proudly sponsored by
|
archiving of SOWACS proudly sponsored by |
|
Hi Steve ans other list members, I've been a lurker for about a week on SOWACS now and have missed some of this discussion on the K_moisture relation. I am assuming that gravimetric measurement of moisture content is being used to calibrate the TDR equipment as this is probably (?) the most precise, repeatable and reliable method. My questions are: 1.: If you use gravimetric moisture measurements with an error of plus-or-minus 2% to calibrate the TDR equipment how can you get a precision of better than 0.02 m3/m3 for TDR by doing your own calibration (assuming bulk density of 1000 kg/m^3 to convert between gravimetric and volumetric moisture contents ). Why not just use Topps as they stand for all mineral soils with organic matter content between say 2 and 18%? The reason I ask is that we have some long term continuous TDR measurements (broken into periods of 90 days or more) of moisture content and the resolution of the record appears to be much greater than plus-or-minus 0.02 m^3/m^3 -- in fact resolution appears to be about 0.002 m^3/m^3. I can post a PCX or GIF piccy of the traces to any interested people but won't do it here as it will just use bandwidth. The soils are silt loams with about 18% clay and smectites as the predominant caly minerals (vertisols ?). 2. Has anyone listed the relevant papers on this list so that other members can follow the more technical aspects of the discussion or get up to speed if they want to ? I am willing to maintain a list of relevant papers and post them periodically if anyone is interested. Comments gratefully received. Regards Grant Upchurch Soil Scientist "Some countries have oil, New Zealand upchurch@nzdri.org.nz has soil" Prof. HS Gibbs (pedologist). >>> srevett@ag.gov wrote on 18/December/1996 11:37pm >>> Dear Josep Maria, My understanding of the literature is that Topp's equation hasn't proven to fit any particular soil 'exactly' but that it fits the ensemble of soil water content and K data very well. Some persons with whom I've discussed this have shown me results that indicated that Topp's equation was biased by 0.02 to 0.05 m^3/m^3 but their wave form interpretation methods were different from what Topp used. When they re-interpreted their data using methods more similar to Topp's they found that Topp's equation was not significantly different from theirs. This issue of wave form interpretation is critical for calibration efforts. Most papers in the literature do not describe the interpretation methods used in enough detail to allow reproduction of the method, or understanding of what was done. Different interpretation methods can easily produce water content differences of 0.03 m^3/m^3 or more. And, simply citing one of Topp's paper is hardly adequate - no single paper does this topic justice. The take home message is that if you want water contents accurate to better than +/- 0.02 m^2/m^2 then do your own calibration. That calibration will include effects both of your particular soil type, and of your wave form interpretation method. If your wave form interpretations are reproducible then your calibration should work. In reference to fitting the data both ways; K=f(theta) and theta = f(K): my experience is that there is little difference if the data exhibit small scatter. If there are important discrepancies, I would re-do the calibration and try to get data with less scatter. Yes, I know this isn't always possible. Best wishes, Steve Steve Evett srevett@ag.gov USDA-ARS, P.O. Drawer 10, Bushland, Texas 79012 U.S.A. (1/2 mile W., Interstate-40 S. access road) Tel. 806-356-5775, FAX: 806-356-5750