Friedl, Mark: Boston University (Project Lead)
Project Funding:
2011 - 2014
NRA: 2010 NASA: NPP Science Team for Climate Data Records
Funded by NASA
Abstract:
This proposal describes research focused on method and data set development in support of a land cover climate data record from NPP VIIRS. Land cover will be generated from NPP via the VIIRS surface type environmental data record (EDR). This EDR will include global maps of land cover at 1-km spatial resolution and quarterly time steps. To generate this data, the NPP surface type EDR will use an approach that is similar to the MODIS land cover type classification algorithm, but which deviates from the MODIS approach in several key aspects. Hence, results from these two products are likely to be quite different. Further, neither the MODIS land cover product nor the VIIRS surface type EDR are designed to accurately track global land cover changes and there is no strategy for continuity of moderate resolution global land cover data sets between VIIRS and MODIS. The activities described in this proposal are designed to address these issues. Specifically we propose a framework and associated set of analyses designed to (1) distinguish stable land cover regions from regions where change may be occurring, (2) identify and label pixels that exhibit spurious change due to poor classification performance, and (3) identify pixels where real change is occurring and to categorize the nature of changes at these pixels according to a framework appropriate to moderate resolution remote sensing observations from instruments such as VIIRS and MODIS. Key elements of the proposed activities include land cover change model development at a set of well- characterized locations with known land cover properties and changes that span a range of mechanisms, rates, and intensities of change. As part of this effort we also propose to develop prototype methods for assessment and validation of regional and global land cover change data sets. The proposed activities will leverage and extend experience, data sets, and algorithms developed from a decade of algorithm development in support of the MODIS land cover product.
Publications:
Glanz, H., Carvalho, L., Sulla-Menashe, D., Friedl, M. A. 2014. A parametric model for classifying land cover and evaluating training data based on multi-temporal remote sensing data. ISPRS Journal of Photogrammetry and Remote Sensing. 97, 219-228. DOI: 10.1016/j.isprsjprs.2014.09.004
Huang, X., Schneider, A., Friedl, M. A. 2016. Mapping sub-pixel urban expansion in China using MODIS and DMSP/OLS nighttime lights. Remote Sensing of Environment. 175, 92-108. DOI: 10.1016/j.rse.2015.12.042
Sulla-Menashe, D., Kennedy, R. E., Yang, Z., Braaten, J., Krankina, O. N., Friedl, M. A. 2014. Detecting forest disturbance in the Pacific Northwest from MODIS time series using temporal segmentation. Remote Sensing of Environment. 151, 114-123. DOI: 10.1016/j.rse.2013.07.042
More details may be found in the following project profile(s):
