Spatial Time-Series: Pollution Pattern Recognition under Irregular Interventions.

Abstract

The Fernald Environmental Restoration Management Corporation (FERMCO) has noted the introduction of arsenic contamination to groundwater around the area of the groundwater recovery system, which captures uranium contaminated groundwater. The introduction of arsenic occurs during high levels of pumping and is particularly sensitive to the western two of the five pumps. Auto-Regressive Moving Average (ARMA) and Spatial-Temporal ARMA (STARMA) empirical analyses are used to model the level of arsenic contamination found through time. The intervention of varied levels of pumping is modeled with a transfer function using analytic techniques to create a causal intervention transfer function input series to give physical meaning to the impulse response weights found. Spatial weights employed in the STARMA modeling are also created using analytic, causal methods. Results suggest a physical interpretation of the relationship between a particular level of pumping and the amount of arsenic to be found at the site in the temporal case, while including the effect of the pumping on a site of interest's neighbors in the spatial-temporal case. Models presented may be employed in the forecasting of arsenic levels at monitoring well sites due to a given change in the operation level of the groundwater recovery system. (AN)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 1995
Accession Number
ADA293830

Entities

People

  • Samuel A. Wright

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Air Force
  • Algorithms
  • Contamination
  • Environmental Restoration And Remediation
  • Geographic Information Systems
  • Groundwater
  • Heuristic Methods
  • Identification
  • Image Classification
  • Information Science
  • Monitoring
  • Pattern Recognition
  • Random Variables
  • Reasoning
  • Recognition
  • Standards
  • Transfer Functions

Readers

  • Environmental Remediation and Restoration.
  • Mathematics or Statistics
  • Neural Network Machine Learning.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference