Date: Thu, 17 Jan 2008 15:07:49 MST To: russell.monson@Colorado.EDU, jia.hu@Colorado.EDU, lynette.laffea@Colorado.EDU, nicole.trahan@Colorado.EDU, jeffrey.beauregard@Colorado.EDU, maggie.prater@Colorado.EDU, marissa.hamed@Colorado.EDU, amy.trowbridge@Colorado.EDU, michael.wilkinson@Colorado.EDU, erin.brown@Colorado.EDU, stanfield.lee@Colorado.EDU, sierra.lovestowell@Colorado.EDU cc: sean.burns@Colorado.EDU, jsun@ucar.edu, sacks@ucar.edu, bowling@biology.utah.edu, lai@sciences.sdsu.edu, blamb@wsu.edu, oncley@ucar.edu, deander@usgs.gov, William.Bowman@Colorado.EDU, todda@Colorado.EDU, markw@culter.Colorado.EDU, Kurt.Chowanski@Colorado.EDU, thedudescientist@hotmail.com, bodenta@ornl.gov, turnip@ucar.edu, schaeffer@biology.utah.edu, zobitz@augsburg.edu, John.B.Miller@noaa.gov, faithann@ntsg.umt.edu, andrew.richardson@unh.edu, hsu@ucar.edu, stephens@ucar.edu, molotch@seas.ucla.edu, musselma@seas.ucla.edu, dewekker@virginia.edu, gochis@rap.ucar.edu, blanken@Colorado.EDU, dmoore1@ucar.edu, losleben@email.arizona.edu, chuixiang.yi@qc.cuny.edu, jing@purdue.edu From: "Sean Burns" Subject: Soil Data at CU/Monson Ameriflux Tower... Hi, Since you have used (or may have an interest) in the data from the CU/Monson Ameriflux tower I am including you in this email about the soil moisture measurements made near our Tower. If you are not interested in soil moisture you can save some time and stop reading right here. Starting with the January 2006 data we are going to start using a cs107 and cs616 sensor for the soil temperature and moisture data in the "climate" data files available from the CU webpage, ie: http://urquell.colorado.edu/data_ameriflux/ This will cause a slight "step" change in the data as we transition to the new sensors---the reason for this change is that the cs616 sensor (installed in Oct, 2005) seems to be produce higher quality data than the cs615 probes we have used in the past. The cs107 temperature sensor (also installed in Fall, 2005) was calibrated by EOL/NCAR and (probably) provides a slightly more accurate measure of soil temperature than the REBS stp-1 platinum resistance thermometers used in the past. The cs107 is more of a "point" measurement while the STP-1 is an average over the length of the sensor (~10cm). Taking a closer look at the soil moisture data is partially due to Noah Molotch (UCLA) examining soil moisture data from our "soil profile pit" and noticing some strange things in these data...he was also interested in getting the data from all 8 of the soil moisture sensors located near the tower...so i've been looking in more detail at these data and have created ASCII files that include data from all the soil moisture sensors that we have (eight cs615s and three cs616s). When sensors were moved it's a bit complicated to present these data...so whenever this occurred (primarily in 2005) I've created separate columns that indicate when a sensor has been moved. More details about the data files are in the header of each data file. The data files can be downloaded from: http://urquell.colorado.edu/data_ameriflux/data_30min/ the "soil" data files are: -rw-r--r-- 1 sburns 6626650 Jan 17 11:54 soil_2005.dat -rw-r--r-- 1 sburns 4347828 Jan 17 11:54 soil_2006.dat -rw-r--r-- 1 sburns 4172113 Jan 17 11:55 soil_2007.dat In most cases i tried to indicate the orientation of the sensor and also whether it was located under a canopy or in a clearing. A resource that was very helpful in evaluating the soil moisture near the Monson tower an "EnviroSMART" Sentek Water Content Profile Probe that was installed by LTER at C1 (ie, Mark Williams/Kurt Chowanski/Mark Losleben/Lucas Z.) in the summer of 2004. This sensor is in a relatively open area just to the west of C1. For more details about this sensor and setup you can contact Kurt Chowanski (Kurt.Chowanski@Colorado.EDU). Anwyay, i will end this email with links to some plots that detail the conclusions we've come to about the soil moisture near our tower: Plot of soil moisture at C1 (from EnviroSMART sensor): http://urquell.colorado.edu/sean/plots/plot_c1_soil_moisture.html * these data are from the LTER/Mountain Climate Program data set...and i think they are pretty cool (and show the usefulness of this probe). Note, i tweaked the 30cm level since those data didn't seem to make a lot of sense. Other than that, these are the "raw" data from the probe---the folks at LTER (not sure who, but probably Kurt/Mark/Lucas) did a calibration of two levels from this sensor prior to installation...and found that the raw data were ~20-30% lower than they should be. * a few of the things i get out of these data---the moisture from the summer rains never gets below 1m (smaller storms typically get down to around 50cm). On two of the years (2005 & 2007) the water table rose up during snowmelt and saturated the lowest levels of the sensor. After looking at this plot a bit i think the reason it didn't happen in 2006 was that we had a very warm (and dry) March...you can see around day 60 (2005) there was some melting...this spread out the duration of the melt period for that year...and did not result in a long steady melt that seemed to occur the other two springs. Or maybe it was just a simple matter of less snowfall in winter 2005/6... A plot comparing soil moisture at C1 with data from out near the CU/Monson tower: http://urquell.colorado.edu/sean/plots/plot_cu_soil_moisture.html -- upper panels are the "soil pit" data -- lower panels are all eight of the cs615 sensors (note, in Oct 2005 these sensors were moved so you might notice a jump in the data around that time). * one important piece of information about the soil pit: it is under a tree canopy...this is (perhaps) part of the reason why the soil pit soil moisture are not as sensitive to the smaller rainstorms that occur at the site. * strange thing #1 about the soil pit data: the sensor at 35cm in our pit (ie, "soil3_hori_35cm_pit_cs615") seems to have values which are much too high...the sensor also seems to be drifting a lot. For example, in 2005 the dry-est values were less than 0.1 m3/m3, in 2006 they were 0.13 m3/m3, and in 2007 they were over 0.2 m3/m3?? I can't explain why these values are changing. * strange thing #2 about the soil pit data: why does the sensor at 5cm (ie, "hori_5cm_pit_cs616") only react to the major storms while the sensor at 15cm ("hori_15cm_pit_cs616") show more of a response to rainfall events. There is a chance that these sensors are "swapped" from the location i think they are in (we'll check this after the snow melts)...but if you look closely at the timing of the snowmelt it's typically "hori_5cm_pit_cs616" that starts showing the melt before "hori_15cm_pit_cs616" (which would mean they are correct). So this would indicate they ARE were i think they are...I'm not sure what's going on with this, but we will confirm locations in the spring. . . * upper panel of lower plots---this compares the "hori_5cm_ucb_cs616" data with the sentek sensor at C1. Both of these sensors are in somewhat open areas and they both appear to be capturing most of the same rain events. the magnitude of the change in soil moisture is different, but this could be a calibration issue and/or a real change due to different spatial locations, soil types, etc. This comparison is essentially why I believe it's better to start using the "hori_5cm_ucb_cs616" data in our climate data files. * the data from eight different cs615 sensors are in the lower two panels of lower plots. These are the original set of sensors that were installed at the site...i added a comment about the data from each sensor to the legend of these plots. The cs615 sensors have a tendency to be sensitive to temperature (see Walker, et al 2004). I tried using the Temperature-correction described in the cs615 manual and it did not come close to removing the temperature dependency...these data could be used for daily qualitative looks at soil moisture at the site. But probably not much more. Here is a closeup look at these data: http://urquell.colorado.edu/sean/plots/plot_soil_moisture_cu_2007.html some of the cs615 sensors have high-freq noise in addition to the temperature-contamination issue. Ref: ---- Walker, J.P, et al. 2004: "In situ measurement of soil moisture: a comparison of techniques". Journal of Hydrology, v293, pp85-99. anyway, this is a rather long-winded email...and some things probably didn't get described properly. My final recommendation would be to replace the flaky cs615 sensors with cs616 sensors (or a sentek sensor if we have extra $)...and do a proper calibration of each sensor (as described in the cs616 manual) before deploying them in the field. Possibly some other types/brands of sensors (see Walker) might give good/better results (I would be interested to hear of any experiences about this). Probably one interesting thing we could do in the forest is a comparison of the open areas to ones under a canopy...and perhaps one could estimate how much water is intercepted by the canopy using such data. If anyone has questions or would like to discuss this in more detail please let me know... thanks, SpB. ------------------------------------------------------------------------------ Sean Burns Department of Ecology and Evolutionary Biology (EEB) Campus Box 334 University of Colorado Boulder, CO 80309-0334 for FEDex: Ramaley Building, Room N122 Internet: sean.burns@colorado.edu http://www.mmm.ucar.edu/people/burns/ Phone: (303) 492-5796 Fax: (303) 492-8699 ------------------------------------------------------------------------------