Characterization of seasonal and inter-annual variability in global water bodies using annual MODIS water maps 2000 - 2016

Accurate maps of surface water resources are critical for long-term resource management, characterization of extreme events, and integration into various science products. Unfortunately, most of the currently available surface water products do not adequately represent inter- and intra-annual variation in water extent, resulting from both natural fluctuations in the hydrologic cycle and human activities. To capture this variability, annual water maps were generated from Terra MODIS data at 250 m resolution for the years 2000 through 2016, using the same algorithm employed to generate the previously released MOD44W Collection 5 static water mask (Carroll et al., 2009). Following efforts to verify the data and remove false positives, the final maps were submitted to the Land Processes DAAC for publication as MOD44W Collection 6.1. Analysis of these maps indicate that only about two thirds of inland water pixels were persistent throughout all 16 years of data, meaning that roughly one third of the surface water detected in this period displayed some degree of inter-annual variation. In addition to the annual datasets, water observations were aggregated by quarter for each year from 2003 through 2016 using the same algorithm and observations from both Terra and Aqua. Analysis of these seasonal maps is ongoing, but preliminary investigation indicates they capture dramatic intra-annual fluctuations of water extent in many regions. In cloudy regions, it is difficult or impossible to consistently measure this intra-annual variation without the twice-daily temporal resolution of the MODIS sensors. While the moderate spatial resolution of MODIS is a constraint, these datasets are suitable for studying such fluctuations in medium to large water bodies, or at regional to global scales. These maps also provide a baseline record of historical surface water resources, against which future change can be compared. Finally, comparisons with the MOD44W Collection 5 static water mask indicate that major changes have occurred in many areas since the early 2000s, rendering these maps an equally valuable update for static water masking applications. ReferencesCarroll, M.L., Townshend, J.R., DiMiceli, C.M., Noojipady, P., & Sohlberg, R.A. (2009). A new global raster water mask at 250 m resolution. Int J Digit Earth, 2, 291-308.