The topic of the 7^th International Conference on Precipitation focused on "observation, estimation, and prediction of precipitation variability at all scales." The conference was sponsored by NSF, NASA and NOAA and was organized by Anna Barros of Harvard and Elfatih Eltahir of MIT.

The conference was organized with days devoted to specific themes. Saturday was Registration and Reception; Sunday was Measurements of Precipitation; Monday was Modeling and Physical Characterization of Precipitation Processes; Tuesday was Characterization of Statistical Properties of Precipitation Fields. Each session consisted of 3 extended Oral Presentations followed by a short break and a large number (about 20) of short oral introductions to the posters after which there was a poster session of 1-11/2 hours. There were two sessions each on Sunday and Monday. The complete program is attached.

The following are my specific, biased personal thoughts on the conference:

• Precipitation Measurements: The NASA Global Precipitation Mission has risen to high priority in NASA. Scheduled Launch is around 2007. It represents an important future research opportunity. Efforts are underway to raise TRMM to a higher orbit to keep it going till GPM launch. There is a lot of work going on at present on integrated satellite radar, radiometer and lightning measurements and comparisons with radar. Work continues on optimized Z-R relations and including vertical profile information, etc. Given the climate and GEWEX interests a fair amount of work related to extended time periods and to larger scales, but most of this was based on observations made at smaller space and time scales. There were at least as many papers which specifically addressed "event" scales. The paper by Yuter built heavily upon Paul Smith's work.. A very interesting series of papers by Holt et al from England described long baseline (10-30km) dual wavelength microwave technique for measuring path-averaged precipitation. Amitai et al also discussed issues of rainrate distributions. Carbone et al presented Hovmoeller diagrams of summertime precipitation episodes over a region from the Continental Divide eastward to the Atlantic and from the Southern Plains to near the Canadian border. Their results showed that most precipitation occurs in progressive systems moving across the country and with a very large diurnal pattern in the west with a displaced, weaker pattern further east. Viginal et al used data from Grand Forks in there radar profile work. Shepherd found urban rainfall anomalies downwind of several cities based on TRMM data (referencing Hjelmfelt, 1982 and other METROMEX work. Incidentally, I met Sandrine Anquentin, U of Grenoble, who has developed a surface parameterization usable in ARPS for Urban areas.

• Modeling and Physical Characterization of Precipitation Processes:

Rodriguez-Iturbe discussed a model study of the role of interannual rainfall fluctuations on ecosystem dynamics (grasses vs mesquite). Lau presented interesting results on Asian monsoon teleconnections for precipitation in U.S. (Tokio-Chicago, wet years; Shanghai-Kansas, dry years). Soden discussed possible evidence of increase in the rate of hydrologic cycling in El Nino years (some discussion elsewhere about possible acceleration for the increased CO2 climate). Arritt et al presented results from the project to intercompare regional climate models which showed that consensus forecasts from different models gave the best forecasts. Findell et al looked at how soil moisture acts to affect convection later in the day. Pal et al looked at impact of regional soil moisture anomalies on storm tracks. The NWS is moving to make the AMBER flash flood warning system available and adaptable to different NWS offices, this will take some time, but the effort is underway. Schneider presented an interesting poster on evaluation of seasonal forecasts from a user perspective. Antquetin presented results of observations and simulations of rainbands over complex topography in Southern France. This work looked a lot like some of the things we've been doing in the Black Hills. Peterson, Rutledge et al gave one of several posters combining satellite lightning data and other information to predict precipitation. Gage showed evidence that disdrometer calibrated profilers could produce very accurate, stable rainfall estimates.

• Characterization of Statistical Properties:

Lebel presented analysis from the Sahel, 90% of rainfall from 12% of precipitation systems. Difference between wet and dry periods is reduced number, not intensity, of these storms. Lovejoy presented evidence that all measures of precipitation on all scales are fundamentally functions of scale and that we can't just average or "disaggregate" but that we need to include a factor based on some "scaling model". The implications and application of this concept need to be discussed--It was suggested that this be a topic of a full session for the next conference. I had to leave fbefore this session finished. However a quick glance at the posters showed that they really were all based on purely statistical approaches.

There were a number of other papers, but the above stood out.

The lightning-satellite-radar precipitation measurement papers showed that there were several well established groups working in the general areas we proposed for NASA-EPSCOR . This is current work in progress, not published work. We proposed to investigate different aspects of these issues, but perhaps not sufficiently dramatically different to bring in another, new effort.

Discussion of USWRP on Monday Night indicated a coming push to look at Flash Floods, but funding is still only $1 Mil. to NCAR and $1 Mil elsewhere (NOAA labs, universities etc.).

NSF has approached the Hydrology community to consider putting together something like UCAR and defining what they need to make progress (sounds like potential for significant base funding effort in the future if the Hydrologists can get there act together).

Coupled modeling is still a priority issue (but given the big-name collaborations developing, if you are not known you've got to show results).

A very interesting trend was the number of cross-country collaborations being developed in a number of areas. Clearly there is growing comfort with "internet collaboration" and a new attitude of "how can we best get this research done and who is the best person to work with --anywhere, rather than "on campus" or "in-state". This has been a theme of the Great Plains Network. I think this approach will increasingly be important in securing research funding in the future.

Back to "Trips and Collaboration" SD Personnel and NASA Centers / NASA-related activities website