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.     

A moving target—incorporating knowledge of the spatial ecology of fish into the assessment and management of freshwater fish populations

Freshwater fish move vertically and horizontally through the aquatic landscape for a variety of reasons, such as to find and exploit patchy resources or to locate essential habitats (e.g., for spawning). Inherent challenges exist with the assessment of fish populations because they are moving targets. We submit that quantifying and describing the spatial ecology of fish and their habitat is an important component of freshwater fishery assessment and management. With a growing number of tools available for studying the spatial ecology of fishes (e.g., telemetry, population genetics, hydroacoustics, otolith microchemistry, stable isotope analysis), new knowledge can now be generated and incorporated into biological assessment and fishery management. For example, knowing when, where, and how to deploy assessment gears is essential to inform, refine, or calibrate assessment protocols. Such information is also useful for quantifying or avoiding bycatch of imperiled species. Knowledge of habitat connectivity and usage can identify critically important migration corridors and habitats and can be used to improve our understanding of variables that influence spatial structuring of fish populations. Similarly, demographic processes are partly driven by the behavior of fish and mediated by environmental drivers. Information on these processes is critical to the development and application of realistic population dynamics models. Collectively, biological assessment, when informed by knowledge of spatial ecology, can provide managers with the ability to understand how and when fish and their habitats may be exposed to different threats. Naturally, this knowledge helps to better evaluate or develop strategies to protect the long-term viability of fishery production. Failure to understand the spatial ecology of fishes and to incorporate spatiotemporal data can bias population assessments and forecasts and potentially lead to ineffective or counterproductive management actions.     

Characterization of As and trace metals embedded in PM10 particles in Puebla City, México

Forty-eight air-filter samples (PM₁₀) were analysed to identify the concentration level of partially leached metals (PLMs; As, Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb and V) from Puebla City, México. Samples were collected during 2008 from four monitoring sites: (1) Tecnológico (TEC), (2) Ninfas (NIN), (3) Hermanos Serdán (HS) and (4) Agua Santa (AS). The results indicate that in TEC, As (avg. 424 ng m^?3), V (avg. 19.2 ng m^?3), Fe (avg. 1,202 ng m^?3), Cu (avg. 86.6 ng m^?3), Cr (41.9 ng m^?3) and Ni (18.6 ng m^?3) are on the higher side than other populated regions around the world. The enrichment of PLMs is due to the industrial complexes generating huge dust particles involving various operations. The results are supported by the correlation of metals (Mn, Cd and Co) with Fe indicating its anthropogenic origin and likewise, As with Cd, Co, Fe, Mn, Pb and V. The separate cluster of As, Fe and Mn clearly signifies that it is due to continuous eruption of fumaroles from the active volcano Popocatépetl in the region.