Using landscape typologies to model socioecological systems: Application to agriculture of the United States Gulf Coast

The complexity of socioecological systems (SES) has posed a persistent challenge to the development of methods for diagnostic and prognostic analyses of global change. We developed a high dimensional statistical framework where cluster analysis was used to characterize regional landscape typologies...

Full description

Saved in:
Bibliographic Details
Published in:Environmental modelling & software : with environment data news 2016-05, Vol.79 (C), p.85-95
Main Authors: Surendran Nair, Sujithkumar, Preston, Benjamin L., King, Anthony W., Mei, Rui
Format: Article
Language:English
Subjects:
Adaptive capacity
Agroecosystem
Clusters
Coastal environments
Corn
GEOSCIENCES
Gulfs
High-dimensional statistics
Landscape typology
Landscapes
Mathematical models
Modelling
Regional
Socioecological systems
Socioecological typology
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The complexity of socioecological systems (SES) has posed a persistent challenge to the development of methods for diagnostic and prognostic analyses of global change. We developed a high dimensional statistical framework where cluster analysis was used to characterize regional landscape typologies and those typologies are linked to the outcome of interest through regression modeling. For demonstration, we applied the framework to agroecosystem of the United States Gulf Coast to evaluate the determinants of spatial variability in crop yield. Regional biophysical typologies (BPT; integrated climate, soil, and topography clusters) and socioecological typologies (SET; BPT combined with socioeconomic clusters) were developed. The SET corn model (R2 = 0.89) outperformed the BPT corn model (R2 = 0.72) and a county fixed-effect model (R2 = 0.53), which reflects the socioeconomic influence over agricultural productivity. The SET model also showed similar predictive skill for soybean and cotton yield. Therefore impact analysis for agroecosystems can lead to incorrect conclusions if biophysical factors are not examined jointly with socioeconomic factors. •The socioecological typology method was used for identifying and attributing regional crop yield variability.•The models demonstrated high predictive skill.•Useful for integrating multi-scaled characteristic of socioecological systems.
ISSN:1364-8152
1873-6726
DOI:10.1016/j.envsoft.2016.01.008