Comment on “heat flow from four new research drill holes in the Western Cascades, Oregon, U.S.A.” by S. E. Ingebritsen, M. A. Scholl and D. R. Sherrod [Geothermics 22, 151–163 (1993)]

Recently Ingebritsen et al. (1993) presented four new heat flow data points along the High Cascades-Western Cascades boundary region in an area extensively studied and discussed by Blackwell et al. (1978, 1982, 1990a, other references are listed in these papers). Ingebritsen et al. (1993) propose that the heat flow values from their four new sites are inconsistent with heat flow contours based on over 50 points presented in Blackwell et al. (1978, 1982, 1990a), but instead are consistent with an alternativeʹcontouring of the heat flow data described by Ingebritsen et al. (1992). They further conclude that the results support the model of a narrow magmatic heat source flanked by an area of high heat flow due to lateral groundwater flow, an alternative model proposed and rejected by Blackwell et al. (1978, 1982, 1990a). They discount the model favored by Blackwell et al. (1978, 1982, 1990a) that utilizes a wide conductive heat source with irregular fluid flow thermal anomalies related to the details of the local geology. We argue here that the results from the four new heat flow sites described by Ingebritsen et al. (1993) further confirm the heat flow and geothermal gradient distribution described on the basis of the extensive heat flow data set collected by Blackwell et al. (1978, 1982, 1990a). Furthermore, as discussed extensively in these references and summarized briefly below, the evidence does not favor groundwater flow as a significant factor affecting the observed regional heat flow pattern, even though convective disturbances are present and are fully described by Blackwell et al. (1990a). The regional crustal thermal regime of the Cascade volcanic arc in the northwestern U.S. and southwestern Canada is arguably the most completely characterized of any 703 704 D. D. Blackwell and G. R. Priest volcanic arc (see additional papers by Blackwell et al., 1990b; Lewis et al., 1988). The results are of possible world-wide interest in terms of characterizing the thermal structure of volcanic arcs and their associated volcanism. The results are also interesting from a more immediate view, the geothermal energy resources of the U.S. and Canada. Thus it is important to come to closure on interpretations of data that reflect on the thermal character of this important province. To put these new data in perspective, a brief recap of the history of investigations is helpful. The first heat flow measurements in the Cascades were made in 1976 in the vicinity of the hot springs along the High Cascades-Western Cascades physiographic boundary (Fig. 1) in Oregon. Wells were drilled to a depth of 100-150 m specifically for heat flow analysis. To our great surprise, the data in general were quite homogeneous in value (the average was 105 + 9 mW m -2) and the thermal regime appeared to be dominated by conduction in all rocks with significant diagenetic or hydrothermal alteration. This generally included most rocks older than about 5 Ma, but also some younger rocks in local areas. Very high values typical of geothermal systems were found only in immediate proximity of the known hot springs. In the Oregon Cascades, because of the continued igneous activity, extensive alteration of rocks now found at the surface is usually found on a regional basis in exposed rocks between 2 and 5 Ma in age. Uniform high heat flow was unexpectedly observed up to 30 km to the west of the Quaternary volcanic axis before the expected transition to the low heat flow region characteristic of the outer arc block to the west was identified. Several models of heat source distribution in the Cascades and possible groundwater flow regimes that might explain the results were presented (Blackwell et al., 1978, 1982), including the exact model proposed later by Ingebritsen et al. (1992; see also |ngebritsen et al., 1989). In stark contrast to the results in the wells in the altered rocks, little or no temperature change with depth was found in 100-150 m deep wells in the mid-to-late Quaternary rocks. The most straightforward interpretation is that these rocks are affected by convective washout (cooling) by meteoric water. Later, results from deeper wells reinforced these observations. Subsequent drilling in the heat flow high demonstrated that sites in rocks with intermediate alteration give internally consistent heat flow values that are generally 50 + 25% of the regional value of about 100 mW m 2 (Steele et al., 1982; Blackwell et al., 1990a). This somewhat systematic variation is related to the decrease in permeability associated with increased alteration. Some skepticism greeted these results, including some in our own minds. In 1979 seven additional holes were drilled specifically at high elevation, and away from the hot springs, to check the 1976 results. The results of these new data (described by Blackwell et al., 1982) verified the conductive nature of the thermal field in shallow holes in the >5 Ma rocks and the presence of a high heat flow well (about 30 km) to the west of the volcanic arc axis. In addition, as part of other projects, three deeper wells were drilled in younger rocks and were included in the 1982 data set. These deep holes in the younger rocks indicated a zone of thermal washout by groundwater flow to depths of up to 300 m where unaltered rocks occurred at the surface, with heat flow values below the washout zone also being 100 + 10 mW m -e in two of the three holes. In parallel studies a number of wells were drilled in volcanic rocks, to depths of up to ! .8 kin, north of the area shown in Fig. 1 near Mt Hood and in the Southern Washington Cascades (Steele etal., 1982; Blackwell etal., 1990b). The deeper holes in both areas proved the reliability of the heat flow values determined in the shallow holes.