Assessment of grassland use intensity by remote sensing to support conservation schemes

Grassland is a land cover in the area of conflict between agriculture and conservation, where intensification of land use is a major threat to grassland biodiversity. Grassland use intensity is a key factor for the conservation value of grassland, and detailed spatial data on grassland use intensity is needed to improve strategies for biodiversity conservation. A new remote sensing-based approach using multi-temporal high resolution RapidEye satellite data was developed in the present study that makes a large-scale assessment of grassland use intensity possible. RapidEye is a constellation of five satellites with 6.5 m spatial resolution, which allows frequent and timely image acquisition targeted at specific growing seasons. Semi-natural grassland, extensively used grassland, intensively used grassland and tilled grassland could be reliably differentiated at the management plot level in a study area in southern Germany. Various combinations of images from different observation dates have been tested as classification input and their overall classification accuracies were validated by field data. Best results were achieved using a combination of five multi-temporal scenes with an overall accuracy of 85.7%. A three-scene combination resulted in an overall accuracy of 82.2%. The analysis showed that seasonal aspects are very important when selecting adequate observation dates. Grassland use intensity was also assessed on peatlands using a peat soil map, since land use intensity significantly affects greenhouse gas emissions from peatlands. The results demonstrate the potential of targeted multi-spectral, high spatial resolution remote sensing for the large-scale monitoring of dynamic habitats, which is of vital importance to support various environmental conservation schemes through improved monitoring and reporting capabilities.     

The influence of rapid urbanization and land use changes on terrestrial carbon sources/sinks in Guangzhou,China

Complex changes in carbon sources and sinks caused by rapid urbanization have been observed with extensive changes in the quantity, structure, and spatial pattern of land use types. Based on the modified Carnegie-Ames-Stanford Approach model and on gray relational analysis, we analyzed the influence of land use changes on carbon sinks and emissions in Guangzhou from 2000 to 2012. The aim was to identify suitable options for built-up land expansion that would allow for minimal carbon losses. The key results were as follows: (1) Built-up land increased by 118.91% in Guangzhou city over the study period, with this expansion taking the form of concentric circles extending around the old Yuexiu district. (2) Carbon emissions over the period of analysis significantly exceeded carbon sink capabilities. The total carbon sink decreased by 30.02%, from 535.40 × 10³ t to 374.6 × 10³ t. Total carbon emissions increased by 1.89 times, from 13.73 × 10⁶ t to 39.67 × 10⁶ t; 80% of carbon emissions were derived from energy consumption. (3) There were large differences in the extent of carbon sink losses at different scales of built-up land expansion and land use change. In Guangzhou, the loss of carbon sink is small when cultivated land (though not prime farmland) and water bodies are converted to built-up land on a small scale. The loss of carbon sink is much smaller when grasslands are converted to built-up land on a large scale. However, forested land, which has excellent carbon sink functions, should not be converted. (4) Changes in carbon sinks were mainly affected by natural factors and land urbanization. Changes in carbon emissions were mainly affected by population urbanization, economic urbanization, and land urbanization. (5) To achieve “economical and intensive use of land”, “urban growth boundary” and “ecological red lines” should be determined for government policies on land use management. These factors have great significance for “increasing carbon sinks and reducing carbon emissions” in urban ecological systems.     

Assessing spatial distribution of Coffea arabica L. in Ethiopia's highlands using species distribution models and geospatial analysis methods

Though there is an increase in popularity of predictive modelling for assessing the geographical distribution of species, there is still a clear gap on explaining geospatial methods to derive the presence/absence of species in terms of geospatial extent besides the ambiguity of robust models. In this paper, we evaluate four major species distribution modelling methods: Artificial Neural Network (ANN), Support Vector Machines (SVM), Maximum Entropy (MaxEnt) and Generalized Linear Model (GLM) with pseudo absence and background absence data. To investigate the efficacy of these models, we present a case study using Coffea arabica L. species in Ethiopia as there was no species distribution modelling that has been done at a local scale especially in the coffee growing areas. We made predictions on 75% subsets and validation on 25% of the 112 presence of the species records that were collected from field observation and 0.5 m spatial resolution of true colour aerial photographs. Twelve biophysical explanatory variables; climatic, remote sensing based and landscape variables were employed in modelling. The results show that MaxEnt with pseudo absence data and SVM with background absence have highest area of understory coffee presence prediction with 12.2% and 23.1% area coverage of indigenous forest, respectively. The result from the model performance test using True Positive Rate (TPR) shows that GLM and SVM with pseudo absence data performed highest (TPR = 0.821). MaxEnt and SVM were the robust modelling methods (TPR = 0.964) using background absence data.     

Modelling potential wildlife-vehicle collisions (WVC) locations using environmental factors and human population density: A case-study from 3 state highways in Central California

Roads can exert direct and indirect impacts on ecosystems and organisms. In particular, wildlife-vehicle collisions (WVC) may be a considerable threat for populations of certain wildlife species. Despite such threat, there is still incomplete understanding of the factors responsible for high road mortality. Only a few empirical studies have tested the idea that spatial variation of roadkill is affected by environmental characteristics and socio-demographic factors. This study examines the relationships between WVC involving different taxonomic groups (i.e. ungulate, avian, medium mammal, small mammal) and physical and human population characteristics of landscapes by adapting the ecological model, Maxent, to distribution modelling of carcasses resulting from WVC. We used observations from the California Roadkill Observation System of four taxonomic groups recorded along highways in northern California. Our results indicated that current carcass-observation locations were explained primarily by total forest area (cells) within 500 m buffer and road density within a 500 m neighborhood. Our results found that current carcass-observation locations are modelled well using environmental variables and human population density together. Moreover, a comparison of projected potential roadkill locations based on environmental factors and human population density among different taxonomic groups revealed substantially different distributions. These results indicate potential areas where wildlife populations are at increased risk of coming into contact with traffic and the potential utility of this methodology for modelling current and future distributions of wildlife across landscapes using the Maxent approach.     

Modelling potential impacts of climate change on the bioclimatic envelope and conservation of the Maned Wolf (Chrysocyon brachyurus)

Forecasting the influence of climatic changes on the distribution of the Maned Wolf (Chrysocyon brachyurus) is important for the conservation of the species. We explored the environmental characteristics than best explain the current distribution of the species, modelled the past and present distribution, projected the niche model into the future, and identified suitable areas for conservation. Niche modelling was performed using Maxent and 21 environmental variables. For past conditions, we considered the Last Glacial Maximum (LGM) and the mid-Holocene (MH) climates. For future conditions, we used the A2a greenhouse gas emission scenario for 2050. Four General Circulation Models (FGOALS 1.0, HADCM3, IPSL-CM4 and MIROC 3.2) were used. The resulting niche model (AUC = 0.89 ± 0.02) predicts maximum probability of presence at precipitation of 106 mm during the coldest quarter, of 396 mm during the warmest quarter, and in totally flat areas. The suitable area for the Maned Wolf currently covers 4,320,364 km². For the LGM, there were inter-model differences in predicted areas (from 819,324 km² to 6,395,886 km²) and in geographic location. The MH models showed drastic changes with respect to the present and considerable inter-model variation. Predictions for 2050 show significant (at least 33%) reductions in distribution. Only a minor fraction (39%) of the current distribution can be considered stable for the period LGM-2050. The FGOALS model was the best option for projecting species occurrence into the future because it included the three localities known for the Maned Wolf from the late Pleistocene and predicts stable areas that coincide with spatial patterns of genetic diversity. The FGOALS projection for 2050 predicts a 33% reduction in suitable habitats, indicating some stable areas (central South America) that will probably be key sites for the conservation of the species.     

Comparison of boosted regression tree and random forest models for mapping topsoil organic carbon concentration in an alpine ecosystem

Soil organic carbon (SOC) plays an important role in soil fertility and carbon sequestration, and a better understanding of the spatial patterns of SOC is essential for soil resource management. In this study, we used boosted regression tree (BRT) and random forest (RF) models to map the distribution of topsoil organic carbon content at the northeastern edge of the Tibetan Plateau in China. A set of 105 soil samples and 12 environmental variables (including topography, climate and vegetation) were analyzed. The performance of the models was evaluated using a 10-fold cross-validation procedure. Maps of the mean values and standard deviations of SOC were generated to illustrate model variability and uncertainty. The results indicate that the BRT and RF models exhibited very similar performance and yielded similar predicted distributions of SOC. The two models explained approximately 70% of the total SOC variability. The BRT and RF models robustly predicted the SOC at low observed SOC values, whereas they underestimated high observed SOC values. This underestimation may have been caused by biased distributions of soil samples in the SOC space. Vegetation-related variables were assigned the highest importance in both models, followed by climate and topography. Both models produced spatial distribution maps of SOC that were closely related to vegetation cover. The SOC content predicted by the BRT model was clearly higher than that of the RF model in areas with greater vegetation cover because the contributions of vegetation-related variables in the two models (65% and 43%, respectively) differed significantly. The predicted SOC content increased from the northwestern to the southeastern part of the study area, average values produced by the BRT and RF models were 27.3 g kg⁻¹ and 26.6 g kg⁻¹, respectively. We conclude that the BRT and RF methods should be calibrated and compared to obtain the best prediction of SOC spatial distribution in similar regions. In addition, vegetation variables, including those obtained from remote sensing imagery, should be taken as the main environmental indicators and explicitly included when generating SOC maps in Alpine environments.     

Impact of landscape pattern at multiple spatial scales on water quality: A case study in a typical urbanised watershed in China

Buffer zones along rivers and streams can provide water quality services by filtering nutrients, sediment and other contaminants from the surface. Redundancy analysis was used to determine the influence of the landscape pattern at the entire catchment scale and at multiple buffer zone scales (100 m, 300 m, 500 m, 1000 m and 1500 m) on the water quality in a highly urbanised watershed. Change-point analysis was further applied to estimate the specific locations along a gradient of landscape metric that result in a sudden change in the water quality variable. The landscape characteristics for 100 m buffer zones appeared to have a slightly greater influence on the water quality than the entire catchment. The patch density of urban land and the large patch index of water were recognised as the dominant variables influencing the water quality for a 100 m buffer zone. The result of change-point analysis indicated key interval values of the two landscape metrics within the 100 m buffer zone. When the patch density of urban land was >30–40 n/100 ha and the largest patch index of water was >2.5–3.5%, the watershed water quality appeared to be better protected.     

Energy balance and greenhouse gas emissions in organic and conventional avocado orchards in Mexico

There is a worldwide growing awareness of the negative impacts of the increasing fossil fuel reliance and greenhouse gas (GHG) emissions from agriculture, in particular for intensive crop systems. We analyze the energy balances and greenhouse gas emissions from export-oriented avocado orchards in Mexico. Avocado is a very important export crop and one of the main drivers of land-use change in the country. We compared 12 avocado orchards under organic and conventional management during two production cycles (2010 and 2011) in a representative region of Central Mexico. Our analysis shows no significant differences in energy consumption and GHG emissions between organic and conventional systems with 55 and 56 GJ ha⁻¹, and 3.30 t CO₂ equiv. ha⁻¹ and 3.57 t CO₂ equiv. ha⁻¹, respectively.Organic systems show three times more use of renewable energy than their conventional counterparts. However both systems depend heavily on fossil fuel inputs, machinery and N-fertilizers (synthetic or organic). Also, there is a high heterogeneity in management practices and input application within both systems, which is reflected in a large variation of their energy-related parameters. Given that avocado production is rapidly expanding in Mexico, a move toward organic production without systematically changing toward less fossil fuel dependent agricultural practices would not be sufficient to ensure a sustainable production.     

Environmental assessment of urban mobility: Combining life cycle assessment with land-use and transport interaction modelling—Application to Lyon (France)

In France, greenhouse gas (GHG) emissions from transport have grown steadily since 1950 and transport is now the main source of emissions. Despite technological improvements, urban sprawl increases the environmental stress due to car use. This study evaluates urban mobility through assessments of the transport system and travel habits, by applying life cycle assessment methods to the results of mobility simulations that were produced by a Land Use and Transport Interactions (LUTI) model. The environmental impacts of four life cycle phases of urban mobility in the Lyon area (exhausts, fuel processing, infrastructure and vehicle life cycle) were estimated through nine indicators (global warming potential, particulate matter emissions, photochemical oxidant emissions, terrestrial acidification, fossil resource depletion, metal depletion, non-renewable energy use, renewable energy use and land occupancy). GHG emissions were estimated to be 3.02 kg CO₂-eq inhabitant⁻¹ day⁻¹, strongly linked to car use, and indirect impacts represented 21% of GHG emissions, which is consistent with previous studies. Combining life cycle assessment (LCA) with a LUTI model allows changes in the vehicle mix and fuel sources combined with demographic shifts to be assessed, and provides environmental perspectives for transport policy makers and urban planners. It can also provide detailed analysis, by allowing levels of emissions that are generated by different categories of households to be differentiated, according to their revenue and location. Public policies can then focus more accurately on the emitters and be assessed from both an environmental and social point of view.     

Modelling the spatial distribution of linear landscape elements in Europe

Linear landscape elements, such as ditches, hedgerows, lines of trees and field margins, provide important habitats and ecosystem services and function as ecological infrastructure for species within agricultural landscapes. Spatial maps of the distribution of these elements are needed to better represent landscape structure within regional scale environmental assessments. We present wall-to-wall maps for green lines, ditches and grass margins for Europe, using spatial modelling of ground observations on linear features from the 2009 LUCAS (land use/cover area frame statistical survey) database. We compare different spatial interpolation methods, ranging from spatial autocorrelation-based methods to methods that explain the occurrence of elements based on biophysical and socio-economic information. Our results are 1 km² resolution maps of the occurrence of linear landscape elements for Europe. Independent validation of green lines based on aerial photographs showed the best results for interpolation based on regionally estimated regressions relating the occurrence of landscape elements to environmental and socio-economic location factors. The results confirm the importance of the underlying biophysical and socio-economic factors on the presence and abundance of linear landscape elements. However, the total explanatory strength of the considered factors is moderate and a considerable uncertainty in the exact distribution remains.     

Procrustes analysis as a tool for land management

Generalized Procrustes analysis (GPA) is a multivariate technique that involves transformations of data matrices to provide optimal comparability. We propose GPA to quantify the concordance among sets of variables that characterize natural, human and productive subsystems. When the land use fits in with the physical support of agricultural production, people's well-being should be evident in a high concordance between the land use and the social conditions. In a situation of instability each set of variables operates in diverse directions resulting in lower resilience and sustainability. Two GPA were performed, between physical support and land use data sets (concordance = 67.4%), and between land use and social conditions data sets (concordance = 65.3%). The interplay between the pair of concordance values constitutes a bi-dimensional index which serves as an ecological indicator. Based on bootstrap confidence interval, the 49 counties of the Pampa Ecoregion, Argentina, were classified in medium, high or low concordance. The lack of concordance is an indicator of imbalances which may contribute to guide environmental management.     

Wildfire spatial pattern analysis in the Zagros Mountains,Iran: A comparative study of decision tree based classifiers

Knowledge of wildfire behavior is of key importance for planning and allocating resources to fire suppression efforts. In this study, we analyzed the spatial pattern of wildfires with five decision tree based classifiers, including alternating decision tree (ADT), classification and regression tree (CART), functional tree (FT), logistic model tree (LMT), and Naïve Bayes tree (NBT). The classifiers were trained using historical fire locations in the Zagros Mountains (Iran) from the years 2007–2014 and a set of fifteen explanatory variables (i.e., slope degree, aspect, altitude, plan curvature, topographic position index (TPI), topographic roughness index (TRI), topographic wetness index (TWI), mean annual temperature and rainfall, wind effect, soil type, land use, and proximity to settlements, roads, and rivers) that were first optimized with a twostep process using multicollinearity analysis and the Gain Ratio variable selection method. The classifiers were then validated using the Kappa index and several statistical index-based evaluators (i.e., accuracy, sensitivity, specificity, precision, and F-measure). The global performance of the classifiers was measured using the ROC-AUC method. In this comparative study, the ADT classifier demonstrated the highest performance both in terms of goodness-of-fit with the training dataset (accuracy = 99.8%, AUC = 0.991) and the capability to predict future wildfires (accuracy = 75.7%, AUC = 0.903). This study contributes to the suite of research that evaluates data mining methods for the prediction of natural hazards.     

Modelling spatial patterns in harbour porpoise satellite telemetry data using maximum entropy

The distribution of harbour porpoises in EU waters is poorly understood, and modelled predictions of their distributions could inform the strategic spatial planning of future exploitation of the marine environment to avoid potential conflicts. We analysed satellite telemetry data from 39 harbour porpoises Phocoena phocoena in inner Danish waters using a modelling tool rooted in maximum entropy: Maxent. Maxent does not require absence data and has been shown to be effective for data characterised by small sample size, sampling bias and locational errors. For each season we used an iterative bootstrapping procedure to randomly select among the most precise records from each of the 39 tagged individuals, and ran Maxent on pooled records based on explanatory environmental variables hypothesised to serve as good proxies for harbour porpoise prey abundance. Among our environmental variables, distance to coast and bottom salinity had the most explanatory power, and their response shapes were relatively consistent across most seasons. The predictive power of the models (assessed by ROC‐AUC) ranged from 0.70 to 0.86 within seasons. The southern Kattegat, the Belt Seas, most western part of the Baltic Sea and the Sound were predicted to have relatively high probabilities of occurrence across seasons. In contrast, the central part of Kattegat and the Baltic Sea south and east of Limhamn and Darss Ridge consistently showed low probabilities of occurrence. Areas with the lowest probabilities of occurrence were generally characterised by high predictive uncertainty. Our methods have implications for the analyses of satellite tagged animals in terrestrial and marine environments. By coupling a bootstrapping procedure with Maxent we circumvented some of the statistical challenges presented by satellite telemetry data to generate spatial predictions within the inner Danish waters.     

Recent Lateral Expansion of Sphagnum Bogs Over Central Fen Areas of Boreal Aapa Mire Complexes

Ecosystems - We investigated recent changes in spatial patterning of fen and bog zones in five boreal aapa mire complexes (mixed peatlands with patterned fen and bog parts) in a multiproxy study....     

Assessing <Emphasis Type="Italic">Octopus vulgaris</Emphasis> distribution using presence-only model methods

The distribution of the common octopus (Octopus vulgaris) in the Mediterranean and Eastern Atlantic waters is evaluated using two presence-only analyses: The maximum entropy model (Maxent) and the ecological niche factor analysis (ENFA). Maxent predicts those geographical areas that satisfy the environmental or abiotic requirements of a species while ENFA explores the niche and habitat preferences of O. vulgaris. The analyses were implemented recovering the spatial information from 213 octopus presence data collected from surveys and bibliographical records. Together, these analyses provided reasonable estimates of the species distribution and the octopus habitat. Among the gathered set of explanatory environmental variables, sea bottom temperature, sea bottom salinity, surface dissolved oxygen and sea surface chlorophyll-α appear as the main variables involved in O. vulgaris distribution. These results were confronted with the available literature.     

Creating spatially continuous maps of past land cover from point estimates: A new statistical approach applied to pollen data

Reliable estimates of past land cover are critical for assessing potential effects of anthropogenic land-cover changes on past earth surface-climate feedbacks and landscape complexity. Fossil pollen records from lakes and bogs have provided important information on past natural and human-induced vegetation cover. However, those records provide only point estimates of past land cover, and not the spatially continuous maps at regional and sub-continental scales needed for climate modelling.We propose a set of statistical models that create spatially continuous maps of past land cover by combining two data sets: 1) pollen-based point estimates of past land cover (from the REVEALS model) and 2) spatially continuous estimates of past land cover, obtained by combining simulated potential vegetation (from LPJ-GUESS) with an anthropogenic land-cover change scenario (KK10). The proposed models rely on statistical methodology for compositional data and use Gaussian Markov Random Fields to model spatial dependencies in the data.Land-cover reconstructions are presented for three time windows in Europe: 0.05, 0.2, and 6 ka years before present (BP). The models are evaluated through cross-validation, deviance information criteria and by comparing the reconstruction of the 0.05 ka time window to the present-day land-cover data compiled by the European Forest Institute (EFI). For 0.05 ka, the proposed models provide reconstructions that are closer to the EFI data than either the REVEALS- or LPJ-GUESS/KK10-based estimates; thus the statistical combination of the two estimates improves the reconstruction. The reconstruction by the proposed models for 0.2 ka is also good. For 6 ka, however, the large differences between the REVEALS- and LPJ-GUESS/KK10-based estimates reduce the reliability of the proposed models. Possible reasons for the increased differences between REVEALS and LPJ-GUESS/KK10 for older time periods and further improvement of the proposed models are discussed.     

Nonlinear features and complexity patterns of vegetation dynamics in the transition zone of North China

Normalized Difference Vegetation Index (NDVI) has been commonly used to estimate terrestrial vegetation distribution and productivity. In this study, we adopted recurrence quantification analysis (RQA) to investigate the spatial patterns of determinism of the vegetation dynamics ecological-geographical transition zones in North China, especially the differences between transition zone and the surrounding areas. The results indicated that there were obvious regional variances in spatial patterns of RQA indices—determinism, laminarity, entropy, and averaged diagonal line length. Remarkable differences of the determinism of NDVI time series also existed between transition zones and the surrounding areas. Moreover, the correlation analysis between the RQA indices and climatic factors suggested that the determinism of the NDVI time series was nonlinearly affected by hydrothermal conditions. Influenced by vegetation patterns, determinism reached the maximum when the annual precipitation is about 400 mm, which is the lower bound of cultivation and forest distribution, and along the 400 mm isohyet is the area where transition zones locate.     

Landscape heterogeneity metrics as indicators of bird diversity: Determining the optimal spatial scales in different landscapes

Species distribution models are often used to study the biodiversity of ecosystems. The modelling process uses a number of parameters to predict others, such as the occurrence of determinate species, population size, habitat suitability or biodiversity. It is well known that the heterogeneity of landscapes can lead to changes in species’ abundance and biodiversity. However, landscape metrics depend on maps and spatial scales when it comes to undertaking a GIS analysis.We explored the goodness of fit of several models using the metrics of landscape heterogeneity and altitude as predictors of bird diversity in different landscapes and spatial scales. Two variables were used to describe biodiversity: bird richness and trophic level diversity, both of which were obtained from a breeding bird survey by means of point counts. The relationships between biodiversity and landscape metrics were compared using multiple linear regressions. All of the analyses were repeated for 14 different spatial scales and for cultivated, forest and grassland environments to determine the optimal spatial scale for each landscape typology.Our results revealed that the relationships between species’ richness and landscape heterogeneity using 1:10,000 land cover maps were strongest when working on a spatial scale up to a radius of 125–250 m around the sampled point (circa 4.9–19.6 ha). Furthermore, the correlation between measures of landscape heterogeneity and bird diversity was greater in grasslands than in cultivated or forested areas. The multi-spatial scale approach is useful for (a) assessing the accuracy of surrogates of bird diversity in different landscapes and (b) optimizing spatial model procedures for biodiversity mapping, mainly over extensive areas.     

Long-term phosphorus and nitrogen removal processes and preferential flow paths in Northern constructed peatlands

Four treatment peatlands were studied in Northern Finland in order to determine peat P, Al and Fe concentration distributions and to find removal parameters for nutrient modelling. The sites had been under loading for 10–16 years. About 20 peat samples for analyses of oxalate-extractable and total P, Al and Fe were collected from the depth 0–10 cm in each peatland. The peat P concentration ranged from 0.097 mg g^?1 to 14 mg g^?1 being 1.7 mg g^?1on average. P accumulated in preferential flow path areas. Although P concentrations were locally high, DSSP (the index of potential soil P release from peat to water) was very low in all studied peatlands, indicating that peat was not saturated by phosphorus. The results indicate that Al-based precipitation chemicals increase substantially P retention capacity of peatland and maintain a stabile P reduction in spite of varying P loads. The results also show long-term phosphorus accumulation in peatlands polishing municipal wastewater from activated sludge treatment. The regression analyses showed that k-value for N removal depends on N loading and hydraulic loading. The first-order area model together with regression analysis of the rate constant result in a good agreement between observed and calculated nitrogen concentration. The NH₄-N loading to the peatland should be below 0.10 mg m² d^?1 in order to achieve a high reduction of 70%.