Forest Cover Types Research Articles

Relationships among forest type, watershed characteristics, and watershed ET in rural basins of the Southeastern US

Evapotranspiration (ET) typically accounts for 60–70% of precipitation in rural basins of the Southeastern United States. Since 1930, substantial reforestation of former croplands has occurred in the Piedmont and Appalachian Highlands in this area, leading to an expected increase in ET and reduction in baseflow. This study examines relationships between basin vegetative cover, abiotic factors, and water-budget partitioning in 45 USGS-gaged rural basins in the Southeastern US. Data are for the 1982–2014 water years with watersheds having ≥40% forest cover, crystalline-rock aquifers, minimal basin water export, and no large reservoirs. Long-term annual ET is calculated using the water-budget equation (ET = P-Q), which ranges from 641 to 971 mm/yr. (median 824). Vegetative cover and other basin variables are regressed against ET to quantify the effects of vegetative and forest types. Budyko analysis is employed to compare the watersheds and to evaluate factors affecting residuals. Regression analysis indicates that ET behavior is best explained by abiotic factors (i.e., precipitation and temperature) but forest-cover type also has some effect. Evergreen forest cover is less common than deciduous or mixed forest but has a positive relationship with ET, while deciduous and total forest have negative relationships with ET. Comparison of water-balance and Budyko-estimated ET indicates that deciduous and total forest are associated with negative residuals while evergreen is not significant. These results show that: forest cover effects on basin ET are complicated; forest-cover type is important for water-yield management in this region, and abiotic basin characteristics exert stronger control than forest cover on ET.

Effects of stand variables on stemflow and surface runoff in pine-oak forests in northern Mexico

The flow of water in temperate forests depends on the amount of precipitation, type of soil, topographic features, and forest cover, among other factors. Unlike the first three, forest cover can be modified by silvicultural treatments, the effects of which manifest in the quality and quantity of water, as well as in the transport of sediments and soil nutrients. The objective of this study was to analyze the effect of some stand variables on surface runoff and stemflow in pine-oak forests of northern Mexico. The stand variables included tree diameter at breast height, basal area, canopy cover, and volume. They were collected in eight 0.1-ha circular plots, measured in 2016 and re-measured in 2018. Nonlinear quantile regression was used to determine the best-fit relationships between the variables. Results indicated that surface runoff was most closely and inversely related to basal area. Stemflow was related to diameter at breast height, while showing no statistical significance. A stemflow funneling ratio did show an inverse, statistically-significant relationship with diameter at breast height. These results can help determine best forest management regimes compatible with the quantity and quality of water fluxes in this type of ecosystem.

Landscape implications of managing forests for carbon sequestration

Abstract We explore the implications of managing forests for the dual purpose of sequestering carbon and producing timber, using a model of the forest sector that includes a Hartman-based representation of forest owners’ behaviour as well as heterogeneity in environmental conditions. We focus on France, where recent policies aim at increasing the carbon sink and where the diversity of forests makes an analysis of spatial dynamics relevant, and we use recent estimates of the shadow price of carbon consistent with the country’s climate commitments. Results suggest that forests may sequester up to 550 MtCO2eq by 2100, driven by changes in harvest levels and species choice, whilst rotation lengths increase overall. A spatial analysis reveals a high spatial variability for these trends, highlighting the importance of considering the local context. Changes in investment patterns affect the spatial distribution of forest cover types: by the end of the century, a majority of regions comprise a larger share of older, multiple-species and mixed-structure forests. Whilst such an evolution may present benefits in terms of biodiversity, ecosystem services provision and resilience, it raises questions regarding the adequacy of such developments with current forest policy, which also aims at increasing harvest levels. An overall mitigation strategy for the forest sector would likely include incentives to energy and material substitution in downstream industries, which we did not consider and may interact with sequestration incentives.

Multiple wildfires with minimal consequences: Low-severity wildfire effects on West Texas piñon-juniper woodlands

Piñon-juniper (PJ) woodlands comprise an important and extensive dry-site forest cover-type that has historically experienced high spatiotemporal variation in fire frequency and extent. We re-measured a network of permanent monitoring plots in a fire-suppressed PJ site the Davis Mountains of West Texas after prescribed fire (Rx) treatments and three subsequent wildfires to 1) quantify wildfire effects on PJ woodland stand structure and fuel loadings, and 2) evaluate the effects of Rx fires followed by wildfire. Although fire weather was extreme during the wildfire years, all three wildfires burned as low-severity fire events. Simultaneous autoregressive modeling revealed that total tree density and basal area declined significantly (P < 0.05) over the time-series, while surface fuel loadings increased significantly (P < 0.05), likely in response to fuel inputs from fire-induced tree mortality. However, tree mortality largely occurred in smaller tree size-classes (i.e. < 25 cm diameter at breast height). Neither fire severity, nor changes in stand structure varied significantly over the time-series among recent fire history types (i.e. no fire, Rx fire only, Rx fire + wildfire, or wildfire only) indicating that Rx fires had little effect on subsequent wildfire effects on forest structure or species composition. The low-severity nature of these recent wildfires under severe fire weather and in the wake of almost a century of fire suppression suggests that just because fires have been absent from a site for decades contemporary wildfires will not always result in fire regime characteristics that differ from historical fires in PJ-dominated areas.

An Analysis of Fast Learning Methods for Classifying Forest Cover Types

ABSTRACT Proper mapping and classification of Forest cover types are integral in understanding the processes governing the interaction mechanism of the surface with the atmosphere. In the presence of massive satellite and aerial measurements, a proper manual categorization has become a tedious job. In this study, we implement three different modest machine learning classifiers along with three statistical feature selectors to classify different cover types from cartographic variables. Our results showed that, among the chosen classifiers, the standard Random Forest Classifier together with Principal Components performs exceptionally well, not only in overall assessment but across all seven categories. Our results are found to be significantly better than existing studies involving more complex Deep Learning models.

Surface soil temperature seasonal variation estimation in a forested area using combined satellite observations and in-situ measurements

Surface soil temperature is the soil temperature from the surface to 10 cm in depth. Surface soil temperature plays a significant role in agricultural drought monitoring, ecosystem energy transfer modeling, and global carbon cycle evaluation. Studies have been proposed to estimate surface soil temperature, but surface soil temperature monitoring within forested areas still poses a significant challenge. In this study, we proposed a surface soil temperature retrieval method using combined satellite observations and in-situ measurements for the Great Dismal Swamp (GDS). The GDS is a U.S. protected area managed and protected by the U.S. Fish and Wildlife Service. It is located along the boundary of Virginia and North Carolina, with maple gum, Atlantic white cedar, and pine pocosin as the main forest cover types. Ground-based surface soil temperature measurements were collected for these forest types from May 2015 to April 2017. Both the Land Remote Sensing Satellite (Landsat) Thermal Infrared Sensor (TIRS) and the Moderate Resolution Imaging Spectroradiometer (MODIS) carry two thermal infrared (TIR) channels. The TIR channels with similar corresponding wavelengths were first fused using an improved fusing model to generate high resolution TIR measurements. Then the enterprise algorithm was applied to calculate land surface temperature (LST) from the fused TIR bands. An improved soil temperature retrieval method was applied to generate surface soil temperature based on LST and vegetation index (VI) within the study area for the three forest types. In-situ measurements were used to build the surface soil temperature retrieval method, and results were then validated. The normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI) were integrated separately as VIs in the model to monitor surface soil temperature. The R2 for retrieved surface soil temperature through satellite observations was 0.76, and the RMSE was 1.96 ℃ when NDVI was integrated in the model; the R2 was 0.78, and the RMSE was 1.85 ℃ when EVI was used.

Estimation of uneven-aged forest stand parameters, crown closure and land use/cover using the Landsat 8 OLI satellite image

This study used the Landsat 8 OLI satellite image and the supervised classification method to estimate uneven-aged forest stand parameters and land use/cover. The spatial success of classification was also investigated. The overall success rates and Kappa values of the classification were, respectively, 74.7% and 0.75 for actual structural type, 84.6% and 0.80 for crown closure, and 88.35% and 0.81 for land use class, whereas the spatial success of classification on the forest cover type map was 36.91% for actual structural type, 64.74% for crown closure, and 41.78% for land use/cover class. The results revealed that the Landsat 8 OLI image can be used to identify stand parameters and land use/cover class. However, because the spatial success rates were below 50% for the actual structural type and land use/cover class of the stand types, it is not suitable for use in spatial classification determination for these classes.

A reversible database watermarking method non-redundancy shifting-based histogram gaps

In relational databases, embedding watermarks in integer data using traditional histogram shifting method has the problem of large data distortion. To solve this problem, a reversible database watermarking method without redundant shifting distortion is proposed, taking advantage of a large number of gaps in the integer histogram. This method embeds the watermark bit by bit on the basis of grouping. First, an integer data histogram is constructed with the absolute value of the prediction error of the data as a variable. Second, the positional relationship between each column and the gap in the histogram is analyzed to find out all the columns adjacent to the gap. Third, the highest column is selected as the embedded point. Finally, a watermark bit is embedded on the group by the histogram non-redundant shifting method. Experimental results show that compared with existing reversible database watermarking methods, such as genetic algorithm and histogram shift watermarking and histogram gap–based watermarking, the proposed method has no data distortion caused by the shifting redundant histogram columns after embedding watermarks on forest cover type data set and effectively reduces the data distortion rate after embedding watermarks.

Spatiotemporal evolution of tropical forest degradation and its impact on ecological sensitivity: A case study in Jinghong, Xishuangbanna, China

Due to rapid urbanization and a growing population, the tropical forest in southwestern China has experienced a dramatic shrinkage, which threatens its biodiversity and imposes limitations to sustainable development. Spatiotemporal change analysis and ecological sensitivity assessment are the important prerequisites for investigating the relationship between eco-environmental quality and human activities. In this study, the tropical forest and other land cover types in Jinghong, China were firstly classified by a machine learning classification algorithm (support vector machine, SVM) with 7 pairs of remote sensing (RS) data (from 1989 to 2018). Then the spatiotemporal change patterns were analyzed. The ecological sensitivity was evaluated based on an index system method (ISM) in which a weighted combination of eleven indicators were produced using an analytic hierarchy process (AHP) method and GIS. Meanwhile four individual sensitivity indicators, including biodiversity sensitivity (BS), water resources sensitivity (WRS), geological hazard sensitivity (GHS) and soil erosion sensitivity (SES) were assessed respectively to create a multi-perspective understanding of the entire ecological sensitivity. The results suggest that the tropical forest experienced a continual decrease from 5631.78 km2 in 1999 to 4216.23 km2 in 2018 with an average change rate of −1.49%. The decreased area was mainly encroached on by human settlements and agriculture, particularly in the south of Jinghong. Furthermore, it could be seen that urbanization is the key driver for the changes to ecological sensitivity with both positive and negative impacts. In Jinghong, the region covered by a tropical forest has a relatively higher comprehensive ecological sensitivity (CES) than that of an urban area. This work shows RS and GIS to be powerful tools providing profound insights to researchers with regard to the spatiotemporal evolution of tropical forests and ecological sensitivity. The results are significant for improving policies in order to keep a sustainable balance in regional ecosystem management.

Dinámica forestal y uso de suelo en las cuencas que integran al municipio Tomatlán, Jalisco

Se analizó la dinámica forestal y usos de suelo en las cuencas hidrográficas que integran al municipio Tomatlán, Jalisco (1997-2017). Para cumplir con este objetivo se clasificaron tres imágenes de satélite: Landsat 5 y 8 (P030 y R046). Como resultado se generaron dos mapas temáticos con siete clases de coberturas forestales y usos de suelo: zonas urbanas, agricultura, pastizal, cuerpos de agua, selva mediana subcaducifolia, selva baja caducifolia, además del bosque templado de coníferas y latifoliadas. El mapa se comparó para calcular el cambio de cobertura presente en los tiempos estudiados. Se determinó la tasa de cambio para el periodo 1997-2017; el porcentaje de coberturas que se mantuvo fue de 77 % y el cambio correspondió a 23 %. Los resultados muestran que, a lo largo del intervalo de tiempo analizado, la cobertura forestal que más ganancias tuvo fue la selva mediana subcaducifolia, con un incremento de 1 %; los cuerpos de agua presentaron una tasa de disminución de 1.2 % por año. Se considera que la ganancia o el incremento de esta cobertura se relaciona con el abandono de áreas agrícolas y pecuarias; lo que ha implicado la regeneración de selva mediana subcaducifolia de segundo crecimiento; mientras que, los cuerpos de agua están disminuyendo, principalmente, en las partes bajas de las cuencas debido al incremento de las zonas agrícolas; lo cual puede tener implicaciones en los cambios del paisaje natural.

Long-term effects of harvest on boreal forest soils in relation to a remote sensing-based soil moisture index

Storing a significant portion of the global carbon (C) stocks, soils of the boreal forest display a high degree of spatial heterogeneity across the landscape, arising from variation in forest structure and landscape morphology, as well as natural and anthropogenic disturbances. Because of this high degree of variability, accurately quantifying C storage in this ecosystem poses a challenge. Forestry is an important feature of Canada’s natural resource-based economy, but there is still considerable uncertainty on how management practices will affect boreal soil C sequestration in the long-term. With increasing pressures due to a changing climate and intensified forest management, developing better tools and techniques to quantify soil C dynamics is of paramount importance. In this study, we measured soil C stocks and associated properties in the forest floors and mineral soils (0–7 cm) of conifer-dominated and deciduous (broadleaf)-dominated boreal forest stands, 17 years following variable retention harvest. We investigated if the Wet Areas Mapping-based depth-to-water (DTW) index, derived from remotely sensed Light Detection and Ranging (LiDAR) data, was related to soil properties of unharvested stands. In addition, relationships between harvest-induced changes in soil properties and the DTW index were examined. In unharvested stands, several forest floor properties were related to DTW. However, these relationships were stand-specific. In unharvested conifer-dominated stands, forest floor C stocks were positively related to site wetness, while in harvested stands there was no relationship; this suggests that C was differentially lost from wetter sites following harvesting. Conversely, in unharvested deciduous-dominated stands, there was no relationship between forest floor C stocks and site wetness, but in harvested stands, wetter sites had higher C stocks. The DTW index was more strongly related to soil properties in the mineral soil (0–7 cm) than in the forest floor. In both forest cover types, mineral soil C and N concentrations, and C stocks, increased with increasing wetness. Relationships between mineral soil properties and DTW were not stand-specific, and were of similar magnitude under deciduous-dominated and conifer-dominated cover. In addition, harvest and forest regeneration had limited impact on the relationships between DTW and mineral soil properties. This study highlights the potential of using remote sensing and the DTW index to model forest floor and mineral soil properties in the boreal forest. In turn, this approach may be utilized in effective forest management strategies that aim to conserve boreal C stocks.

Color-Based Forest Cover Type Image Segmentation using K-Means Clustering Approach

In order to understand forest composition, classifying forest cover type can help research regarding forest resilience, carbon sequestration, and climate change concerns. The purposes of this study were to develop and implement some image processing functions based on the histogram of forest cover type color image, and to classify forest cover type using its color feature sets of image pixels. Color-based image segmentation that is based on the color feature of image pixels assumes that homogeneous colors in the image correspond to separate clusters and hence meaningful objects in the image. The Image Processing Toolbox of MATLAB R2019a was used to convert the original forest cover type image to the enhance contrast image, including histogram of enhance contrast image. Furthermore, It was also used to analyze color-based forest cover type image segmentation using the enhance contrast image for this study. Using K-Means clustering analysis, a three-cluster solution was developed, labeled as Hardwoods (Yellow Color) Cover Type, Hardwoods (Gray Color) Cover Type, and Loblolly Pines Cover Type. There was a significant difference among three different forest cover type clusters in terms of histograms and L*a*b* color space features visually.

Modelling habitat suitability of western tragopan (Tragopan melanocephalus) a range-restricted vulnerable bird species of the Himalayan region, in response to climate change

Climate change is expected to alter the structure and functions of an ecosystem including species composition and its geographical distribution. There is limited understanding on how the habitat of the Himalayan range-restricted species would be affected under the influence of climate change. In the present study, we model the climate change impacts on habitat suitability of western tragopan (Tragopan melanocephalus), a range-restricted and vulnerable bird species in the Indian western Himalayas. The climate change scenarios of IPCC represented by representative concentration pathways (RCPs) viz. RCP 4.5, RCP 6.0, and RCP 8.5 were considered for assessing the habitat suitability for the year 2050 and 2070. Most influencial variables that may be linked to habitat suitability of Tragopan, such as bioclimatic variables, land use (forest cover and forest type), soil characteristics, and topographic variables (elevation, slope, aspect, heat load index) were considered to develop a model using the Maximum Entropy (MaxEnt) algorithm. MaxEnt is a widely used and accepted tool for modeling species distribution. The model’s performance for mapping habitat was evaluated by the Area Under Curve (AUC) (AUC > 90%). The measured TSS value 0.98 and Kappa value 0.71 were elaborated for the aptness of the model for suitable habitat mapping. It was observed that the suitable habitat of the western tragopan would shift towards higher elevations under all RCPs. The study would benefit to biodiversity conservators and policymaker for formulating future strategy and planning for conservation and management of the Himalayan range-restricted bird species. The approach of this study could be replicated with other range-restricted Himalayan bird species for future projections of suitable habitat.

Влияние типа лесного покрова подстилающей поверхности на уровень отраженных от нее радиосигналов спутниковых радионавигационных систем при полетах воздушных судов на малых высотах

The purpose of the study was to analyze the influence of the type of the underlying surface forest cover on the level of SNS radio signals reflected from the surface and, accordingly, on the aircraft user equipment efficiency at low-altitude flying. By mathematical simulation, we analyzed the dependences of the signal-to-noise ratio on the elevation angle of navigation spacecraft of the working constellation and the aircraft flight altitude for different types of forest cover. Findings of research show that the rugged barren terrain is characterized by the highest noise level, while the lowest noise level is typical for the thin forest. This can be explained by the fact that most electromagnetic waves penetrate deep into the forest, where they are absorbed due to multiple reflections. Moreover, the thicker the forest cover, the lower the degree of absorption and the higher the noise level.

Challenges in estimating forest biomass: use of allometric equations for three boreal tree species

Regionally fitted allometric equations for individual trees and root-to-shoot ratio values are normally used to estimate local aboveground and belowground forest biomass, respectively. However, uncertainties arising from such applications are poorly understood. We developed equations for both aboveground and belowground biomass using destructive sampling for three dominant upland boreal tree species in northwestern Alberta, Canada. Compared with our equations, the diameter-based national equations derived for use across Canada underestimated aboveground biomass for Picea glauca (Moench) Voss but gave reasonable estimates for Populus balsamifera L. and Populus tremuloides Michx. The national equations based on both tree diameter and height overestimated aboveground biomass for the Populus species but underestimated it for Picea glauca in our study area. The approach of root-to-shoot ratio proposed by the Intergovernmental Panel on Climate Change (IPCC) overestimated belowground biomass by 16%–41%, depending on forest cover type, in comparison with our values estimated directly on site, with the greatest bias in deciduous-dominated stands. When the general allometric equations for aboveground biomass and the root-to-shoot ratio for belowground biomass were combined to estimate stand biomass, overestimation could be as high as 18% in our study area. The results of our study support the development of improved regional allometric equations for more accurate local-scale estimations. Incorporating intraspecific variation of important traits such as tree taper may be especially helpful.

Genetic and morphological diversity of mouse lemurs (Microcebus spp.) in northern Madagascar: The discovery of a putative new species?

Tropical forests harbor extremely high levels of biological diversity and are quickly disappearing. Despite the increasingly recognized high rate of habitat loss, it is expected that new species will be discovered as more effort is put to document tropical biodiversity. Exploring under-studied regions is particularly urgent if we consider the rapid changes in habitat due to anthropogenic activities. Madagascar is known for its extraordinary biological diversity and endemicity. It is also threatened by habitat loss and fragmentation. It holds more than 100 endemic primate species (lemurs). Among these, Microcebus (mouse lemurs) is one of the more diverse genera. We sampled mouse lemurs from several sites across northern Madagascar, including forests never sampled before. We obtained morphological data from 99 Microcebus individuals; we extracted DNA from tissue samples of 42 individuals and amplified two mitochondrial loci (cytb and cox2) commonly used for species identification. Our findings update the distribution of three species (Microcebus tavaratra, Microcebus arnholdi, and Microcebus mamiratra), including a major increase in the distribution area of M. arnholdi. We also report the discovery of a new Microcebus lineage genetically related to M. arnholdi. Several complementary approaches suggest that the newly identified Microcebus lineage might correspond to a new putative species, to be confirmed or rejected with additional data. In addition, morphological analyses showed (a) clear phenotypic differences between M. tavaratra and M. arnholdi, but no clear differences between the new Microcebus lineage and the sister species M. arnholdi; and (b) a significant correlation between climatic variables and morphology, suggesting a possible relationship between species identity, morphology, and environment. By integrating morphological, climatic, genetic, and spatial data of two northern Microcebus species, we show that the spatial distribution of forest-dwelling species may be used as a proxy to reconstruct the past spatial changes in forest cover and vegetation type.

Influence of residential land cover on childhood allergic and respiratory symptoms and diseases: Evidence from 9 European cohorts

IntroductionRecent research focused on the interaction between land cover and the development of allergic and respiratory disease has provided conflicting results and the underlying mechanisms are not fully understood. In particular, green space, which confers an overall positive impact on general health, may be significantly contributing to adverse respiratory health outcomes. This study evaluates associations between surrounding residential land cover (green, grey, agricultural and blue space), including type of forest cover (deciduous, coniferous and mixed), and childhood allergic and respiratory diseases. MethodsData from 8063 children, aged 3–14 years, were obtained from nine European population-based studies participating in the HEALS project. Land-cover exposures within a 500 m buffer centred on each child's residential address were computed using data from the Coordination of Information on the Environment (CORINE) program. The associations of allergic and respiratory symptoms (wheeze, asthma, allergic rhinitis and eczema) with land coverage were estimated for each study using logistic regression models, adjusted for sex, age, body mass index, maternal education, parental smoking, and parental history of allergy. Finally, the pooled effects across studies were estimated using meta-analyses. ResultsIn the pooled analyses, a 10% increase in green space coverage was significantly associated with a 5.9%–13.0% increase in the odds of wheezing, asthma, and allergic rhinitis, but not eczema. A trend of an inverse relationship between agricultural space and respiratory symptoms was observed, but did not reach statistical significance. In secondary analyses, children living in areas with surrounding coniferous forests had significantly greater odds of reporting wheezing, asthma and allergic rhinitis. ConclusionOur results provide further evidence that exposure to green space is associated with increased respiratory disease in children. Additionally, our findings suggest that coniferous forests might be associated with wheezing, asthma and allergic rhinitis. Additional studies evaluating both the type of green space and its use in relation to respiratory conditions should be conducted in order to clarify the underlying mechanisms behind associated adverse impacts.

Change detection of landscape connectivity arisen by forest transformation in Hazaribagh wildlife sanctuary, Jharkhand (India)

Forest land conversion is the primary driver of biodiversity decline worldwide. Hazaribagh wildlife sanctuary is a region of rich biodiversity in which forests and wildlife are deteriorating fast. The prime reasons for forest degradation and wildlife loss are the landscape connectivity weakening and forest transformation. In the present work, landscape connectivity and forest transformation relationships were analyzed in a spatio-temporal domain. The forest patches as a group of spectral abundance were extracted using the endmember retrieval technique. The connectivity analysis was performed by using a connectivity index in the extracted forest patches. Forest transformation is calculated using a post-classification change detection strategy for five types of forest cover during the four phases of the year (1992–2005, 2005–2010, 2010–2017 and 1992–2017). The forest cover was measured using a forest canopy density model using spectral indices. The landscape connectivity of 80–100% exhibit a rapid increase of 38% in 2005 from 1992 contrary to a 13% decrease in 2010 and 2017. The 23% loss of forest cover from 2005 to 2010 and a 17% loss in 2010–2017 phase of forest transformation weakened the forest connectivity. Forest cover, having a density higher than 40% was more vulnerable to degradation and landscape connectivity loss. The result shows that such declines of forest cover and landscape connectivity will reduce the genetic diversity in the forest, especially the mammalian population.

Growing neural gas with random projection method for high-dimensional data stream clustering

High-dimensional data streams emerge ubiquitously in many real-world applications such as network monitoring and forest cover type. Clustering such data streams differs from traditional data clustering algorithm where given datasets are generally static and can be repeatedly read and processed, thus facing more challenges due to having to satisfy such constraints as bounded memory, single-pass, real-time response and concept drift detection. Recently, many methods of such type have been proposed. However, when dealing with high-dimensional data, they often result in high computational cost and poor performance due to the curse of dimensionality. To address the above problem, in this paper, we present a new clustering algorithm for data streams, called RPGStream, by combining the random projection method with the growing neural gas (GNG) model which is an incremental self-organizing approach, belonging to the family of topological maps such as SOM or neural gas. To gain insights into the performance improvement obtained by our algorithm, we analyze and identify the major influence of random projection on GNG. Although our method is embarrassingly simple just by incorporating the random projection into an exponential fading function of GNG, the experimental results on variety of benchmark datasets indicate that our method can still achieve comparable or even better performance than G-Stream algorithm even if the raw dimension is compressed up to 10% of the original one (e.g., for CoverType dataset, its dimension is reduced from 54 to 5).

Adaptability of the threatened jaguarundi (Herpailurus yagouaroundi Schereber, 1777) to human-altered environments in San Luis Potosí, Mexico

Persistence and recovery of rare species in developing regions with limited protected areas depends upon their adaptability to human-altered habitats. The jaguarundi (Herpailurus yagouaroundi) is classed as threatened in Mexico, and knowledge of its distribution and environmental correlates is necessary for informed recovery efforts. However, little is known of jaguarundi habitat or distribution in interior Mexico, including the state of San Luis Potosí (SLP). We conducted 96 semi-structured interviews around communities, ejidos, and ranches throughout SLP to obtain records of jaguarundi presence and identify environmental correlates and site attributes associated with its occurrence. We evaluated interviews using analytical criteria of credibility, and collected habitat information from 50 reliable occurrences from three of the four geographic regions of SLP. Compared to the SLP landscape, jaguarundi occurrences were located closer to water, closer to roads, at lower elevation, marginally closer to communities, and in areas with greater total edge, edge density, and number of landscape patches. Jaguarundi showed preference for mosaics of tropical forest, agricultural, grassland, and urban (i.e., any community) cover types. Relatively dense hiding or ambush cover was usually present at occurrence sites. Collectively, maximum entropy modeling and logistic regression modeling predicted similar and high likelihood of jaguarundi presence in regions characterized by mosaics of tropical forest, agriculture, grassland, or urban cover types &lt;500 m in elevation and &lt;2 km from roads. These mosaic landscapes tended to be relatively close to communities of moderate population densities and water, and typically support higher small prey densities than less fragmented areas. Jaguarundi were adaptable to at least light–moderate human-related disturbance, and may be benefitted by it because of increased edge and habitat mosaics.