Transcript Temp452
Temperature Prediction ASOS Temperature/Humidity Senor Why Can’t Use Model Output Directly for Temp Forecasts? • Model surface/2m height may be very different than real elevation due to limited horizontal/vertical resolution. (e.g., MM5 36 km surface elevation for Boeing Field is 256 m, should be 5 m) • Model resolution may be inadequate to properly simulate the temperature effects of important features such as: – Narrow gaps, such as the Fraser River Valley or the Columbia River Gorge. – Land/water contrasts, such as around Puget Sound or along the coast. Near Sea Level Gap On Border of WA and OR Portland wind distribution for all days by direction and speed • Wintry precipitation at PDX – Almost exclusively E or SE – Tendency towards E for snow and SE for freezing rain Wind distribution for days with snowfall Wind distribution for days with Freezing Rain December 11-15, 2000 Case Study • Strong winds for nearly four days – Gale force a times • Wintry mix in Western Gorge and exit area. – Little snow accumulation but significant icing. • Selected because: – Data availability was good (esp. ACARS) – Model Initialization ok Domain Definition 36 km grid spacing 12 km grid spacing Portland The Dalles Portland The Dalles Pass Height = 700 m Pass Height = 600 m 4 km grid spacing Portland Cascade Locks The Dalles Pass Height = 400 m 12 km grid spacing Portland The Dalles Pass Height = 600 m 1.33 km grid spacing, Pass Height = 150 m • Portland Cascade Locks Troutdale The Dalles 444.4 m grid spacing, Pass Height = 100 m Portland Troutdale Cascade Locks T on 150 m Surface Portland Troutdale • Model temperatures may be seriously in error due to poor model physical parameterizations, such as for the planetary boundary layer, radiation, surface energy fluxes, cloud and precipitation processes. – Example: overmixing in PBL can result in the inability to maintain shallow cold air masses. – Example 2: improper soil moisture (too warm or dry) can greatly influence temperatures. • As a result of such errors, models can have serious systemic temperature biases. Shallow Fog…Nov 19, 2005 • Held in at low levels for days • MM5 held in the inversion…generally without the shallow mixed layer of cold air a few hundred m deep • MM5 could not maintain the moisture at low levels Temperature Biases • Large temperature biases can occur at certain times: – When there is a shallow layer of cold air (few hundred m deep) that is mixed out. – During transition season (particularly spring, when land surface conditions are problematic) – During summer during warm periods. 00 UTC 12 UTC Downslope Warming • Large warming during downslope flow • Often large over Cascade foothills (North Bend), but apparent all over the world, including to the lee (east) of the Rockies--the Chinook Wind. • In Europe called the Foehn Wind. • Uusually, air comes from mid-levels where potential temperature is higher than at the surface. A drop in dewpoint usually accompanies it. Diabatic Effects Diabatics • Radiation: – direct radiational heating and cooling of the air is relatively small – indirectly, very large through modulation of ground temperature and communication into the boundary layer by turbulence. Surface heating by the solar flux and nighttime cooling by IR flux dominate surface temperatures. – Clouds have a major effect on radiational heating/cooling. Clouds, Radiation and Surface Temps • Clouds lessen warming during day by lessening solar radiation • Clouds decrease cooling at night by intercepting IR radiation and reradiating IRback to the surface. • Thin cirrus…only minor influence (1-2F) • Thick altostratus--5-10F influence or more. IR Opacity of Air Can Influence Surface Temps • IR opacity is proportional to humidity. • Increased water vapor content increases opacity--better absorption and emission in IR. • Less water vapor results in better IR cooling under clear conditions. That is why deserts cool very rapidly at night. Why Washington DC stays hot all night. Conduction and Turbulence Effects in the BL. • Snohomish, WA • April 4, 2005 SE Everett Hoquiam, WA: July 15, 2002 18 16 70 Wind Speed (kts) Temperature (ЎF) 65 12 60 10 55 8 6 50 4 45 2 0 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Hour (LDT) Temperature (ЎF) Wind Speed (knots) 14 Comparison of the daily temperature cycle between observations (blue dashed) and the model (red triangles) for a coastal location (Boston, MA) on 10 August 1997 impacted by the passage of a sea breeze front. 2-m air temperature and 10-m wind velocity in the afternoon of a sea breeze event on April 18, 2005. The land-sea temperature contrast drives the onshore sea breeze. Local spatial variations in sea surface temperature (SST) can leave a signature on the sea breeze, especially when complex coastlines create multiple sea breeze fronts (e.g., the sea breeze penetrating from Long Island Sound in the image above). In our multiply-nested COAMPS simulations, nest 4 (1.33 km resolution) and nest 5 (0.44 km resolution) were enhanced to account for urban effects and to include time-varying (hourly) high-resolution SST's from Alan Blumberg's New York Harbor Observing and Prediction system (NYHOPS). The diurnal heating of the ocean surface is thus represented in our simulations. We are in the process of quantifying the impact of these effects on the sea breeze circulation for several distinct events, and evaluating the forecast skill of the system. Southerly Buster…SE Australia…just like the push MOS Compared to the NWS: Temperature Cool Season Mi. Temp – 12 UTC Cycle Average Over 80 US stations 7 6 5 4 3 2 '66-67 '71-72 '76-77 '81-82 '86-87 '91-92 '96-97 '01-02 Cool Season 24-h GUID. 24-h LCL 48-h GUID. 48-h LCL Temperature MAE (F) for the seven forecast types for all stations, all time periods, 1 August 2003 – 1 August 2004. Large one-day temp changes MAE for each forecast type during periods of large temperature change (10F over 24-hr), 1 August 2003 – 1 August 2004. Includes data for all stations. MAE for each forecast type during periods of large departure (20F) from daily climatological values, 1 August 2003 – 1 August 2004. Number of days each forecast is the most accurate, all stations. In (a), tie situations are counted only when the most accurate temperatures are exactly equivalent. In (b), tie situations are cases when the most accurate temperatures are within 2F of each other. Looser Tie Definition Number of days each forecast is the least accurate, all stations. In (a), tie situations are counted only when the least accurate temperatures are exactly equivalent. In (b), tie situations are cases when the least accurate temperatures are within 2F of each other. Looser Tie Definition Highly correlated time series Time series of MAE of MAX-T for period one for all stations, 1 August 2003 – 1 August 2004. The mean temperature over all stations is shown with a dotted line. 3-day smoothing is performed on the data. Cold spell Time series of bias in MAX-T for period one for all stations, 1 August 2003 – 1 August 2004. Mean temperature over all stations is shown with a dotted line. 3-day smoothing is performed on the data. MAE for all stations, 1 August 2003 – 1 August 2004, sorted by geographic region. MOS Seems to have the most problems at high elevation stations. Bias for all stations, 1 August 2003 – 1 August 2004, sorted by geographic region. Precipitation Comparisons Brier Scores for Precipitation for all stations for the entire study period. Brier Score for all stations, 1 August 2003 – 1 August 2004. 3-day smoothing is performed on the data. Prob. Of Precip.– Cool Season (0000/1200 UTC Cycles Combined) 0.7 Brier Score Improvement over Climate 0.6 Guid POPS 24 hr Local POPS 24 hr Guid POPS 48 hr Local POPS 48 hr 0.5 0.4 0.3 0.2 0.1 0 1966 1969 1972 1975 1978 1981 1984 Year 1987 1990 1993 1996 1999 2002