Feedback to
"Review of the ARM SGP CART Site Measurement August 10, 1998 Feedback has been received concerning the SGP measurements review document issued on June 30, 1998. It is contained below. From Marv Wesely, Instrument Team Leader: CLOUD WORKING GROUP Cloud base observations: A plan has been developed to install a MPL-HR at the central facility. Installation of Vaisala ceilometers is planned in the near future for the boundary facilities, but none at the central facility. MMCR: Many of the issues raised involve research. No schedule of which I am aware has been set for upgrading the processing computer; an estimate of the costs needs to be made. Changing the fundamental product of the MMCR to include the Doppler spectrum was not previously advocated, to my knowledge. The benefits of doing so are not clear, particularly since the "bug" problem has not been devised; Jay Mace was leading the effort in this regard. I had not heard the concern expressed about the effects of the radome on the MMCR observations. All of these issues need to be put on the list of activities for the MMCR. At the current time, standardization of the modes of operation and calibration issues represent the higher priority issues. Whole sky imager: One-minute data products are currently envisioned only for IOPs. If the one-minute data products are needed routinely, the increase of data volume is large. Is it "nice" or really "necessary?" Cloud observations away from the central facility: Developing HSIs (hemispheric sky imagers) for routine operation could be expensive. Even if they were installed at the boundary facilities, the coverage would be far short of CART-wide coverage. The approach of using satellite data together with WSI observations at the central facility needs to be considered more seriously. Additional input from Martin Platt: The situation with the MPL is not particularly encouraging. A plan for conversion to high resolution systems has been developed. Addition of dual-polarization capabilities is not likely in the current round of activities but might be considered later when the situation stabilizes. To satisfy the need for a zenith-pointing fast visible radiometer, the NFOV radiometer is planned for installation in August 1998 [now installed]. SHORTWAVE RADIATION WORKING GROUP Temperature: Currently, radiosondes are the only routine capability for measurement of temperature in clouds. A commercial passive microwave system for temperature profiles might be considered, although its vertical resolution is fairly course. Broadband radiometry: Acceleration of the pace of GRAMS implementation represents a problem that apparently needs more resources. UV spectrometry: The need for a broadband UV radiometer that measures the total energy in wavelengths between 0.3 to 0.4 micrometers can be met with a commercial device, according to Joe Michalsky. ARM/CART will investigate acquisition of this device. CSPHOT: Something could perhaps be done to increase the angular scans of the CSPHOT. This would have to be negotiated with NASA. Column Vapor and Liquid: These issues probably need to be addressed by Jim Liljegren. AEROSOL WORKING GROUP Measurement of aerosol chemical composition and of CCN spectrum has been considered more than once during past ARM discussions, and some relevant Science Team proposals have been submitted but not funded. The ARM Science Team community would need to put these activities at a higher priority than implied in the past, if such measurement capabilities were to be developed. SURFACE FLUX WORKING GROUP The current operation of the ECORs is hampered by the lack of a data ingest and the resulting long lag time between problems appearing and their identification by the instrument mentor. The priority of the development of ECOR data ingest routines is high, but progress is limited by unavailability of DSIT personnel until later in 1998. I support the need for NDVI measurements. My experience with the hand-held devices is that they require fairly skilled operation. Simplified approaches for routine observation could conceivably be developed, but requires an effort for which resources would need to be allocated. SINGLE-COLUMN MODEL WORKING GROUP Surface flux over cropland: I agree. See discussion above. From Mike Splitt, SGP Site Scientist Team: I read the "Review of the ARM SGP CART Site Measurement Capability Relative to the Needs of ARM's Scientific Working Groups" and have the following comments for your consideration as I think they apply mainly to the Cloud and SCM Working Groups for the near term. Investigation of use of WSR-88D for cloud and cloud water definition for the ARM SGP Site is in its infancy. I have in the past ordered WSR-88D data though OU/CIMMS and have conducted VAD analysis of the data to provide vertical profiles of wind speed and divergence for some select periods. Shortly, I will be receiving a data set for the Summer 1997 SCM IOP which includes many radars from within the SGP CART area. Mark Miller, Steven Lazarus, and myself are coordinating efforts to begin integrating radar data into cloud analyses for the SGP Site and will be using this data set for a trial run. Mark Miller (and others) have done extensive work comparing the WSR-88D data to a MMCR, and Steve Lazarus has begun integrating WSR-88D data into the ARPS analysis system (ADAS) over the Salt Lake City area (the analysis incorporates many inputs including radar and satellite data). Thus, we are in a position to start work on creating test products for ARM use. We hope to get some results for presentation at the upcoming SCM Workshop this fall. While results from this work on obviously not yet available. I think it might be sensible to include mention of the WSR-88D data as a potential data source need. Considering the problems with cloud specification mentioned by Jay Mace and others, I think the radar data can certainly go a long way to helping define the site-wide three-dimensional cloud structure. From Tony Del Genio, SGP Site Advisory Committee: I quickly read the data assessment document on your Web site. The one thing I'd like to emphasize is cloud optical depth. It is briefly mentioned in a couple of places within the document, but not assigned particular importance. It is the parameter that links cloud properties to radiation. As was mentioned in the document, cloud optical depth can be retrieved from the MFRSR, and there are candidate algorithms to do so. This should be given a high priority. From Don Slater, Shortwave Working Group: I have read over most of your CART-needs document; let me say that it is nice to see a composite of so many people's idea's and must be very useful for defining future CART directions. Specifically, in regards to the shortwave, I have to agree with what Warren has supplied you. I would like to stress (Warren, correct me if I'm wrong on any of this!) several thoughts: I am not sure what is the complicating factor in getting the GRAMS to produce operational values (w/m2 or w/m2/nm), but it is very important that it start producing numbers in the near future. These numbers need to be carefully compared to other reliable instruments to give an idea of difference residuals, and under what conditions these residuals vary. Ells Dutton has begun such a project, but I encourage it to become an ARM priority. Warren has mentioned the NFOV, but I think it is important to stress the need for cloud optical depth. Hopefully the NFOV will provide something along these lines. As Joe Michalsky says: "Calibrations, Calibrations, Calibrations!", and he is correct. The broadband instruments are in pretty good shape as far as I know (although the EF's may be on their own a little too much to truly trust everything), but we need to come as close as possible to having a common calibrator for the spectrometers. Joe has shown the arrogance of calibrations if taken without question, so it might be easier to at least have a consistent error among spectrometers (one calibration source) rather than a collection of independent errors. Warren has also commented about the RSS and the UV RSS not having sufficient signal-to-noise below about .4 micron due to silicon detectors; I don't know what the UV RSS uses, but a back-thinned, AR coated silicon detector can reach down quite a ways to the 0.3 micron. This is not true for a front illuminated system. One question regarding the WSI - I have worked with whole sky imagers in the past (specifically for very low light level nighttime work) and hope that the occurrence of light scattering within the optical path has not been forgotten. This is not an unimportant nuisance; with bright clouds and the solar disc within the image, even internal reflections of a percent or two may contribute significantly to overall hazing of the image. I say this with the future all-sky radiances product in mind. I think that spectral radiances over the entire hemisphere vs. modeled radiances may hold real clues to the now famous SGP Bermuda Triangle for photons. After all, shouldn't this give us information regarding aerosols? To do this, however, there may be a need to operate with additional filters in the WSI spread throughout the visible and near-IR. After seeing some of Pilewskie's spectral fluxes and their apparent dependence on surface spectral albedos, maybe the albedo of the every-changing SGP surroundings needs to be better assessed (a first go could be the BDRF IOP). Additional Comment from Warren Wiscombe, SGP Site Advisory Committee and STEC Member: I ran out of steam on my "SW input needs" white paper, and never finished it. However, I agree with what Don Slater has to say, especially about the GRAMS. The comparisons of GRAMS vs. Ells Dutton's best radiometers last summer, under two weeks of clear and clouds conditions, revealed offsets or maybe 10 W/m^2 and differences up to 30 W/m^2, as shown by the histograms we posterized at the Gordon Conference. We need a database like this from SGP that we can trust, for a much longer period, and soon. One thing I would add: we need to start routine operational ozone column measurements at the SGP. The Microtops was used during the big IOP last fall. I have provided Ted Cress with suggestions on how to take measurements with it (basically, 3 times a day, at noon and two hours each side of noon, but with flexibility to move the observation if the Sun is obscured by cloud). I, too, would like to see optical depth, but I don't want to get it from shortwave observations -- that's a vicious circle. We need to get it from independent parts of the spectrum, e.g. microwave for liquid water path plus radar from mean effective radius. One final thing I would say about SW IOP planning: I now feel we should schedule regular SW IOPs just as the SCM group does. It seems to draw a line in the sand, without which things proceed at a much slower pace. Sometimes just scheduling the SW IOP has the greatest impact on getting things finished! I would therefore like to schedule at least one a year forevermore. We will move the season around to sample different conditions. We are due next for a spring and winter outing. My inclination is to go for a spring 1999 outing, then a winter 2000 outing. That gives us some really different clouds than we have seen up till now. |