Rapid Prototyping Capability
Introduction
The Institute for Technology Development (ITD), Science Systems and Applications,
Incorporated (SSAI) and Computer Sciences Corporation (CSC) have formed a “Stennis Team” at the NASA Stennis Space Center (SSC)
to support the Mississippi State University
GeoResources Institute in a Rapid Prototyping Capability Project funded by the Mississippi Research Consortium (MRC). ITD is
participating in a related study that uses MODIS data products to perform crop type assessment. It is titled “A Phenological
Approach for Mapping Crop Types using Hypertemporal MODIS NDVI Data Products” and is intended to support the Production Estimates
and Crop Assessment Division (PECAD) of the U.S. Dept. of Agriculture’s (USDA) Foreign Agricultural Service (FAS) to predict
production estimates.
The Rapid Prototyping Capability (RPC) has two primary functions. The first is to extend the Crop Type Mapping project to
include simulated VIIRS data. The other function is to assist in the development of an RPC node at the NASA Stennis Space Center
(SSC).
Project Description
The Crop Type Mapping project uses Moderate Resolution Imaging Spectroradiometer (MODIS) data to perform crop type
assessment. The RPC project extends the crop type mapping to use simulated Visible Infrared Imaging Spectrometer (VIIRS) data
instead of MODIS data products. The VIIRS data will be simulated from MODIS data products. The spatial resolution of the VIIRS
Visible Near-InfraRed (VNIR) bands are 400 meters compared to the 250 meter spatial resolution of the MODIS VNIR bands. The red
and infrared bands of the VNIR sensors are used to compute the Normalized Difference Vegetation Index (NDVI) equation. NASA is
in the process of phasing out the MODIS sensor and continuing its missions with the VIIRS sensor. Applications currently using
MODIS data should be tested with VIIRS data sets to determine if the VIRRS data sets are adequate to continue the experiments
supported by MODIS. The purpose of this processing is to determine if and/or to what extent the crop type mapping accuracy
realized by using the MODIS data is diminished when using VIIRS data.
In addition, the purpose of this project is to assist in the development of an RPC node. The RPC node will consist of an
operational computing environment that will allow for the easy and efficient prototyping of experiments designed to reveal the
degree to which NASA sensor data or simulated sensor data can address specific scientific questions. The equipment for the RPC
Node is housed at a building at SSC. However, the intent is that the RPC Node can be replicated at any of the NASA centers.
Crop Type Mapping using VIIRS Simulations
CSC/SSAI has developed two products that directly impact the project. They are the Applications Research Toolbox (ART) and the
Time Series Product Tool (TSPT). The ART is an integrated set of algorithms and models that allows users to perform a suite of
simulations and statistical trade studies on remote sensing systems. Specifically, the ART provides the capability to generate
simulated multispectral image products at various scales from high spatial hyperspectral and or multispectral image products. A
Time Series Product Tool (TSPT) has been created to generate layerstacks of various data products to assist in time series
analysis and to create videos of sequences of temporal images.
The Stennis Team includes ITD and CSC/SSAI. CSC/SSAI is building upon the ART and TSPT to create high performance VIIRS
simulation capability that can be used to rapidly prototype products for evaluation. The VIIRS simulator will be a systems
level simulator that will emulate the spatial, spectral, radiometric properties defined by the VIIRS specification. Products
will be simulated through system level modification of existing similar products as well as the use of Algorithm Theoretical
Basis Documents. The TSPT will be modified to account for the differences in potential cloud statistics and quality assurance
flags.
ITD is using these hypertemporal simulated VIIRS data sets in application activities. These activities include using ART
and TSPT to build the data sets and then to use them in classification algorithms. Accuracy assessments will be performed to
compare with accuracy assessments of classifications performed with MODIS data products.
RPC Node Creation
The Stennis Team will support the MSU GRI in implementing and connecting Rapid Prototyping Capability nodes to the hub at
NASA Stennis Space Center (SSC). This support will include providing an interface between the University team’s systems and
NASA SSC’s implementation, architecture and IT support. NASA SSC is providing the physical space needed to establish these
nodes. Initial case studies and tests will be supported as well as providing other technical support for NASA SSC to operate the
Rapid Prototyping Capability.
The tools described can provide a rich data set to evaluate NASA Sun-Earth System Sensors. In order to show the utility of
these simulated and enhanced Sun-Earth Systems data sets, several evaluations will be performed. These evaluations will address
five science questions within the national applications of Agricultural Efficiency, Aviation, Air Quality, Carbon Management,
Coastal Management, Disaster Management, Ecological Forecasting, Energy Management, Homeland Security, Invasive Species, Public
Health, or Water Management.
The evaluations will be characterized by a quick assessment of the ability of NASA products and models to improve the
effectiveness of Decision Support Tools. The objective of the enhancing Decision Support Tools is to move research results
into operational use. To that end, government agencies interested in the operational capabilities of Earth-Sun Sensors will be
engaged in the process.
The evaluation process will use the data assimilation, modeling and analysis tools available within the
RPC node. Interaction with the relevant government agency will provide input to assess the requirements of the evaluation.
Included in the final evaluation for a given focus area will be reports which are built around the System Engineering Evaluation
Process.
|