Repeat precision gravity surveys were conducted at Dutch Flats in the last weeks of July, August, September, and October of 2003. All sites were located next to monitoring wells using pounded re-bar to mark the station location. A portable steel plate was placed over the re-bar (without touching) to provide a stable platform for the gravity meter. Height offset between the plate and re-bar was checked each measurement, to correct for slight changes in relative height. Stations were located at wells 1A, 1B, 1C, 1D, 1E, 1F, 1G, 1M, 2B, 2C, 2D, 2F, 3B, 3C, 3E, and 3F. Gravity reference stations were located at the absolute gravity stations Morrill CA (MORRILL_CA) and Scotts Bluff National Monument Baseline (SBNM). These absolute gravity stations were installed for this project, and measured in July and September 2003, during the relative gravity campaigns. Contact Daniel Winester at the National Geodetic Survey (NGS) for more information on the absolute gravity stations. Gravity data acquisition was done in the same general manner as described in Allis et al [2000]; the gravity meter (Scintrex CG-3M 711420) measures at a station for 12 minutes, storing the average and standard deviation of 30 1-second samples. This time series is corrected for sensor tilt (read from internal electronic levels), sensor temperature, solid Earth tides (after Tamura [1987]), linear instrument drift, and non-linear instrument drift. Time series are averaged together using a weighted arithmetic average, ignoring the first 3 minutes of data to remove spring relaxation effects. Linear instrument drift is computed by the meter, using user-supplied constants which are updated on a monthly basis. Non-linear instrument drift is corrected through multiple loops between stations, and a discontinuous empirical function derived from a linear least-squares inversion. Final gravity values have a computed standard error of the mean that is typically 2-5 uGal; for ease of interpretation, the lower signal bound is considered 10 uGal. GPS receivers were used to locate the stations (top of the re-bar) during each gravity survey; two or more GPS receivers were used in a rapid-static mode, with a target occupation length of 1 hour for rover stations. Processing could be done using Trimble Geomatics Office, or another package capable of rapid-static differential GPS. For the July and September campaigns, the local base position (Morrill CA) is best taken from the NGS data collected by Daniel Winester for absolute gravity measurements. The lack of motion at both Scotts Bluff National Monument and Morrill CA benchmarks indicates there is no need for elevation corrections at the reference stations. Rover stations (those at wells) may still contain a subsidence-induced gravity signal; it is extremely unlikely this signal is larger than 9 uGal. This is not significant in the context of the coarse interpretation done here. Gravity changes over the course of the surveys is shown in Table XX. Changes are listed as differences from July 2003, in microGals (0.001 mGal = 1x10^-8 m/s^2); the second number is two standard errors of the mean, in microGals. Note that surveys took two days to complete, although data are presented in a single line; loop closures between the days and differences between repeated stations at the beginning and end of each day are used to tie the multiple days into a single survey. All surveys hold Morrill CA constant inside and between surveys; NGS absolute gravity results show no significant change at Morrill CA between July and September. NGS also show SBNM to be stable over the July-September time frame, which the relative results show (with less confidence due to the August survey). There is nothing in the processing algorithm that requires SBNM to be stable, hence it forms a useful check on the stability and repeatability of the relative surveys. With the rough processing done here, the results are certainly good to within 10 uGal, and this limit could almost certainly be lowered by a more detailed examination of the outlying surveys. This accuracy is also without any elevation corrections due to possible subsidence (which is unlikely, but possible). Table XX. Gravity changes relative to July 2003. Changes are the first number, and are in uGal. Second numbers are 2 s.e., in uGal. For space concerns, the table is broken into 3 parts; stations 1A-1M, stations 2B-3F, and the reference stations. Ellipses (...) indicate multiple readings at a site for a given survey; this is due to the looping for non-linear drift corrections. SBNM has only occupation in Oct 2003 due to wind problems with the first occupation. Relative Gravity Readings (uGal) DATE | 1A | 1B | 1C | 1D | 1E | 1F | 1G | 1M ---------- | ----------- | ----------- | ----------- | ----------- | ----------- | ----------- | ----------- | ----------- 2003/07 | 0 2 | 0 5 | 0 7 | 0 3 | 0 2 | 0 8 | 0 6 | 0 3 ... | 0 2 | 0 2 | 0 2 | 0 2 | 0 2 | 0 2 | 0 3 | 0 2 2003/08 | 12 2 | 23 1 | 13 2 | 10 1 | 6 1 | 21 1 | 10 1 | 12 1 ... | 11 2 | 23 2 | 13 2 | 10 2 | 6 2 | 20 2 | 9 2 | 12 2 2003/09 | 0 8 | 20 2 | 16 4 | 9 2 | 4 2 | 22 1 | 7 2 | 15 4 ... | 0 4 | 20 2 | 16 4 | 9 3 | 4 4 | 22 2 | 7 5 | 15 5 2003/10 | -8 3 | 15 3 | 24 4 | 15 3 | 11 5 | 30 3 | 10 3 | 21 3 ... | -7 4 | 15 3 | 24 4 | 16 3 | 12 3 | 30 4 | 10 3 | 21 3 DATE | 2B | 2C | 2D | 2F | 3B | 3C | 3E | 3F ---------- | ----------- | ----------- | ----------- | ----------- | ----------- | ----------- | ----------- | ----------- 2003/07 | 0 9 | 0 7 | 0 2 | 0 1 | 0 2 | 0 2 | 0 3 | 0 2 ... | 0 5 | 0 17 | 0 2 | 0 1 | 0 3 | 0 3 | 0 5 | 0 3 2003/08 | 4 7 | 3 1 | 9 3 | 0 2 | -27 3 | 3 2 | -8 1 | -10 2 ... | 5 4 | 3 8 | 10 2 | 0 2 | -27 4 | 3 5 | -8 2 | -10 3 2003/09 | -11 2 | | | | -47 3 | | | ... | -11 4 | | | | -48 2 | | | 2003/09 | | 6 3 | -1 2 | -2 2 | | 6 2 | 0 2 | 2 5 ... | | 5 5 | -1 4 | -2 2 | | 6 3 | 0 4 | 2 7 2003/10 | -21 7 | -41 6 | -24 8 | -3 4 | -71 7 | 8 4 | -1 4 | 1 2 ... | -22 7 | -42 7 | -25 5 | -4 3 | -72 7 | 7 4 | -2 3 | 0 3 DATE | MORRILL_CA | SBNM ---------- | ----------- | ----------- 2003/07 | 0 2 | 0 2 ... | 0 2 | ... | 0 1 | ... | 0 2 | 1 2 ... | 0 2 | 1 2 ... | 0 2 | 2003/08 | -1 2 | ... | -2 1 | ... | -2 1 | ... | 3 3 | -10 3 ... | 0 3 | -9 2 ... | 0 3 | ... | 0 3 | -9 0 ... | 0 2 | 2003/09 | -1 2 | ... | -1 2 | ... | 2 3 | ... | 2 2 | ... | -1 3 | -7 1 ... | -1 2 | -8 2 ... | -1 4 | 2003/10 | 1 3 | ... | 4 3 | ... | -2 3 | ... | 1 3 | ... | 0 5 | ... | -1 3 | ... | -2 4 | ... | -1 3 | ... | 0 4 | -7 2 ... | 0 17 | Plotting the gravity changes in map view shows a coherent picture that agrees at least qualitatively with expected trends. Figures 1 through 3 [see files] show the gravity changes, with respect to July, for August to October. The colorbars are in mGal (1000 uGal = 1 mGal), and there are 25 contour levels covering the entire range of the data. The contour intervals and colormap change for each figure. Note that the gravity changes follow the expected pattern of initial buildup near the canals (2B, 3B, 1F), filling of the inter-canal region, and then draining near the canals after the irrigation canals drain. More detailed analysis of the gravity changes can certainly be done, but would first entail a more careful look at the surveys to insure the highest data quality possible. Quantitative interpretation should also use a 3D model to get the best results from the combined gravity and well data. Comparison between the gravity and well data has not been done here, but should correlate. References: Allis, R.G., P. Gettings, D.S. Chapman, 2000. Precise Gravimetry and Geothermal Reservoir Management. Proc. 25th Workshop on Geothermal Reservoir Engineering, Stanford University, pp 179-188. Tamura, Y. 1987. A Harmonic Development of the Tide-Generating Potential. Bulletins d'Informations Marees Terrestres, vol 99, pp 6813-6855.