Polar Weather Prediction David H. Bromwich Session V YOPP Modelling Component Tuesday 14 July 2015 A special thanks to the following contributors: Kevin W. Manning, Jordan G. Powers, Keith M. Hines, Dan Lubin, Karen Pon, Jonathan Wille, and Aaron B. Wilson
The Antarctic Mesoscale Prediction System Provides customized NWP support for United States Antarctic Program forecasters Forecast model is the Weather Research and Forecasting Model (WRF ARW), optimized for the Antarctic environment (Polar WRF) Funded by the National Science Foundation Collaboration between National Center for Atmospheric Research/Mesoscale and Microscale Meteorology Laboratory and the Ohio State University/Byrd Polar and Climate Research Center Primary goals are to support USAP forecasters and their needs, with secondary aims to support research and education efforts in Antarctic meteorology Real time forecasts running since October 2000, through many updates Real time products disseminated primarily through the AMPS web page (http://www2.mmm.ucar.edu/rt/amps/) and the Antarctic IDD network AMPS archive recent years available through Earth System Grid Limited support for special projects: South Georgia Island Wave Experiment (SG WEX), Antarctic Cloud Microphysics Campaign, 2ODIAC, ORCAS, AVOCET
30 km 10 km 3.3 km 1.1 km 3.3 km
Special Project Support ORCAS The O 2 /N 2 Ratio and CO 2 Airborne Southern Ocean Study (Jan Feb 2016) Expand Domain 1 of our New Zealand/Palmer run Extra output for trajectory calculations Extra archives AMOMFW 2015 4
Ongoing Testing and Development AMPS Ensemble forecasts Still experimental: Relatively low resolution (10 km grid), with few ensemble members [O(15)] Useful for assessing predictability of a forecast situation AIRS assimilation Overall improvement in Cold Start with AIRS Hybrid Ensemble/Variational data assimilation Flow dependent Background Error Covariances from Ensembles Hybrid technique weights the forecast error between the variational and ensemble methods.
Example Ensemble Output: Strong Wind Frequency and Maximum Wind
Arctic NWP Example: Arctic Radiation Icebridge Sea & Ice Experiment (ARISE) August October 2014 Polar WRF 3.5.1 Real Time Forecasts (00 and 12 UTC) 280 x 310 x 53 Grid (8 km resolution) NCEP Water/Ocean/Sea GFS for initial Ice conditions and lateral boundary conditions NISE sea ice concentration; NCEP SST LANDUSE Grassland Shrubland Ice A Sample of Objectives: Forest Tundra Measure spectral and broadband radiative flux, quantify surface characteristics, cloud properties, and other atmospheric state parameters under a variety of Arctic conditions (including open water, sea ice, and land ice), and coinciding with satellite overpasses when possible. Acquire detailed measurements of land and sea ice characteristics to help bridge a gap in NASA satellite observations of changing Arctic ice conditions.
Issues with the Cloud Microphysics Over open water more cloud water Over sea ice less cloud water Polar WRF modeled clouds during the ARISE field campaign in the Beaufort Sea are mostly liquid water in agreement with ARISE observations. Simulated cloud water magnitudes have not yet been verified against ARISE observations, but may suffer from the same excess simulated for ASCOS. ASCOS over Sea Ice August 2008 During ASCOS modeled cloud water (green) at North Pole is much larger than observed (red) indicating issues with the Morrison microphysics scheme.
Cloud Physics is a Prydz Bay Low Bi Polar Problem 22 Jan 2013 AMPS cloud base product MODIS Terra satellite visible image Cloud east of AIS (A) identified by model. Cloud over AIS (B) not identified by cloud algorithm but low level moisture present. Insufficient model moisture for cloud west of AIS (C). AMPS low level RH product
ARM West Antarctic Radiation Experiment A Joint NSF DOE ARM Mobile Facility Campaign Objectives 1. Improve understanding of mechanisms governing West Antarctic energy balance and climate change Influence of subtropical and tropical teleconnections Influence of local cloud radiative forcing and feedbacks 2. Assessment and improvement of cloud physical parameterization in climate model simulations for the coldest climate regime What factors govern cloud physics in a very cold and very pristine environment year around? Deployment Plan Figure adapted from October 2015 September 2016 Nicolas and Bromwich (2011) AMF2 at McMurdo Station ( Central Facility ) Detailed cloud and aerosol observations with the most advanced atmospheric science equipment available today. October 2015 15 January 2015 (Summer) West Antarctic Ice Sheet (WAIS) Divide ( Extended Facility ) Observations of cloud, upper air and surface energy budget
Boundary Layer Physics in the Polar Regions 5 km 30m: Temp, RH, Wind Speed, Wind Direction, Net Longwave and Shortwave Radiation 15m: Temp, Wind Speed, Wind Direction 7.5m: Temp, RH, Wind Speed, Wind Direction 4m: Temp, Wind Speed, Wind Direction 2m: Temp, Wind Speed 1m: Temp, Wind Speed 1.67 km Alexander Tall Tower!
Positive Histogram Critical transition range is 4 8 m s 1 AMPS overestimates the strength of the inversion for stronger wind speeds AMPS Inv > Tower Inv
Negative Histogram Critical transition range is 4 8 m s 1 AMPS underestimates the strength of the inversion for weaker wind speeds AMPS Inv < Tower Inv
Recommendations for YOPP Possible AMPS contributions to YOPP Assimilation of extra YOPP observations (obs put on GTS) AMPS analyses and forecasts archived at NCAR and available for scientific investigations via web Forecast plots for YOPP field campaigns as resources allow NWP Physics Airborne campaigns needed to collect additional data for the improvement of cloud microphysics in the Arctic and Antarctic e.g., manned aircraft, UAVs Coordinated with ground based observations (ships, IASAO observatories, etc.) Need more observations like Alexander Tall Tower! (and taller!) to capture PBL characteristics in the Polar Regions array of towers