Ensemble-based prediction and diagnostics during the predict field experiment


НазваEnsemble-based prediction and diagnostics during the predict field experiment
Дата конвертації20.04.2013
Розмір445 b.
ТипПрезентации


Ensemble-based prediction and diagnostics during the PREDICT field experiment

Sharan Majumdar (RSMAS / U. Miami)

Ryan Torn (SUNY at Albany)

Fuqing Zhang (Penn State)

Acknowledgments: all PREDICT PIs and Forecasters

The problem

  • Only ~20% of tropical waves develop into depressions.

  • Need to advance understanding of the physical differences between developing and non-developing systems.

  • Tropical cyclone formation is a multi-scale process, which depends on

    • Large-scale environment in which it is embedded
    • Wave’s structure (low-level vorticity, mid-level moisture)
    • Convective-scale processes: hot towers etc.
  • Seek to investigate utility of ensemble forecasts

    • Global; Regional; Convective-scale


Offer a longer-term outlook (>1 week) on the potential for a tropical disturbance to develop.

  • Offer a longer-term outlook (>1 week) on the potential for a tropical disturbance to develop.

    • Scatter of forecasts depicting critical values of
    • Area-averaged vorticity
    • Geopotential thickness anomaly
    • Okubo-Weiss parameter
    • Probabilities of exceedance of critical values of
    • Vertical wind shear
    • Lower-tropospheric relative humidity
    • Upper level divergence / lower-level convergence
  • Offer hypotheses on processes that lead to errors in large-scale conditioning (and thereby inaccurate representations of smaller-scale processes)



Example: Genesis of Karl (AL13; PGI44L)

  • Monsoon-like trough predicted to meander northward from the northern coast of South America into the southern Caribbean.

  • ECMWF ensemble forecasts of

    • (TOP) 700-850 hPa relative vorticity averaged over disk of radius 300 km
    • (BOTTOM) 200-850 hPa thickness anomaly: Z(r=250 km) – Z(r=1000 km)
  • All initialized on 00 UTC, 8 Sept 2010

  • 7 days prior to genesis































Example: Genesis of Karl (AL13; PGI44L)

  • 7-day forecasts of

    • Probability that 700 hPa RH > 70%
    • Okubo-Weiss parameter (vorticity^2 – strain rate^2): indicator of strongly curved flow with minimum horizontal shearing deformation






























Example: Genesis of Karl (AL13; PGI44L)

  • Genesis occurred just before 00 UTC, 15 Sep

  • Next two “loops”

    • 9- through 0- day ensemble forecasts of area-averaged relative vorticity valid at genesis time
    • Same but displayed as PDFs










































How to verify ensemble forecasts?

  • Verify each ensemble member individually to determine whether they meet some threshold criteria for genesis.

  • Could examine area-averaged vorticity and thickness anomaly exceeding a critical value



Regional ensembles: WRF/EnKF







Future Work

  • Produce community real-time ensemble products in 2011 from NCEP GEFS (+ others?)

  • Determine viable objective criteria for genesis; explore tracking mechanisms

  • Evaluate TIGGE ensemble predictions versus their own respective analyses

  • Comparison with Torn ensemble sensitivity (WRF)

  • Detailed investigation of processes that are adequately / inadequately captured in ensemble members: what are the key limitations?



NCEP GFS Analyses 2007-9: TD





Схожі:

Ensemble-based prediction and diagnostics during the predict field experiment iconDepartamento de Geofísica Universidad de Chile Field campaign vocals chilean Upwelling Experiment (CUpEx)

Ensemble-based prediction and diagnostics during the predict field experiment iconKnowledge-based systems. Knowledge-based systems
В найбільш широкому розумінні будь-яка система штучного інтелекта повинна спиратися на знання
Ensemble-based prediction and diagnostics during the predict field experiment iconЗміст Вступ
Доказова медицина (ДМ) (evidence-based medicine) розділ медицини, що базується на доказах і передбачає пошук, порівняння, аналіз...
Ensemble-based prediction and diagnostics during the predict field experiment iconProbabilistic Verification of Ensemble Forecasts of Tropical Cyclogenesis

Ensemble-based prediction and diagnostics during the predict field experiment iconLightning Prediction Objectives: Lightning Prediction Objectives

Ensemble-based prediction and diagnostics during the predict field experiment iconThe Trans-regional Network on Early Diagnostics in Educational Desadaptation Транскордонна мережа ранньої діагностики навчальної дезадаптації дітей

Ensemble-based prediction and diagnostics during the predict field experiment iconExpert Peer Review of Indicators, Field Protocols and Report

Ensemble-based prediction and diagnostics during the predict field experiment iconNs initial Spin and Kick (and Magnetic Field) Dong Lai

Ensemble-based prediction and diagnostics during the predict field experiment iconSO441 Synoptic Meteorology Numerical weather prediction a bit of important history

Ensemble-based prediction and diagnostics during the predict field experiment iconPrediction of the 10 July 2004 Beijing Flood with a High-Resolution nwp model Ying-Hwa Kuo1 and Yingchun Wang2


Додайте кнопку на своєму сайті:
dok.znaimo.com.ua


База даних захищена авторським правом ©dok.znaimo.com.ua 2013
звернутися до адміністрації
dok.znaimo.com.ua
Головна сторінка