T46 Construction and Geology of the Massachusetts Water Resources Authority Tunnel, Eastern Massachusetts II

Monday, 5 November 2001
1:30 PM - 5:30 PM

CORRELATION OF LINEAMENTS TO GROUND WATER INFLOWS IN THE MWRA TUNNEL

Paper number: T46-3
Program: Technical Sessions
Time: 1:30 PM – 5:30 PM

Authors

Abstract

Lineaments derived from three image types (1:80,000 black and white (BW), 1:58,000 color infrared (CIR), and 1:250,000 side-looking airborne radar (SLAR)) were compared to water-bearing features within a 9.6 km section of tunnel being constructed through foliated crystalline metamorphic bedrock in a glaciated region of eastern Massachusetts. Lineaments drawn by three observers during two independent trials (N=9137) were reduced to three sets (one per image type) of coincident lineaments (N=794). Thirty-five coincident lineaments crossed the tunnel. Nineteen discrete flow zones, each producing less than or equal to 19 L/min, were identified in the tunnel and used to quantify the reliability of lineament analysis as a method of predicting water-bearing features in glaciated metamorphic rocks.

Thirteen (68%) of the flow zones correlate with coincident lineaments, six zones correlate with more than one image type, and one zone correlates with all three image types. Overall, it is difficult to distinguish lineaments that will be successful in predicting water-bearing zones from those that will be unsuccessful without considering other corroborating evidence. Most of the observed flow (80%) correlates with northwest-trending coincident lineaments. However, the majority of the flow (67%) associated with these lineaments is produced from structures that strike to the north or northeast. In addition, only fifteen of the thirty-five coincident lineaments correlate with the flow zones indicating that twenty lineaments are not associated with any appreciable flow. Six flow zones are undetected by the lineament analysis.

In this study, BW lineaments are able distinguish high-yield through-going structures (at the 90% confidence level) with greater reliability than the SLAR or CIR lineaments. However, linking bedrock type, overburden type, topographic position, and proximity to surface water bodies with lineament analysis improves the predictive capability of the lineament method.