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.     

Treatment of fluoride-contaminated water. A review

Environmental Chemistry Letters - Delivering the right amount of fluoride to drinking water protects the teeth from decay and reduces the risk of cavities. Nonetheless, fluorosis has been...     

IRRIGATOR INVOLVEMENT IN THE IMPLEMENTATION OF AGRICULTURAL NONPOINT SOURCE POLLUTION CONTROL PROGRAMS1

ABSTRACT: Recently, Congress designated irrigated agriculture under the “nonpoint source” category, covered by Section 208 of P.L. 92-500 and involves the use of “best management practices.” Generally, the most appropriate solutions for pollution abatement from irrigated agriculture involve the delivery and use of water rather than the treatment of irrigation return flows. 1. Technological alternatives should be utilized that are sensitive to local conditions and acceptable to the farmers. 2. Informational and educational programs to assist farm operators individually and collectively must be instituted prior to the start of the project; imaginatively conceived, and continuously modified and upgraded if motivation for change is to be encouraged. 3. Technical assistance personnel should be given short courses in skills needed for working effectively with irrigators. 4. Communication techniques used for working with farmers as individuals and groups should be designed into the implementation program and evaluated. 5. Credibility and trustworthiness of Federal and state agencies in the eyes of the irrigators provide the important final ingredient in effectively implementing change and reducing nonpoint source pollution from irrigated agriculture.     

Headwater streams: neglected ecosystems in the EU Water Framework Directive. Implications for nitrogen pollution control

The European Union Water Framework Directive (WFD) aims to achieve the “good status” of waters by 2015, through monitoring and control of human impacts on “bodies of surface water” (BSWs), discrete elements for quality diagnosis and management. Headwater streams, however, are frequently neglected as they are not usually recognised as BSW. This poses limitations for the management of river catchments, because anthropogenic impacts on headwaters can constrain the quality of downstream rivers. To illustrate this problem, we compared nitrate levels and land use pressures in a small agricultural catchment with those recorded in the catchment in which it is embedded (Ega), and in the Ebro River Basin (NE Spain) comprising both. Agriculture greatly influenced water nitrate concentration, regardless of the size of the catchments: R² = 0.91 for headwater catchments (0.1–7.3 km²), and R² = 0.82 for Ebro tributary catchments (223–3113 km²). Moreover, nitrate concentration in the outlet of a non-BSW small river catchment was similar to that of the greater downstream BSW rivers. These results are of interest since, despite representing 76% of the length of the Ega catchment hydrographical network, only 3.1% of the length of the headwater streams has been identified as BSWs. Human activities affecting headwater streams should therefore be considered if the 2015 objective of the WFD is to be achieved.     

A model for estimating the potential biomagnification of chemicals in a generic food web: Preliminary development

GOAL, SCOPE AND BACKGROUND: Bioaccumulation and biomagnification of organic pollutants have been increasingly assessed and modeled during the last years. Due to the complexity of these processes and the large variability of food webs, setting generic assessments for these parameters is really difficult. Equilibrium models, based on a compound's lipophylicity, are the main tool in regulatory proposals, such as for identifying Persistent, Bioaccumulative and Toxic Substances (PBTs), although a refinement has been claimed by the scientific community. Toxicokinetic studies offer an alternative for these estimations, where biomagnification is modeled as a succession of bioaccumulation processes, each one regulated by toxicokinetic parameters. METHODS: A review of kinetic models covering species belonging to different trophic levels and with different ecological behavior has been conducted. The results were employed for setting a conceptual model for estimating the biomagnification potential in a generic food web, which was mathematically implemented through system dynamic models developed under data sheet software. Crystal Ball was then employed for allowing Monte Carlo based probabilistic calculations. Bioaccumulation laboratory assays have been performed to estimate toxicokinetic parameters in mussels (Mytilus edulis) with two PAHs (chrysene and benzo[a]pyrene). The contamination was delivered via food. The exposure period lasted more than one month followed then by a depuration phase. The contaminant content was determined on an individual basis on five replicates. RESULTS AND DISCUSSION:. The reviewed information suggested the development of a tiered conceptual biomagnification model, starting with a simplified food chain which can be refined to more realistic and complex models in successive levels. CONCLUSIONS: The mathematical implementation of the conceptual model offers tools for estimating the potential for bioaccumulation and biomagnification of chemicals under very different conditions. The versatility of the model can be used for both comparative estimations and for validating the model. RECOMMENDATIONS AND PERSPECTIVES: Since bioaccumulation and biomagnification processes are crucial elements for a proper risk assessment of chemicals, their estimation by mathematical models has been widely tested. However, inregulatory assessments, too simplistic models are still being used quite often. The biomagnification model presented in this study should be amore accurate alternative to these models. In comparison to other previously published biomagnification models, the present one covers the time variation of bioaccumulation using just a few toxicokinetic parameters.     

Reproduction and biochemical responses in Enchytraeus albidus (Oligochaeta) to zinc or cadmium exposures

To better understand chemical modes of action, emphasis has been given to stress responses at lower levels of biological organization. Cholinesterases and antioxidant defenses are among the most used biomarkers due to their crucial role in the neurocholinergic transmission and in cell homeostasis preventing DNA damage, enzymatic inactivation and lipid peroxidation. The main goal of this study was to investigate the effects of zinc and cadmium on survival and reproduction of E. albidus and to assess metals oxidative stress potential and neurotoxic effects at concentrations that affected reproduction. Both metals affected the enchytraeids’ survival and reproduction and induced significant changes in the antioxidant defenses as well as increased lipid peroxidation, indicating oxidative damage. This study demonstrates that determining effects at different levels of biological organization can give better information on the physiological responses of enchytraeids in metal contamination events and further unravel the mechanistic processes dealing with metal stress.     

Temporal patterns of atmospheric mercury species in northern Mississippi during 2011–2012: Influence of sudden population swings

Gaseous elemental mercury (GEM), gaseous oxidized mercury (GOM) and particulate bound mercury (PBM) were measured on the University of Mississippi campus from July 2011 to June 2012. It is believed to be the first time that concentrations of atmospheric mercury species have been documented in northern Mississippi, and at a location with relatively large and sudden swings in population. The mean concentration (±1SD) of GEM was 1.54 ± 0.32 ng m⁻³; levels were lower and generally more stable during the winter (1.48 ± 0.22) and spring (1.46 ± 0.27) compared with the summer (1.56 ± 0.32) and fall (1.63 ± 0.42). Mean concentrations for GOM and PBM were 3.87 pg m⁻³ and 4.58 pg m⁻³, respectively; levels tended to be highest in the afternoon and lowest in the early morning hours. During the fall and spring academic semesters concentrations and variability of GOM and PBM both increased, possibly from vehicle exhaust. There were moderate negative correlations with wind speed (all species) and humidity (GOM and PBM). Backward air mass trajectory modeling for the ten highest peaks for each mercury species revealed that the majority of these events occurred from air masses that passed through the northern continental US region. Overall, this study illustrates the complexity of temporal fluctuations of airborne mercury species, even in a small town environment.     

Combined effects of nitrogen addition and organic matter manipulation on soil respiration in a Chinese pine forest

The response of soil respiration (Rs) to nitrogen (N) addition is one of the uncertainties in modelling ecosystem carbon (C). We reported on a long-term nitrogen (N) addition experiment using urea (CO(NH₂)₂) fertilizer in which Rs was continuously measured after N addition during the growing season in a Chinese pine forest. Four levels of N addition, i.e. no added N (N0: 0 g N m⁻² year⁻¹), low-N (N1: 5 g N m⁻² year⁻¹), medium-N (N2: 10 g N m⁻² year⁻¹), and high-N (N3: 15 g N m⁻² year⁻¹), and three organic matter treatments, i.e. both aboveground litter and belowground root removal (LRE), only aboveground litter removal (LE), and intact soil (CK), were examined. The Rs was measured continuously for 3 days following each N addition application and was measured approximately 3–5 times during the rest of each month from July to October 2012. N addition inhibited microbial heterotrophic respiration by suppressing soil microbial biomass, but stimulated root respiration and CO₂ release from litter decomposition by increasing either root biomass or microbial biomass. When litter and/or root were removed, the “priming” effect of N addition on the Rs disappeared more quickly than intact soil. This is likely to provide a point of view for why Rs varies so much in response to exogenous N and also has implications for future determination of sampling interval of Rs measurement.

     

Mitigation of hydrogen fluoride aerosols by dry powders

Fluid bed tests were used to determine overall reaction rates for eight different dry powders of low hazard and toxicity before and after reaction with HF. In flow chamber tests, a representative oxide, hydroxide and carbonate were used to measure the efficiency of mitigation of an HF aerosol cloud. The results show that dry powders provide a valid alternative to water spray mitigation. Powder efficiencies are higher than water efficiencies at constant weight ratio.

At present, no attempt has been made to provide any technical designs or layouts for powder mitigation systems. However, the technology for smaller systems is available through manufacturers of dry powder chemical fire extinguisher equipment. When these powders are kept dry and under nitrogen atmosphere, a shelf life of several years can be expected.     

Cover estimation versus density counting in species-rich pasture under different grazing intensities

Two methods for monitoring of grassland vegetation were compared: visual estimation of plant cover (C) and plant densities counting (D). C and D were performed in monthly intervals for three vegetation growing seasons after imposing different grazing regimes on abandoned grassland in 1998. Species scores obtained from paired redundancy analyses (RDA) of C and D data were compared and Spearman's rank correlations were used to show if the two methods give comparable results. Results of C and D were highly correlated in the first two growing seasons only. In the third season, correlation was substantially lower as the sward structure was more heterogeneous due to creation of differently defoliated patches especially under extensive grazing. Presence of the same plant species with different habit in frequently and in infrequently grazed patches, reduced significance of Spearman's rank correlations. Cover estimation can fully substitute plant density counting in grassland with lower proportion of frequently and infrequently grazed patches only, but caution should be used when comparing different management regimes in long term analyses.