Transcript Slide 1
GC51A-0449 Resolving Terrain Effects in Borehole Temperature Profiles Will Gosnold and Shannon Heinle, Department of Geology and Geological Engineering, Grand Forks, ND 58202-8358 [email protected] The types of terrain effects on surface temperature that disturb the T-z profile we can discern from air photos, satellite imagery, topo maps, and historical records include: change in land cover, proximity to water bodies, and topography. Degrees (C) 1.5 2 2.5 0 0 50 50 100 100 150 Figure 6 200 250 300 350 150 meters into clear area 150 meters into forest 25 meters into open area 25 meters into forest Depth (m) 0 0.5 150 1 1.5 2 2.5 3 Figure 7 Figure 9 200 250 300 150 meters into forest 150 meters into open area 25 meters into open area 25 meters into forest 350 The time-transient effects of borehole proximity to the edge of a clearing in the case of clearing of old growth forest circa (Fig. 6) and regrowth of forest (Fig. 7) can be modeled. However, too much is unknown about the amount and timing of surface temperature change to permit precise corrections to borehole T-z profiles. Figure 10 Figure 11 0 0 0 -50 -50 -50 -100 -100 -100 -150 Depth (m) 1 Depth (m) 0.5 Degrees (C) Depth (m) 0 Depth (m) We have tested the accuracy of borehole temperature profiles in tracking surface air temperatures on short timescales (< 20 y) by direct comparison of multiple borehole logs and the meteorological record from nearby climate stations. We have conducted similar tests on long timescales (~100 y) by comparing synthetic borehole T-z profiles generated from climate data with measured T-z profiles. In both tests, we found that the borehole temperature profiles accurately record the surface air temperature record. Examples the long-term tests in which borehole sites show warming trends that parallel the meteorological record are shown below in the Figures 1 & 2. However, when we examined the global set of borehole data used for climate reconstructions, we found large scatter in the data with some borehole sites showing cooling where the meteorological record indicates warming. We suspected that because the data were not screened initially for terrain and cultural effects, a screening for these factors might yield a better data set. Subsequent examination of the locales of the 130 sites in US segment of the global borehole data set by remote sensing, topo maps, and Google Earth indicates that at least 33 sites may have terrain effect disturbances to the T-z profiles. We have used remote sensing and thermal modeling of some of the sites to determine if corrections can be applied to the data. [email protected] -150 -150 -200 -200 -200 -250 -250 -250 -300 -300 0 50 100 150 200 Distance from lake (m) 250 300 -300 0 50 100 150 200 250 300 0 Distance from Lake (m) 50 100 150 200 250 300 Distance from lake (m) Comparison of T-z profiles and subsurface isotherms affected by proximity to a water body. The models are based on a site in Canada (Image 1). Fig. 9 shows the lake effect without climate change. Fig. 10 shows the effect of 2ºC warming and Fig. 11 shows the effect of 2ºC cooling. An appropriate profile from the model for no climate change could be used to correct the observed T-z profile. Figure 1 Image 1 Figure 2 Image 2 Synthetic T-z profiles generated from USHCN annual temperature data match observed T-z profiles in the mid-continent of North America. Figure 8 Figure 3 Figure 4 Figure 5 Significant scatter in the global borehole T-z profiles (Fig. 3) is largely a function of data location by latitude, but some may be due to terrain and cultural disturbances that were not screened out of the data. Inversion of some of the data from the mid-continent of North America (Fig. 4), a region in which the data were screened, generates a relatively scattered, although consistent, set of results. Some early analyses of composite borehole T-z inversions with no screening for terrain effects yielded “spaghetti” plots as simulated in Fig. 5. Figure 8 shows the effects of changes in land cover on subsurface temperature at two sites in Nebraska and one in Vermont. The Fremont, NE site converted from overgrowth with 1.5 m tall weeds to a heavily grazed pasture between 1981 and 1996. The Wayne, NE site experienced increasing tree shading during the same period. The Vermont site experienced cooling between 1965 and 1992, but the air temperature record shows warming. The T-z profile labeled Model shows the theoretical effect the air temperature should have had on the 1965 T-z profile in 1992. Surface temperature changes at the sites cannot be documented precisely enough to permit corrections to the borehole T-zs. Modeled T-z profiles (Fig. 13) and isotherms (Fig. 14) across a valley in Utah (Image 2). The site is one of the US sites in the borehole database and has an obvious disturbance to the T-z profile. Conclusions: Topographic corrections to T-z profiles can be modeled relatively accurately and are the most likely candidates for use in improving the borehole database. In all other situations, the amount and timing of surface temperature changes is insufficiently known to permit precise corrections. Alternatively, a careful screening of the data and exclusion of potentially disturbed sites could improve the overall results. This research is supported by National Science Foundation Award ATM-0318384