The Mauna Kea Weather Center: the need for custom forecasts

1
The Mauna Kea Weather Centerthe need for custom
forecasts

• Steven Businger, Tiziana Cherubini and R. Lyman

2
The Mauna Kea Weather Centerthe need for custom
forecasts
1 - Forecasts of Observing Quality

• Telescope mirror temperature
• Telescope wind shake
• Precipitable water
• Seeing and Cn2

MKWC HPC System
3
2 - Weather Hazard Mitigation

• Tropical cyclones
• Cold frontal passages
• Upper level troughs/lows
• Strong subtropical highs (strong summit winds)
• Kona lows

4
Brief History of Weather Center

• Memorandum of understanding between UH
  Meteorology IfA established the Mauna Kea
  Weather Center in July 1998.
• Three principal objectives
• (i) Provide weather forecasts and nowcasts for
  MKO.
• (ii) Determine and forecast conditions that
  provide the best astronomical observing
  conditions.
• (iii) Communicate forecasts, meteorological data,
  and imagery to observatories.

5
Current Status

• MKO forecasts issued twice daily, Monday through
  Friday
• Twice Daily WRF runs
• Satellite and model data feeds from NWS Unidata
• Satellite and model graphics presented on web
  site
• Comprehensive data archive maintained
• New WRF products under development
• WRF case study research ongoing

6
Synergy with Broader Community
7
Synergy with Broader Community

• WRF is an NCAR- and community-supported research
  and forecast model.
• Data assimilation effort (LAPS) for WRF in
  collaboration with NOAA ESRL.
• Data sharing, web distribution, and data archive
  development in collaboration with Unidata and UH
  Meteorology.
• GPS IPW in collaboration with UH Geophysics and
  NOAA.
• Calibration and assimilation of lightning data in
  collaboration with ONR and NASA.
• COSMIC project refractivity data from
  limb-sounding technique
• Utilization of high performance computer
  resources available at NCAR and MHPCC.

8
mkwc.ifa.hawaii.edu

• Two Linux Servers provide
• Data ingest
• Data assimilation and WRF input
• Graphic/Web
• Redundant product distribution
• Archive function

9
Silicon Mechanics HPC

• The MKWC HPC system is comprised of 16 compute
  nodes, 128 CPUs (Intel Xeon L5420 Quad-Core
  2.50GHz), with high-speed communication links
  between nodes (Infiniband cards and switches).
  The system includes a RAID-6 storage component.

10
Key Variables in Twice-DailyMKWC Forecasts

• Cloud cover, fog, precipitation
• Summit winds and temperature
• Precipitable water
• Seeing, Cn2, and wind profiles

11
Research Challenge Seeing Forecasting

• Obtain fine vertical and horizontal resolution
  forecasts of temperature, wind profiles and
  turbulence related variables from WRF ? Cn2
  algorithm prediction
• Provide optical turbulence parameters by
  integrating the Cn2 profiles
• Validate and refine the optical turbulence
  algorithm

12
Learning from the daily forecast experience

• The MKWC has been issuing operational
  quantitative seeing forecasts since the summer of
  2006.
• A posteriori, the forecaster qualitatively/quantit
  atively validates his forecast, thus gaining
  experience.
• An inventory of the daily data received at the
  MKWC is under construction. This will allow
  observed weather conditions to be correlated to
  observed seeing conditions.

13
Learning from the daily forecast experience

• Quantitative validation

14
MKWC Goals

• Study sensitivities of Cn2/seeing forecasts.
• Tune model physics configurations to increase
  overall accuracy.
• Incorporate 3-D VAR data assimilation into model
  assess impacts.
• Investigate weather patterns leading to
  especially good or bad seeing.
• Comprehensive WRF forecast validation.

15
MKWC Goals

• Increase the skill of conventional and seeing
  forecasts with help of validation statistics.
• Provide forecast variables with finer temporal
  and spatial resolution.
• Issue longer-term seeing forecasts.

16
Book on SeeingThe measurement and impact of
Atmospheric Turbulence and Refractivity on the
Propagation of Extraterrestrial Radiation

• Introduction
• 1. Atmospheric Turbulence
  
  Authors
• 1.1 Atmospheric turbulence from the perspective
  of a meteorologist...(Raman)
• 1.2 Atmospheric turbulence for
  astronomy....(Vernin)
• 2. Instrumentation for Observing Optical
  Turbulence
• 2.1 Remote optical turbulence sensing present
  and future...(Tokovinin)
• 2.2 Standard and commonly used optical
  turbulence profilers ..(Chun et al.)
• 2.3 Seeing by site monitors versus VLT image
  quality ...(Sarazin et al.)
• 3. Adaptive Optics - Interferometry
• 3.1 Introduction to Adaptive Optics The Quest
  for Image Quality.(Businger, Tokovinin et
  al.)
• 4. Modeling Optical Turbulence
• 4.1 The Missing Link Between Meteorology and
  Astronomy...(Simons Roy)
• 4.2 Optical Turbulence Modeling and Forecast.
• Towards a new era for ground-based
  astrononmy.(Masciadri)
• 4.3 An operational perspective for modeling
  optical turbulence.(MKWC)

17
Status

• Most of the papers have been reviewed.
• Formatting
• Glossary
• Editing issues
• Deadlines
• Draft by the end of the year
• Published by spring 2009

18

• Thank you to the contributors

19
Questions?