. The study proved the usefulness of MODIS land cover type data in monitoring large areas periodically for quick decision-making.

Integrated Evaluation of Vegetation Drought Stress through Satellite Remote Sensing

In the coming decades, Bulgaria is expected to be affected by higher air temperatures and decreased precipitation, which will significantly increase the risk of droughts, forest ecosystem degradation and loss of ecosystem services (ES). Drought in terrestrial ecosystems is characterized by reduced water storage in soil and vegetation, affecting the function of landscapes and the ES they provide. An interdisciplinary assessment is required for an accurate evaluation of drought impact. In this study, we introduce an innovative, experimental methodology, incorporating remote sensing methods and a system approach to evaluate vegetation drought stress in complex systems (landscapes and ecosystems) which are influenced by various factors. The elevation and land cover type are key climate-forming factors which significantly impact the ecosystem’s and vegetation’s response to drought. Their influence cannot be sufficiently gauged by a traditional remote sensing-based drought index. Therefore, based on differences between the spectral reflectance of the individual natural land cover types, in a near-optimal vegetation state and divided by elevation, we assigned coefficients for normalization. The coefficients for normalization by elevation and land cover type were introduced in order to facilitate the comparison of the drought stress effect on the ecosystems throughout a heterogeneous territory. The obtained drought coefficient (DC) shows patterns of temporal, spatial, and interspecific differences on the response of vegetation to drought stress. The accuracy of the methodology is examined by field measurements of spectral reflectance, statistical analysis and validation methods using spectral reflectance profiles.

Potential Distribution of Colonizing Nine-Banded Armadillos at Their Northern Range Edge

The nine-banded armadillo (Dasypus novemcinctus) has become a recent addition to the local fauna of Illinois as a response to habitat alteration and climate change. This range expansion has resulted in the presence of armadillos in areas not predicted by earlier models. Although these models have been revised, armadillos continue to move north and have reached areas of heavy agricultural use. We identified conditions that favor the presence of armadillos and potential corridors for dispersal. Identifying the distribution of the armadillo in Illinois is a vital step in anticipating their arrival in areas containing potentially sensitive wildlife populations and habitats. Armadillo locations (n = 37) collected during 2016–2020 were used to develop a map of the potential distribution of armadillos in southern Illinois. Environmental data layers included in the model were land cover type, distance to water, distance to forest edge, human modification, and climactic variables. Land cover type was the most important contributing variable to the model. Our results are consistent with the tenet that armadillo activity and dispersal corridors are centered around riparian areas, and that forested cover may provide corridors an agricultural mosaic.

Combined Effects of Wildfire and Vegetation Cover Type on Volcanic Soil (Functions and Properties) in a Mediterranean Region: Comparison of Two Soil Quality Indices

Mediterranean regions are the most impacted by fire in Europe. The effects of fire on soil greatly vary according to several factors such as vegetation cover type, but they are scarcely studied. Therefore, this research aimed at evaluating the combined impacts of fire and vegetation on single soil characteristics and on the overall soil quality and functionality through two soil quality indices, simple additive (SQI) and a weighted function (SQIFUNCT). In order to reach the aims, burnt and unburnt soils were collected under different vegetation cover types (herbs and shrubs, black locust, pine and holm oak) within the Vesuvius National Park. The soils were analyzed for the main abiotic (water and organic matter content, total C, N, Ca, K, Cu and Pb concentrations, C/N ratio) and biotic (microbial and fungal biomasses, basal respiration, β-glucosidase activity) characteristics. On the basis of the investigated soil characteristics, several soil functions (water retention, nutrient supply, contamination content, microorganism habitat and activities), and the soil quality indices were calculated. The results showed that the impact of fire on soil quality and functionality was mediated by the vegetation cover type. In fact, fire occurrence led to a decrease in water and C/N ratio under herbs, a decrease in C concentration under holm oak and a decrease in Cu and Pb concentrations under pine. Although the soil characteristics showed significant changes according to vegetation cover types and fire occurrence, both the additive and weighted function soil quality indices did not significantly vary according to both fire occurrence and the vegetation cover type. Among the different vegetation cover types, pine was the most impacted one.