Supporting Work

Integrated Suite of Models From Prior NASA Funding

Our prior NASA awards supported the development of an integrated suite of models to be used in conjunction with remotely sensed as well as targeted in situ observations with the objectives of describing processes controlling fluxes on land, their coupling to riverine systems, and the delivery of materials to estuaries and the coastal ocean. The work focused on the Mississippi River- Atchafalaya River Basin (MARB) and northern Gulf of Mexico (GOM). The resulting models are capable of reproducing observations in both the river and coastal ocean sub-domains and have the capacity of hindcasting and forecasting land use and climate change scenarios. However, further evaluation, improvement, expansion, and documentation are much desired. A major objective of our present proposed work is to expand the spatial domain of the model framework to include the regions of the southern and southeastern U.S. and the South Atlantic Bight (SAB). The South, Southeast and MARB regions are undergoing significant and distinct anthropogenic and climate-related changes. Indeed, the upland areas of the MARB have experienced large-scale land cover and land use change since the mid-1800’s [Chen et al., 2006] and are likely to undergo further rapid development in the coming years [Alig, 2009; Napton et al., 2010]. Terrestrial land cover and land use are major factors influencing the strength of the terrestrial carbon sink [Chen et al., 2006; Houghton et al., 1999; Reay et al., 2007; Tian et al., 2012c] and the delivery of sediment [Vörösmarty et al., 2003], organic carbon [Meybeck, 1993], dissolved inorganic carbon (DIC) [Cai et al., 2008; Raymond et al., 2008], nutrients [Howarth et al., 2011] and other constituents from land to rivers, and eventually to the ocean [Ciais et al., 2008].

With support of our prior NASA projects, our Auburn team (Tian and co-workers) developed and applied the Dynamic Land Ecosystem Model (DLEM; Figure 1) to quantify terrestrial carbon storage and fluxes including characterization and quantification of soil carbon and vegetation biomass and land-atmosphere fluxes of CO₂ and CH₄ in the terrestrial ecosystems of MARB and southeastern US, and lateral fluxes of water, nutrients and organic and inorganic carbon from land to the GOM [Tian et al., 2010a; Tian et al., 2011a; Tian et al., 2012b; Zhang et al., 2012]. The DLEM (Fig. 2) is a process-based land ecosystem model that uses spatially referenced information on atmospheric chemistry, climate, elevation, soils, and land cover/land use to estimate daily carbon (CO₂, CH₄), nitrogen (N₂O), and water fluxes and pool sizes in terrestrial ecosystems and the riverine fluxes of carbon (DOC, POC and DIC), nutrients and water from land to coastal ocean. The DLEM includes five major components that focus on different processes: (1) biophysics, (2) plant physiology, (3) soil biogeochemistry, (4) dynamic vegetation and land-use, and (5) disturbance, land use and management. Recently, with the support of our prior NASA IDS project, we have improved the DLEM by incorporating the processes of soil erosion with the modified universal soil loss equation (MUSLE) water routine process based on global river network data sets (GTN30) [Vörösmarty et al., 2000], and a simplified nitrogen removal process in the river systems [Alexander et al., 2000; Wollheim et al., 2008]. We expanded the DLEM model with a Nutrient Export (NE) component to track the leached carbon (DOC, POC), and nitrogen (DON, DIN, and PON) from terrestrial ecosystems to aquatic systems [Liu et al., 2009].

In addition to model improvement, Tian (Co-I) and his team have developed a series of gridded datasets, including time-invariant drivers (soil properties, DEM, river network) and time variant drivers (climate, atmospheric chemistry, land use and land management practices), For the entire domain of MARB and southeastern US, we developed input data with a spatial resolution of 5 arc-minutes (about 9.2 km × 9.2 km at the equator) as input for the DLEM model. A detailed description of these data sets has been provided in our published work [Liu et al., 2013]. In addition, Pan (Co-I) and her team have collected and classified multiple time periods of remote sensing images including Landsat MSS and TM/ETM+ for the southeastern US, as well as MODIS and AHVRR data for the entire land domain.

The NCSU team (R. He and co-workers) constructed a regional coupled ocean circulation and marine ecosystem model. This 3-dimensional coastal modeling system is built upon a regional-scale circulation model (Fig. 2). The model domain covers the entire SAB and its upstream Gulf of Mexico areas (hereafter SABGOM), allowing the Loop Current/Florida Current/Gulf Stream dynamics to be properly represented. This regional model was implemented based on the Regional Ocean Modeling System [Haidvogel et al., 2008] (ROMS). Its computational kernel includes high-order advection and time-stepping schemes, weighted temporal averaging of the barotropic mode to reduce aliasing into the slow baroclinic motions, and conservative parabolic splines for vertical discretization. A redefinition of the pressure-gradient term is also applied in ROMS to reduce the pressure-gradient truncation error, which has previously limited the accuracy of terrain-following coordinate models. The spatial resolution of the SABGOM ROMS is 5 km. Vertically, it has 36 layers weighted to better resolve surface and bottom boundary layers. At SABGOM ROMS eastern and southern open boundaries, deep ocean temperatures, salinities and transport values are specified using the data-assimilative Global Hybrid Coordinate Ocean Model [Chassignet et al., 2007]. We augmented the HyCOM open boundary transports with barotropic tides from the global analysis of Egbert and Erofeeva [2002]. Surface meteorological forcing is obtained from the NCEP North America Regional Reanalysis (NARR, 3hourly, 32 km resolution). At the land boundary, real-time river runoff data from USGS gauges are used to provide freshwater input. SABGOM ROMS has been used to perform both a long-term hindcast to study the inter-annual variability in the coastal circulation and extreme events such as the 2003 South Atlantic Bight cold event [Hyun and He, 2010] and the coastal ocean response to Hurricanes [Zambon, 2009; Zambon et al., in review]. Coupled with SABGOM circulation is a marine ecosystem model described by Fennel et al. [2006] and Hofmann et al. [2011]. This ecosystem model (see Figs. 2 and 3) is a descendant of the Fasham et al. [1990] formulation that explicitly includes nitrogen and carbon cycles.

Results from our previous work are being served on a Google Earth geo-spatial portal which can be accessed here. Currently, available information is primarily for ocean products (currents, salinity, nutrients, chlorophyll, and primary production). As part of our current NASA CMS project, we are populating the data portal with terrestrial information including dissolved inorganic carbon leaching. Other products being developed include freshwater, evapotranspiration, and dissolved organic carbon and nitrogen leaching.