Tools for and barriers to terrestrial ecosystem stewardship

We compare ecosystem models applicable to the study of land management. We contrast models across the following metrics: 1) structural transparency, 2) model complexity, 3) scale of application, 4) strength of vegetation description, 5) strength of biogeochemistry description, 6) strength of hydrologic description, 7) representation of land management, 8) availability of validation studies, and 9) integration with socio-economic land-use drivers. Our analysis emphasizes the following questions: 1) Do ecosystem models accurately characterize the dynamics of managed landscapes? Is model sensitivity well characterized? Are sufficient observations available for model validation? 2) Are ecosystem modeling tools used to test ecologically or economically motivated land management scenarios? 3) How accessible are ecosystem modeling tools? Is robust application of an ecosystem model limited to the expert development team? We find the following broad trends across land management models. Though many ecosystem models are freely available and have a user-friendly interface, ecosystem models are generally applied by a small community of model developers. Commonly applied land management models cover a range of complexity in terms of vegetation, hydrologic, and biogeochemical processes that are explicitly modeled. The accuracy of modeled vegetation growth is typically robust across models. However, model validation against biogeochemical cycle and hydrologic data is more limited in scope, often due to lack of observations, and reveals challenges for model accuracy. With some exceptions, ecosystem models have predominantly been used to test management scenarios which conform to conventional land management practices. Ecosystem model application provides an opportunity to question conventional land management subsidies and would benefit from collaboration with socio-economic disciplines.