Project Funding: 2015 - 2018

NRA: 2013 NASA: Land Cover / Land Use Change   

Funded by NASA

Abstract:
How broad-scale factors impinge on local decision making and translate into land use change is not well understood. This is particularly true of the vast rangelands of the Altai Mountains in central Asia where grazing was and remains the dominant form of agriculture and land use. Critical questions remain about what communities, donors, and policymakers can do to promote desirable co-management outcomes in grazing systems of this ecologically similar region but politically complex region which remains in upheaval following the collapse of the former Soviet Union and the heavy subsidies it once provided to herding societies. Our study will take a nested approach that first contrasts long-term, broad-scale LCLUC based spectral dynamics derived from Landsat imagery over the last 30 years for the same high montane grasslands occupied by Kazakh peoples herding livestock across four countries with strikingly different political systems Mongolia, Russia, Kazakhstan and China (regional modeling). We will then integrate remotely sensed data on recent LCLUC with semi-structured interviews of local herders at the local level among grazing areas in a transboundary region shared between Mongolia and Russia along the Sailyugem Range (local modeling). The local comparison will be used to operationalize Ostrom’s diagnostic framework for social-ecological systems to elucidate mechanisms by which household- and community-scale factors influence the exploitation status of rangelands and thereby identify the mechanisms behind how social-ecological grazing systems influence LCLUC in the region. Importantly, we have validated on Mongolian grasslands during 2013 that MODIS-derived EVI accurately reflects on-the-ground conditions of great relevance to rangeland management and herders (particularly % bare ground and a qualitative index of rangeland quality derived from herders traditional ecological knowledge). Extending this validation work, we will explore the use of small unmanned aerial vehicles for more efficient delivery of high quality information for validating RS data on rangeland quality in mountainous environments and associated grazing systems. This extensive database describing in detail on-the-ground rangeland conditions at 100 1 km2 plots in the Mongolian Altai will enable us to develop a new approach to convert information extracted from satellite images into knowledge on change patterns, distribution and intensity of rangeland conditions for mountainous regions. More specifically, we will develop a novel multi-sensor fusion methodology based on expected spatio-spectral behavior of ground features. Study outcomes will include better monitoring and understanding of pastureland change, important for guiding land-use policy and management in the region, and a new processing protocol for integrating multi-sensor information for grassland mapping. Moreover, we will integrate research with education and outreach in substantive ways by involving > 100 undergraduate and graduate students at SUNY ESF on the project and contributing the images to Zooinverse where citizen scientists will be engaged to help interpret imagery (and where we can contrast the outcomes of crowdsourced versus classroom-based estimates of rangeland indicators derived from high-resolution imagery). In terms of professional outreach we will distribute data products, tools, and concepts developed to users through the NYView consortium-based web server and target regional experts and policy makers concerned with rangeland management issues in the vast Altai mountain region via a workshop we will host in Mongolia during the final year of the project. To assist us with project activities we will be partnering with three WWF offices from the local region.