27.04.2013 Views

Figure I Generalized map of the Wilbur Mining ... - University of Utah

Figure I Generalized map of the Wilbur Mining ... - University of Utah

Figure I Generalized map of the Wilbur Mining ... - University of Utah

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

J<br />

0<br />

10<br />

X<br />

a.<br />

^20<br />

30<br />

2") 22<br />

/<br />

,/<br />

\<br />

\<br />

\<br />

38°C/km-J<br />

300m TEST WELLS<br />

\ \<br />

\(J1<br />

24<br />

K<br />

Two 300m test wells were drilled on <strong>University</strong><br />

land adjacent to Las Alturas under <strong>the</strong> supervision<br />

<strong>of</strong> L. Chaturvedi <strong>of</strong> <strong>the</strong> Civil Engineering<br />

Department at NMSU. The wells were sited on <strong>the</strong><br />

basis <strong>of</strong> <strong>the</strong> interpretation <strong>of</strong> <strong>the</strong> data from <strong>the</strong><br />

shallow gradient hole pr<strong>of</strong>ile, at <strong>the</strong> locations<br />

shown in <strong>Figure</strong> 1, based on <strong>the</strong> following logic.<br />

The first test well, DTI, was sited 100m east<br />

(downdip on <strong>the</strong> assumed fault) <strong>of</strong> <strong>the</strong> apparent<br />

peak <strong>of</strong> <strong>the</strong> anomaly defined by <strong>the</strong> temperature<br />

gradient pr<strong>of</strong>ile. With this well it was hoped to<br />

intersect and stay in <strong>the</strong> anomaly source to total<br />

depth. The second well, DT2, was drilled 0.4<br />

miles west <strong>of</strong> DTI along <strong>the</strong> pr<strong>of</strong>ile, between<br />

shallow wells HMSU-7 and -8. This site was selected<br />

so that if <strong>the</strong> interpretation <strong>of</strong> <strong>the</strong> source<br />

<strong>of</strong> <strong>the</strong> anomaly as a narrow fault-controlled zone<br />

<strong>of</strong> rising hot water were incorrect, this well<br />

would intersect a more laterally extensive source<br />

at a greater depth from DTI, but at a closer<br />

distance to <strong>the</strong> potential user, <strong>the</strong> NMSU campus.<br />

If <strong>the</strong> narrow fault source interpretation were<br />

correct, however, DT2 would provide information<br />

about <strong>the</strong> lateral flow from <strong>the</strong> system, and act as<br />

a site for a reinjection well, if required. Wells<br />

DTI and DT2 were drilled in December 1978 and<br />

January 1979 and completed as temperature test<br />

wells at depths <strong>of</strong> 300 and 360m respectively with<br />

two-inch water filled casing.<br />

Temperature data from <strong>the</strong> two 300m test wells<br />

are shown in <strong>Figure</strong> 4. The first well, DTI,<br />

TEMP CO<br />

26 28 30 32 34 36<br />

NMSU SH/lLLOW GEOTHERMAL TEST<br />

BOREHOLE TEMPERATURE LOGS<br />

\ , \ ^ ^ 0 \ ^ LOGS WITHIN 72HRS OF DRILLING<br />

\ V ^ ^ ^<br />

\^x<br />

\ " N V ^ V \ /-446°C/km<br />

320°C/krn—1 TT L-4 33 .c/km<br />

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!