B. Snowfall and Freezing Precipitation
[Background]
[Short-term forecasting] [ASOS drizzle
detection algorithm]
[Snow Gauge and Windshield Testing] [Hotplate
Snowgauge]
RAP has a successful history of involvement with airport and aircraft operations dealing with the impact of snow and freezing precipitation. The operation of aircraft during snow and freezing rain or freezing drizzle conditions is a significant safety issue due to the rapid loss of lift and increase in drag produced by ice on an aircraft. For example, a rough ice coating of only 0.8 mm on a plane’s wing can result in a 25% loss of lift and increase in drag. Snow and freezing rain accumulations on taxiways and runways also impact the safety and efficiency of ground operations.
The main goal of this research is to improve nowcasts and forecasts of snowfall and freezing precipitation. The work is being done in the context of a system called the Weather Support to Deicing Decision Making (WSDDM) that has been developed over the past several years under FAA sponsorship. The WSDDM system is now operational in the New York metropolitan area, with observing systems located at all three major airports. The research emphasis during FY 2001 has been on evaluating the existing WSDDM nowcasts and in developing better nowcasts and forecasts for this system using numerical modeling and Doppler radar. RAP also continued to collaborate with scientists at the Desert Research Institute, Reno, Nevada on the development and evaluation of instruments designed to more accurately measure snowfall rates in real time.
2. Short-term forecasting of snowbands using numerical models and Doppler Radar
The feasibility of using the MM5 four-dimensional variational data assimilation (4DVAR) system to assimilate Level II radar data into a mesoscale model was investigated by M. Xu, A. Crook and R. Rasmussen. Observing System Simulation Experiments (OSSE) as well as real data experiments were conducted for a snowstorm event.
Results from the OSSEs show that, using 30-minute observations of u, v and reflectivity, the recovery of the wind and rain/snow water mixing ratio is reasonably accurate. On the other hand, 60-70% errors in the unobserved fields (e.g. temperature and water vapor) remain. With partial success in the retrieval, the derived fields are able to improve the forecast of snowbands, especially in the first 3 hours. Assimilating radial velocities instead of u and v degrades the retrieval and forecast, but the assimilation still has a positive impact on the forecast.
Real data experiments using WSR-88D observations show that assimilating radar reflectivity alone makes definite, though limited, improvement on the subsequent forecasting. Further experiments are needed in order to effectively assimilate radial velocity data using MM5-4DVAR.
3. ASOS drizzle detection algorithm
During the past two years C. Wade has been working on an algorithm to detect drizzle on NWS Automated Surface Observing System (ASOS) stations. Drizzle is not currently reported on ASOS stations unless an observer augments the observation. The differentiation between rain, drizzle and snow is important in the detection and forecasting of in-flight icing conditions. When temperatures are near freezing, drizzle or freezing drizzle at the surface suggests the presence of super-cooled, large droplets aloft.
The drizzle algorithm that is being proposed to the NWS is based on the raw one-minute data collected by the ASOS LEDWI sensor. The Particle channel on LEDWI gives an indication of the size of the largest particle to fall through the 50-mm diameter LEDWI beam during a given minute. When Particle channel data are viewed against the values in the LEDWI’s Low channel, a pattern emerges that suggests that a functional relationship exists between the Low and Particle channels that can be used to identify when drizzle is occurring. To eliminate false reports of drizzle that can occur as a result of atmospheric turbulence, the value of this function is used in combination with data from other ASOS sensors.
An analysis of precipitation type reported at more than 200 U.S. surface stations from 1961-1990 has shown that drizzle is well correlated with low ceilings and near-saturated conditions. Therefore, a requirement of the algorithm is that ceilings be overcast below 2000 feet and that the temperature-dew point spread be less than or equal to 4 degrees (F). If these conditions are met, and if the functional relationship described above indicates drizzle, then the precipitation type is likely drizzle. Drizzle intensity (light, moderate or heavy) is determined from the magnitude of the values in LEDWI’s High-frequency channel.
The NWS is currently evaluating the algorithm at its Test and Evaluation site near Sterling, VA. If the algorithm proves successful as an indicator of drizzle, it can be used in combination with the output from the ASOS icing sensor to distinguish between freezing rain and freezing drizzle.
4. Snow Gauge and Windshield Testing for the Climate Reference Network
The Climate Reference Network (CRN) is a program designed to improve the quality of instrumentation used to make observations of wind, temperature and precipitation used in climate studies. The accurate measurement of winter (frozen) precipitation has long been a problem due to the difficulty of measuring the liquid equivalent of snow. Precipitation gauges designed to measure rain often under measure snowfall. This is due to the fact that the gauge acts as an obstacle to the airflow and that the slower falling snowflakes tend follow the airflow around the gauge rather than going into the gauge’s orifice. The single Alter wind shield used on many gauges improves the catch somewhat, but may still result in significant under measurement (50% or more) during windy events.
In an effort to improve the accuracy of these winter precipitation measurements, the CRN contracted with RAP during the past two years to evaluate current off-the-shelf snow gauge technology in combination with various wind shields. The tests were conducted at NCAR’s Marshall facility, located 5 miles southeast of Boulder. RAP engineer J. Cole is manager of the Winter Precipitation Facility at Marshall and installed 7 GEONOR snow gauges in 7 different wind shields for the CRN studies. In addition, Jeff developed and tested a controlled heating mechanism for the gauges so that snow would not stick to the gauges and block their openings. The mechanism uses a temperature sensor mounted on the inside surface of the gauge’s orifice and heat tape wrapped around the outside surface of the orifice. The temperature is constantly monitored. When the temperature falls below +1C, the heat turns on and warms the orifice. When the temperature rises above +2C, the heat turns off. Thus, the temperature of the orifice is maintained just above freezing.
The heating tests were very successful, insuring that snow was melted the instant it struck the gauge’s orifice. Of the 7 wind shields tested, the shield that resulted in the greatest catch was the Double-Fence Intercomparison Reference (DFIR) shield developed in Russia in the 1960’s. The DFIR shield was used extensively in WMO (World Meteorological Organization) tests conducted around the world during the 1980’s and 1990’s. The DFIR shield consists of two concentric rings of fencing at 12m and 4m in diameter, with the snow gauge at the center. The snow gauge typically has its own wind shield around it at a diameter of about 1.5 m. Thus, the GEONOR snow gauge in the DFIR was a triple-shielded gauge. Unfortunately, the size of the DFIR shield may be too large for some locations. Figure B1 shows a photograph of the DFIR shield at the Marshall facility.
Figure B1. Photograph of the DFIR wind shield at NCAR's Marshall facility. At the center of teh DFIR is a GEONOR snow gauge in an Alter shield. Other wind shields and instrumentation at the winter test facility are shown in the background.
A hotplate snowgauge has been jointly developed by NCAR (R. Rasmussen, J. Cole) and Desert Research Institute (J. Hallett, R. Purcell) that provides a method to measure liquid equivalent snowfall rate every minute. One of the main motivations for this work is the need for improved methods to measure liquid equivalent snowfall rates in support of aircraft deicing operations at airports. The hotplate snowgauge does not require glycol or oil, and is expected to be relatively inexpensive. The principle of operation is to measure the amount of heat necessary to melt and evaporate all the snow or rain striking the top surface of the hotplate. The system has an upper and lower plate heated to nearly identical constant temperatures. Currently, the top plate is heated to 75 °C, and the lower plate to a slightly cooler temperature. The plates are maintained at constant temperature during wind and precipitation conditions by increasing or decreasing the current to the plate heaters. During normal windy conditions without precipitation, the plates cool nearly identically due to their identical size and shape. During precipitation conditions, the top plate cools due to the melting and evaporation of precipitation while the bottom plate is only effected by the wind.
The difference between the power required to cool the top plate compared to the bottom plate is proportional to the precipitation rate. The initial design of the plates had a smooth upper and lower surface. During previous testing at Marshall it was determined that snow would “skate” off the upper surface during high wind conditions and underestimate the snowfall rate during these periods. In order to overcome this problem, three concentric walls were added to both the top and bottom plates. These concentric walls help prevent snow or rain impacting the plate at an angle from sliding off during high wind conditions. Tests of this new design showed good comparison to the liquid equivalent measured from co-located GEONOR snowgauges. However, it was also determined that radiational heating of the top plate could lead to false indications of precipitation. Research has shown that a net radiometer or on/off sensor can be used to correct this problem.
The main research performed this year was the development of an algorithm to correct for the undercatch of the hotplate during high wind conditions.
The photo of the original (top portion of the photo) and coil hotplate (lower hotplate) is shown belowin Figures B2 and 3:
Figure B2. During the fall of 2001 a new frisbee® hotplate was installed. A photograph of this hotplate is shown below in Figure B3.
