A Tool for Classification and Regression Using Random Forest Methodology: Applications to Landslide Susceptibility Mapping and Soil Thickness Modeling

Classification and regression problems are a central issue in geosciences. In this paper, we present Classification and Regression Treebagger (ClaReT), a tool for classification and regression based on the random forest (RF) technique. ClaReT is developed in Matlab and has a simple graphic user interface (GUI) that simplifies the model implementation process, allows the standardization of the method, and makes the classification and regression process reproducible. This tool performs automatically the feature selection based on a quantitative criterion and allows testing a large number of explanatory variables. First, it ranks and displays the parameter importance; then, it selects the optimal configuration of explanatory variables; finally, it performs the classification or regression for an entire dataset. It can also provide an evaluation of the results in terms of misclassification error or root mean squared error. We tested the applicability of ClaReT in two case studies. In the first one, we used ClaReT in classification mode to identify the better subset of landslide conditioning variables (LCVs) and to obtain a landslide susceptibility map (LSM) of the Arno river basin (Italy). In the second case study, we used ClaReT in regression mode to produce a soil thickness map of the Terzona catchment, a small sub-basin of the Arno river basin. In both cases, we performed a validation of the results and a comparison with other state-of-the-art techniques. We found that ClaReT produced better results, with a more straightforward and easy application and could be used as a valuable tool to assess the importance of the variables involved in the modeling.     

Seasonal Differences in the Levels of Gaseous Air Pollutants in the Vicinity of a Waste Dump

Hydrogen sulphide, ammonia, nitrogen dioxide, mercaptans and sulphur dioxide (H₂S, NH₃, NO₂, R-SH, SO₂) concentrations were measured at the location in the vicinity of the waste dump to determine the air pollution level of these pollutants prior to the operation of the Mobile Thermal Treatment Plant. Samples were collected over one year period. Seasonal differences, and the influence of meteorological parameters (temperature, relative humidity, pressure and wind direction) on the air pollution levels were studied. Results show relatively low concentrations of H₂S, NO₂, R-SH and SO₂, while NH₃ levels were higher compared to the guideline values. Good weather conditions (high air pressure and low relative humidity) are connected to long range transport of NO₂, while higher temperatures result in elevated NH₃ and R-SH concentrations. Because of the predominant northeast wind direction (the same as the waste dump direction), the contribution of air pollution from the direction of the waste dump at the measuring site is significant, but that does not necessarily mean that the pollutants originated from that source.     

Physico-Chemical Quality of Drinking Water in Rohtas District of Bihar

Water Quality Survey of Rohtas district of Bihar was conducted. Samples were collected from differentsources and analysed. 209 samples were collectedfrom 196 villages. Results of water quality surveyidentified the problem areas in respect of high iron,manganese, fluoride, nitrate and brackishness of water in the district.     

Landscape Correlates of Breeding Bird Richness Across the United States Mid-Atlantic Region

Using a new set of landscape indicator data generated by the U.S.EPA, and a comprehensive breeding bird database from the National Breeding Bird Survey, we evaluated associations between breeding bird richness and landscape characteristics across the entire mid-Atlantic region of the United States. We evaluated how these relationships varied among different groupings (guilds) of birds based on functional, structural, and compositional aspects of individual species demographics. Forest edge was by far the most important landscape attribute affecting the richness of the lumped specialist and generalist guilds; specialist species richness was negatively associated with forest edge and generalist richness was positively associated with forest edge. Landscape variables (indicators) explained a greater proportion of specialist species richness than the generalist guild (46% and 31%, respectively). The lower value in generalists may reflect finer-scale distributions of open habitat that go undetected by the Landsat satellite, open habitats created by roads (the areas from which breeding bird data are obtained), and the lumping of a wide variety of species into the generalist category. A further breakdown of species into 16 guilds showed considerable variation in the response of breeding birds to landscape conditions; forest obligate species had the strongest association with landscape indicators measured in this study (55% of the total variation explained) and forest generalists and open ground nesters the lowest (17% of the total variation explained). The variable response of guild species richness to landscape pattern suggests that one must consider species' demographics when assessing the consequences of landscape change on breeding birds.     

Accumulation of Heavy Metals in Epiphytic Lichens Near a Municipal Solid Waste Incinerator (Central Italy)

The epiphytic lichen Parmelia caperata was usedas biomonitor in the area of a municipal solid wasteincinerator (Poggibonsi, central Italy) to investigatethe levels and the spatial distribution of the heavymetals Al, Cd, Cr, Cu, Fe, Hg, Pb and Zn. Levels ofAl, Cu and Hg were similar to those in unpollutedareas, whereas high values were found for Cr, Zn andespecially Cd. The distribution pattern of the lastthree metals and the exponential relationship of theirconcentrations with distance from the incinerator,showed that the disposal plant is a local source ofatmospheric pollution due to Cd, Cr and Zn. For thesemetals, long-term hazard should be seriously taken into account.     

Application of Defined Co-Immobilized Microbial Consortium as a Ready-to-Use Seed Inoculum in Bod Analysis

A number of microorganisms were isolated from sewage. Biochemical Oxygen Demand (BOD-5 day) analysis was carried out by individual pure cultures. The cultures giving higher or equal BOD values as compared to reference GGA solution were selected for the formulation of a defined mixed microbial consortium. This microbial consortium was co-immobilized on calcium-alginate beads. Four synthetic and six industrial samples were tested for BOD by using immobilized beads as well as sewage as source of seeding materials. BOD values obtained with beads for all the synthetic as well as industrial samples were fairly comparable with those obtained with sewage. Reusability of prepared microbial beads was also checked with different synthetic and industrial samples and was compared with reference GGA solution. The same microbial beads can be reused three times for different BOD-5 day estimations. It is recommended that immobilized microbial beads can be used as a ready-to-use seeding material for BOD analysis.     

Inputs of chemicals from recreational activities into the Norwegian coastal zone

Wastewater treatment works effluent is often considered to be one of the most important point sources of a wide range of anthropogenic contaminants to aquatic systems, however, this paper discusses other potential sources. With the aim of establishing the important sources of UV filters, insect repellent and biocides to the aquatic environment, samples were collected from sites with direct (bathing areas and marinas) and indirect (sites receiving wastewater effluent) human influences. Sunscreens containing UV filters are used in large volumes during the summer months and often applied shortly before a person enters the water for swimming activities. The results presented here demonstrate that washing directly from the skin is an important point source of 4 UV filters to the Oslofjord. The insect repellent, diethyl toluamide, was also measured and it was concluded that washing from the skin was not such an important point source into the fjord. Concentrations of the biocide Irgarol 1051 were also measured and were elevated in the small boat marina and surrounding enclosed area. This work demonstrates that man recreational water-based activities are a diffuse source of some contaminants into coastal and fjord environments and this study provides an initial assessment of the levels being released.     

A comparison of speciated atmospheric mercury at an urban center and an upwind rural location

Gaseous elemental mercury (GEM), particulate mercury (PHg) and reactive gaseous mercury (RGM) were measured every other hour at a rural location in south central Wisconsin (Devil's Lake State Park, WI, USA) between April 2003 and March 2004, and at a predominantly downwind urban site in southeastern Wisconsin (Milwaukee, WI, USA) between June 2004 and May 2005. Annual averages of GEM, PHg, and RGM at the urban site were statistically higher than those measured at the rural site. Pollution roses of GEM and reactive mercury (RM; sum of PHg and RGM) at the rural and urban sites revealed the influences of point source emissions in surrounding counties that were consistent with the US EPA 1999 National Emission Inventory and the 2003-2005 US EPA Toxics Release Inventory. Source-receptor relationships at both sites were studied by quantifying the impacts of point sources on mercury concentrations. Time series of GEM, PHg, and RGM concentrations were sorted into two categories; time periods dominated by impacts from point sources, and time periods dominated by mercury from non-point sources. The analysis revealed average point source contributions to GEM, PHg, and RGM concentration measurements to be significant over the year long studies. At the rural site, contributions to annual average concentrations were: GEM (2%; 0.04 ng m(-3)); and, RM (48%; 5.7 pg m(-3)). At the urban site, contributions to annual average concentrations were: GEM (33%; 0.81 ng m(-3)); and, RM (64%; 13.8 pg m(-3)).     

Identification and quantification of microcystins from a Nostoc muscorum bloom occurring in Oukaïmeden River (High-Atlas mountains of Marrakech, Morocco)

Atmospheric BTEX [benzene, toluene, ethylbenzene, o-xylene and (m + p)-xylene] concentrations have been determined in the Naples metropolitan area (NMA) and in two suburban areas located on the north within about 25 km, during 2006. The pollutants were collected by passive samplers (24-h samplings), and analysed by GC-MS. In all the areas analysed the average atmospheric benzene concentrations were higher than the limit value fixed by the European Union for 2010 and in NMA the average concentration (9.8 microg m(-3)) also exceeded the limit fixed for 2006. High linear correlation coefficients between the average daily concentrations of the different BTEX are indicative of a single major source, most likely the vehicular traffic. The temporal and spatial distribution of BTEX relative concentrations suggest that massive emissions in NMA negatively affect the quality of the air in northern suburban areas, prevalently during the hottest months of the year, probably due to transport by local seasonal winds. The ratios between BTEX daily concentrations showed a clear dependence on the intensity of solar actinic flow, indicating a major role of photochemical processes in the air cleaning from these volatile organic pollutants.     

Predicting the biological condition of streams: use of geospatial indicators of natural and anthropogenic characteristics of watersheds

We developed and evaluated empirical models to predict biological condition of wadeable streams in a large portion of the eastern USA, with the ultimate goal of prediction for unsampled basins. Previous work had classified (i.e., altered vs. unaltered) the biological condition of 920 streams based on a biological assessment of macroinvertebrate assemblages. Predictor variables were limited to widely available geospatial data, which included land cover, topography, climate, soils, societal infrastructure, and potential hydrologic modification. We compared the accuracy of predictions of biological condition class based on models with continuous and binary responses. We also evaluated the relative importance of specific groups and individual predictor variables, as well as the relationships between the most important predictors and biological condition. Prediction accuracy and the relative importance of predictor variables were different for two subregions for which models were created. Predictive accuracy in the highlands region improved by including predictors that represented both natural and human activities. Riparian land cover and road-stream intersections were the most important predictors. In contrast, predictive accuracy in the lowlands region was best for models limited to predictors representing natural factors, including basin topography and soil properties. Partial dependence plots revealed complex and nonlinear relationships between specific predictors and the probability of biological alteration. We demonstrate a potential application of the model by predicting biological condition in 552 unsampled basins across an ecoregion in southeastern Wisconsin (USA). Estimates of the likelihood of biological condition of unsampled streams could be a valuable tool for screening large numbers of basins to focus targeted monitoring of potentially unaltered or altered stream segments.