home>>projects >>WDC's CMP at Umatilla

washington demilitarization company’s comprehensive monitoring program
at the U.S. army’s umatilla
chemical weapons depot

GiSdT (Guided Interactive Statistical Decision Tools) was developed by Neptune and Company to provide statistical guidance, perform analyses, and to offer an avenue for sharing data and analyses with stakeholders. We have implemented GiSdT to evaluate the monitoring data that are currently being collected at a former chemical munitions facility. This work is being performed for the Washington Demilitarization Company’s (WDC) Comprehensive Monitoring Program (CMP) at the U.S. Army’s Umatilla Chemical Weapons Depot.

Data are available from several different media, including soil, air and biota. The primary transport mechanism is airborne deposition. The data are being collected as part of a monitoring program that provides data every quarter. Each quarter approximately 3500 new samples are collected and analyzed for a suite of analytes including chemical agent. The data are updated each quarter into an SQL database that N&C stores and maintains. The Umatilla-GiSdT application accesses the data through SQL queries. Consequently, the statistical analyses are updated as soon as the new data are loaded into the system. The interactive and Open Source web-based tools that are used are very powerful for this type of monitoring application. We have also automated the reporting of quarterly data into the required quarterly results such that all standard reports and analyses are run automatically, and resources are then only spent on describing interesting results or trends.

This application includes spatio-temporal data for soil, air and biota. The data are overlaid on GIS images (shape files). Scaleable Vector Graphics (SVG) are used to enhance the spatial presentation and to allow the user to click on a specific location to access the temporal information for that location. The temporal information is then presented using temporal plots, and is analyzed using parametric and non-parametric statistical methods. EDA of the data is also available in the form of box plots, histograms, quantile plots and density estimates. Comparisons are made to baseline conditions so that the project team can quickly determine which data exceed thresholds, and can also determine if there is a spatial or temporal trend of concern that might cause exceedance of the thresholds at some point in the future.

Neptune provides all data results and statistical analysis input for quarterly reports. We then conduct more thorough and intricate analyses for the annual reports. The annual reports provide our opportunity to present complex statistical analyses and explain the implications of our findings.