12
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: not found

      Predicting nitrogen loading with land-cover composition: how can watershed size affect model performance?

      1 ,
      Environmental management
      Springer Nature

      Read this article at

      ScienceOpenPublisherPubMed
      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

          Watershed-wide land-cover proportions can be used to predict the in-stream non-point source pollutant loadings through regression modeling. However, the model performance can vary greatly across different study sites and among various watersheds. Existing literature has shown that this type of regression modeling tends to perform better for large watersheds than for small ones, and that such a performance variation has been largely linked with different interwatershed landscape heterogeneity levels. The purpose of this study is to further examine the previously mentioned empirical observation based on a set of watersheds in the northern part of Georgia (USA) to explore the underlying causes of the variation in model performance. Through the combined use of the neutral landscape modeling approach and a spatially explicit nutrient loading model, we tested whether the regression model performance variation over the watershed groups ranging in size is due to the different watershed landscape heterogeneity levels. We adopted three neutral landscape modeling criteria that were tied with different similarity levels in watershed landscape properties and used the nutrient loading model to estimate the nitrogen loads for these neutral watersheds. Then we compared the regression model performance for the real and neutral landscape scenarios, respectively. We found that watershed size can affect the regression model performance both directly and indirectly. Along with the indirect effect through interwatershed heterogeneity, watershed size can directly affect the model performance over the watersheds varying in size. We also found that the regression model performance can be more significantly affected by other physiographic properties shaping nitrogen delivery effectiveness than the watershed land-cover heterogeneity. This study contrasts with many existing studies because it goes beyond hypothesis formulation based on empirical observations and into hypothesis testing to explore the fundamental mechanism.

          Related collections

          Author and article information

          Journal
          Environ Manage
          Environmental management
          Springer Nature
          1432-1009
          0364-152X
          Jan 2013
          : 51
          : 1
          Affiliations
          [1 ] Department of Biology, University of Florida, Gainesville, FL 32611, USA. tz@ufl.edu
          Article
          10.1007/s00267-012-9897-3
          22773114
          674b0741-2aff-49da-99a6-eb6a39c99cdb
          History

          Comments

          Comment on this article