Application Design

Overview

The SouthFACT application supplies users with three main features: 1) a free account and subscription service, 2) a web mapping viewer, and 3) a custom request service.

LandsatFACT Applications

By creating an account, users can define area(s) of interest and request to be notified when new products are available. The Forest Change Viewer displays the latest (and historical) change products, allowing users to navigate and explore change products online. If users with an account have a need to identify changes over a specific time period and area of interest, the Custom Request service can process requests and send the data for download to the user.

 

To provide these features, SouthFACT makes requests of the USGS Earth Resources Observation and Science (EROS) Data Center on behalf of the user, seamlessly downloading the imagery for processing. As new imagery becomes available from Landsat 7 and Landsat 8, the application determines whether the imagery is at least partially free of clouds before downloading and processing to produce the Vegetation Indices. Usable imagery is compared to the most recently acquired imagery for the same World Reference System (WRS) - 2 in order to identify differences in vegetation on the landscape. The most recent vegetation changes across the 13 Southern Region States can be observed by the user in the Forest Change Viewer.

 

Software Architecture

The SouthFACT application is portioned into a Client / Server architecture composed of general purpose open source products (configured for this project) as well as JavaScript, PHP, and Python scripts constructed specifically for this project.

 

The Apache web server directs client requests through the open source Web Map Server (WMS) provided by the MapServer platform from the Open Source Geospatial Foundation. WMS provides the change products shown in the Forest Change Viewer along with spatial and geographic information stored in the PostGIS extension to the open source PostgreSQL database. PHP scripts developed for this project use the SOAP interface to the EROS system to acquire Landsat imagery and record a history of the downloaded images in PostgreSQL. Python scripts developed for this project search the image metadata history in PostgreSQL and retrieve the most appropriate raster (image) data from the server’s file system for processing. Python scripts then identify differences between Landsat images that are not overly obscured by clouds using the Project Methods. User account information is managed by the open source content management system Drupal.