38
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Random Forest Classification Method for Predicting Intertidal Wetland Migration Under Sea Level Rise

      , , , ,
      Frontiers in Environmental Science
      Frontiers Media SA

      Read this article at

      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Intertidal wetlands such as mangrove and saltmarsh are increasingly susceptible to areal losses related to sea level rise. This exposure is potentially offset by processes that might enable wetlands to accrete in situ or migrate landward under sea level rise, and planning policies that might open new opportunities for migration. We present and demonstrate a method to predict intertidal wetland distribution in the present-day landscape using random forest classification models, and use these models to predict the intertidal wetland distribution in future landscapes under specified sea level scenarios. The method is demonstrably robust in predicting present-day intertidal wetland distribution, with moderate correlation or better between predicted and mapped wetland distributions occurring in nearly all estuaries and strong correlation or better occurring in more than half of the estuaries. Given the accuracy in predicting present-day wetland distribution the method is assumed to be informative in predicting potential future wetland distribution when combined with best available models of future sea level. The classification method uses a variety of hydro-geomorphological surrogates that are derived from digital elevation models, Quaternary geology or soils mapping and land use mapping, which is then constrained by a representation of the future sea level inside estuaries. It is anticipated that the outputs from applying the method would inform assessments of intertidal wetland vulnerability to sea level rise and guide planning for potential wetland migration pathways.

          Related collections

          Most cited references53

          • Record: found
          • Abstract: not found
          • Article: not found

          Random Forests

            Bookmark
            • Record: found
            • Abstract: not found
            • Article: not found

            Building Predictive Models inRUsing thecaretPackage

              Bookmark
              • Record: found
              • Abstract: found
              • Article: found
              Is Open Access

              The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation

              Background To evaluate binary classifications and their confusion matrices, scientific researchers can employ several statistical rates, accordingly to the goal of the experiment they are investigating. Despite being a crucial issue in machine learning, no widespread consensus has been reached on a unified elective chosen measure yet. Accuracy and F1 score computed on confusion matrices have been (and still are) among the most popular adopted metrics in binary classification tasks. However, these statistical measures can dangerously show overoptimistic inflated results, especially on imbalanced datasets. Results The Matthews correlation coefficient (MCC), instead, is a more reliable statistical rate which produces a high score only if the prediction obtained good results in all of the four confusion matrix categories (true positives, false negatives, true negatives, and false positives), proportionally both to the size of positive elements and the size of negative elements in the dataset. Conclusions In this article, we show how MCC produces a more informative and truthful score in evaluating binary classifications than accuracy and F1 score, by first explaining the mathematical properties, and then the asset of MCC in six synthetic use cases and in a real genomics scenario. We believe that the Matthews correlation coefficient should be preferred to accuracy and F1 score in evaluating binary classification tasks by all scientific communities.
                Bookmark

                Author and article information

                Journal
                Frontiers in Environmental Science
                Front. Environ. Sci.
                Frontiers Media SA
                2296-665X
                July 13 2022
                July 13 2022
                : 10
                Article
                10.3389/fenvs.2022.749950
                f61345f8-4268-4c08-8c1c-1ceda8061e49
                © 2022

                Free to read

                https://creativecommons.org/licenses/by/4.0/

                History

                Comments

                Comment on this article