Empirical social-ecological system analysis: from theoretical framework to latent variable structural equation model.

The social-ecological system (SES) approach to natural resource management holds enormous promise towards achieving sustainability. Despite this promise, social-ecological interactions are complex and elusive; they require simplification to guide effective application of the SES approach. The complex, adaptive and place-specific nature of human-environment interactions impedes determination of state and trends in SES parameters of interest to managers and policy makers. Based on a rigorously developed systemic theoretical model, this paper integrates field observations, interviews, surveys, and latent variable modeling to illustrate the development of simplified and easily interpretable indicators of the state of, and trends in, relevant SES processes. Social-agricultural interactions in the Logone floodplain, in the Lake Chad basin, served as case study. This approach is found to generate simplified determinants of the state of SESs, easily communicable across the array of stakeholders common in human-environment interactions. The approach proves to be useful for monitoring SESs, guiding interventions, and assessing the effectiveness of interventions. It incorporates real time responses to biophysical change in understanding coarse scale processes within which finer scales are embedded. This paper emphasizes the importance of merging quantitative and qualitative methods for effective monitoring and assessment of SESs.