North-Central Section - 39th Annual Meeting (May 19–20, 2005)

Paper No. 5
Presentation Time: 8:00 AM-12:00 PM


RAY, Sushmita1, EDWARDS, R. Lawrence1, ALEXANDER Jr, E.Calvin1 and DORALE, Jeffrey A.2, (1)Geology & Geophysics, Univ. of Minn, 310 Pillsbury Dr. SE, Minneapolis, MN 55455, (2)Geoscience, Univ of Iowa, 121 Trowbridge Hall, Iowa City, IA 52242,

We examined fluorescent banding and detrital layers in stalagmites from Spring Valley Caverns in southeastern Minnesota to construct a 1400 year flooding record with annual resolution. The fluorescent bands were imaged with a laser confocal microscope, using excitation and emission wavelengths of 488 nm and 522 nm respectively. Each band comprised a fluorescent and dark couplet varying in width between 50 and 200 microns. 13 230Th dates were obtained for the 1400 year old section with errors ranging from ± 8 to ± 40 years depending on the age. The fluorescent part of the couplet was counted to obtain a band age. Given a ± 3 % error in band counting, the number of bands agrees within error with the 230 Th dates, indicating that the bands are annual. In addition to the dark portion of the couplet, we noticed yet another type of band, which was darker and had an average thickness of 9 microns. These dark bands were composed of detrital particles, which when analyzed with electron microprobe indicated mixtures of ilmenite, plagioclase, quartz, dolomite and clay minerals. We hypothesize that these clay layers may represent major rainfall events that flooded the cave, inundated the growing speleothem and deposited detritus on its growing surface. To test this hypothesis, a time series of these clay layers was constructed for the past 100 years and compared with the instrumental record of precipitation from Spring Valley. Each of the flood events corresponds to a year when there was an extreme rainfall event. However, there are some extreme rainfall events that do not appear in the cave record. These observations suggest that these clay layers may be used as a proxy for flood events.

We found 97 flood events between 600 and 1998 A.D., an average of 6 per century. The number of flood events per century was high from 600 to 900 A.D. averaging 9 events per century. Between 900 and 1700 A.D. flood events were less, about 5 per century with no flood events recorded between 1260 and 1360 A.D. From ~ 1700 to 1998 A.D. the frequency increased to earlier values again averaging 9 events per century. Our record shows that the frequency of flood events changed over the span of the Little Ice Age (1450-1850 A.D.), being low in the beginning and increasing to earlier values after 1700 A.D.