Future Plans  Refine Machine Learning:  Investigate optimal pressure level to use as input  Investigate use of neural network  Add additional input.

Slides:



Advertisements
Similar presentations
Improvements to Statistical Intensity Forecasts John A. Knaff, NOAA/NESDIS/STAR, Fort Collins, Colorado, Mark DeMaria, NOAA/NESDIS/STAR, Fort Collins,
Advertisements

RAMMT/CIRA Tropical Cyclone Overview THE DVORAK TECHNIQUE Introduction Visible Technique IR Technique Strengths and Weaknesses Lab Exercise: Visible Pattern.
A Blended, Multi-Platform Tropical Cyclone Rapid Intensification Index
Robert DeMaria.  Motivation  Objective  Data  Center-Fixing Method  Evaluation Method  Results  Conclusion.
Future Plans  Refine Machine Learning:  Investigate optimal pressure level to use as input  Investigate use of neural network  Add additional input.
Acknowledgments: ONR NOPP program HFIP program ONR Marine Meteorology Program Elizabeth A. Ritchie Miguel F. Piñeros J. Scott Tyo Scott Galvin Gen Valliere-Kelley.
Hurricane center-fixing with the Automated Rotational Center Hurricane Eye Retrieval (ARCHER) method Tony Wimmers, Chris Velden University of Wisconsin.
Microwave Imagery and Tropical Cyclones Satellite remote sensing important resource for monitoring TCs, especially in data sparse regions Passive microwave.
CORP Symposium Fort Collins, CO August 16, 2006 Session 3: NPOESS AND GOES-R Applications Tropical Cyclone Applications Ray Zehr, NESDIS / RAMM.
Diagnosing Tropical Cyclone Structure Presented by John Knaff with input and efforts from A. Schumacher, R. DeMaria, G. Chirokova, C. Slocum.
Exercise – Constructing a best track from multiple data sources NATIONAL HURRICANE CENTER JACK BEVEN WHERE AMERICA’S CLIMATE AND WEATHER SERVICES BEGIN.
Forecasting Tropical cyclones Regional Training Workshop on Severe Weather Forecasting and Warning Services (Macao, China, 9 April 2013)
Analysis of High Resolution Infrared Images of Hurricanes from Polar Satellites as a Proxy for GOES-R INTRODUCTION GOES-R will include the Advanced Baseline.
Formation of a tropical cyclone eye is often associated with intensification [1]. Currently, determination of eye formation from satellite imagery is generally.
ATMS 373C.C. Hennon, UNC Asheville Observing the Tropics.
Application of the Computer Vision Hough Transform for Automated Tropical Cyclone Center-Fixing from Satellite Data Mark DeMaria, NOAA/NCEP/NHC Robert.
1 Tropical cyclone (TC) trajectory and storm precipitation forecast improvement using SFOV AIRS soundings Jun Tim Schmit &, Hui Liu #, Jinlong Li.
Advanced Applications of the Monte Carlo Wind Probability Model: A Year 1 Joint Hurricane Testbed Project Update Mark DeMaria 1, Stan Kidder 2, Robert.
Applications of ATMS/CrIS to Tropical Cyclone Analysis and Forecasting Mark DeMaria and John A. Knaff NOAA/NESDIS/STAR Fort Collins, CO Andrea Schumacher,
Team Lead: Mark DeMaria NOAA/NESDIS/STAR Fort Collins, CO
Improvements in Deterministic and Probabilistic Tropical Cyclone Wind Predictions: A Joint Hurricane Testbed Project Update Mark DeMaria and Ray Zehr NOAA/NESDIS/ORA,
Lessons Learned from the Deployment and Integration of a Microwave Sounder Based Tropical Cyclone Intensity and Surface Wind Estimation Algorithm into.
Tropical Cyclone Applications of GOES-R Mark DeMaria and Ray Zehr NESDIS/ORA, Fort Collins, CO John Knaff CIRA/CSU, Fort Collins, CO Applications of Advanced.
New and Updated Operational Tropical Cyclone Wind Products John A. Knaff – NESDIS/StAR - RAMMB, Fort Collins, CO Alison Krautkramer – NCEP/TPC - NHC, Miami,
TC Intensity Estimation: SATellite CONsensus (SATCON) Derrick Herndon, Chris Velden, Tony Wimmers, Tim Olander International Workshop on Tropical Cyclone.
Improvements in Deterministic and Probabilistic Tropical Cyclone Surface Wind Predictions Joint Hurricane Testbed Project Status Report Mark DeMaria NOAA/NESDIS/ORA,
CIMSS TC Intensity Satellite Consensus (SATCON) Derrick Herndon and Chris Velden Meteorological Satellite (METSAT) Conference Ford Island Conference Center.
A Brief Digression: Waterspouts Szilagyi (2005, 2009) Waterspout Nomogram 850 hPa T: ~3-7°C SST: ~19-21°C.
STATISTICAL ANALYSIS OF ORGANIZED CLOUD CLUSTERS ON WESTERN NORTH PACIFIC AND THEIR WARM CORE STRUCTURE KOTARO BESSHO* 1 Tetsuo Nakazawa 1 Shuji Nishimura.
Possible impacts of improved GOES-R temporal resolution on tropical cyclone intensity estimates INTRODUCTION The Advanced Baseline imager (ABI) on GOES-R.
Statistical Evaluation of the Response of Intensity to Large-Scale Forcing in the 2008 HWRF model Mark DeMaria, NOAA/NESDIS/RAMMB Fort Collins, CO Brian.
A Comparison of Two Microwave Retrieval Schemes in the Vicinity of Tropical Storms Jack Dostalek Cooperative Institute for Research in the Atmosphere,
Hurricane Intensity Estimation from GOES-R Hyperspectral Environmental Suite Eye Sounding Fourth GOES-R Users’ Conference Mark DeMaria NESDIS/ORA-STAR,
Tropical cyclone products and product development at CIRA/RAMMB Presented by Cliff Matsumoto CIRA/CSU with contributions from Andrea Schumacher (CIRA),
Center for Satellite Applications and Research (STAR) Review 09 – 11 March 2010 Image: MODIS Land Group, NASA GSFC March 2000 Improving Hurricane Intensity.
Ocean Surface Winds Research Summary: Meteorological Applications Mark DeMaria, NOAA/NESDIS, Fort Collins, CO Ocean Surface Winds Workshop NCEP/Tropical.
The ARCHER automated TC center-fixing algorithm: Updates on real-time operations, accuracy and capabilities Anthony Wimmers and Christopher Velden Cooperative.
NHC/JHT Products in ATCF Buck Sampson (NRL, Monterey) and Ann Schrader (SAIC, Monterey) IHC 2007 Other Contributors: Chris Sisko, James Franklin, James.
Andrew Heidinger and Michael Pavolonis
Revisiting estimation of TC winds from IR Imagery.
The Impact of Lightning Density Input on Tropical Cyclone Rapid Intensity Change Forecasts Mark DeMaria, John Knaff and Debra Molenar, NOAA/NESDIS, Fort.
AMSU Product Research Cooperative Institute for Research in the Atmosphere Research Benefits to NOAA: __________________ __________________________________________.
CIMSS/NESDIS-USAF/NRL Experimental AMSU TC Intensity Estimation: Storm position corresponds to AMSU-A FOV 8 [1 30] Raw Ch8 (~150 hPa) Tb Anomaly: 5.36.
Can Dvorak Intensity Estimates be Calibrated? John A. Knaff NOAA/NESDIS Fort Collins, CO.
Tropical Cyclone Rapid Intensity Change Forecasting Using Lightning Data during the 2010 GOES-R Proving Ground at the National Hurricane Center Mark DeMaria.
Improving Intensity Estimates Using Operational Information John Knaff NOAA/NESDIS Regional and Mesoscale Meteorology Branch Fort Collins, CO.
Exploring a Global Climatology of Tropical Cyclone Eye Sizes ETHAN WRIGHT: UNC ASHEVILLE RESEARCH ADVISOR: DR. CHRISTOPHER HENNON 04/22/2015.
Development of a Rapid Intensification Index for the Eastern Pacific Basin John Kaplan NOAA/AOML Hurricane Research Division Miami, FL and Mark DeMaria.
Improved Statistical Intensity Forecast Models: A Joint Hurricane Testbed Year 2 Project Update Mark DeMaria, NOAA/NESDIS, Fort Collins, CO John A. Knaff,
AODT The Advanced Objective Dvorak Technique JHT Progress Report - Latest Advancements Timothy Olander, Christopher Velden, James Kossin, Anthony Wimmers,
TC Projects Joint Hurricane Testbed, Surface winds GOES-R, TC structure – TC Size TPW & TC size (Jack Dostalek) IR climatology – RMW/wind profile Proving.
Overview of CIRA and NESDIS Global TC Services Presented by John Knaff NOAA/NESDIS Regional and Mesoscale Meteorology Branch Fort Collins, CO USA For The.
The National Hurricane Center GOES-R Proving Ground Mark DeMaria NOAA/NESDIS, Fort Collins, CO GLM Science Meeting, Huntsville, AL September 26,
New Tropical Cyclone Intensity Forecast Tools for the Western North Pacific Mark DeMaria and John Knaff NOAA/NESDIS/RAMMB Andrea Schumacher, CIRA/CSU.
TOWARD AN OBJECTIVE SATELLITE-BASED ALGORITHM TO PROVIDE REAL-TIME ESTIMATES OF TC INTENSITY USING INTEGRATED MULTISPECTRAL (IR AND MW) OBSERVATIONS Jeff.
Preparing for GOES-R and JPSS: New Tropical Cyclone Tools Based on 30 Years of Continuous GOES IR Imagery John A. Knaff1, Robert T. DeMaria2, Scott P.
Tropical Cyclone Forecasting and Monitoring
Training Session: Satellite Applications on Tropical Cyclones
Mark DeMaria and John A. Knaff - NOAA/NESDIS/RAMMB, Fort Collins, CO
Automated Objective Tropical Cyclone Eye Detection
Accounting for Variations in TC Size
AUTOMATED TROPICAL CYCLONE EYE DETECTION USING DISCRIMINANT ANALYSIS
GOES-R Risk Reduction Research on Satellite-Derived Overshooting Tops
Cal/Val Activities at CIRA
Objective Methods for Tropical Cyclone Center
Training Session: Satellite Applications on Tropical Cyclones
TC Intensity Estimation: SATellite CONsensus (SATCON)
Advanced Dvorak Technique
Statistical Evaluation of the Response of Intensity
Validation of CIRA Tropical Cyclone Algorithms
Presentation transcript:

Future Plans  Refine Machine Learning:  Investigate optimal pressure level to use as input  Investigate use of neural network  Add additional input parameters from retrieved wind field  Investigate use of ATMS instead of AMSU (Fig. 4 and 5)  Using MIRS processing system from ATMS for T and q retrievals instead of statistical retrievals  Add VIIRS Image Processing:  Use microwave storm center location to sub-sect VIIRS data  Find center estimate based on image gradient features in VIIRS data (Fig. 4 and 5)  Spiral patterns fitted to Harris corner/edge detection locations (Collins)  IR image processing method from Ritchie et al. (2011)  Feed estimates into machine learning algorithm to determine final storm center estimate References Bessho, K., M. DeMaria, and J. A. Knaff, 2006: Tropical Cyclone Wind Retrievals from the Advanced Microwave Sounding Unit: Application to Surface Wind Analysis. Journal of Applied Meteorology and Climatology, 45, Collins, Robert. “Harris Corner Detector.” Penn State University. Lecture Ritchie, E. A., G. Valliere-Kelley, M. F. Piñeros, and J. S. Tyo, 2011: Improved tropical cyclone intensity estimation using infrared imagery and best track data. Wea. Forecasting, 27, Wimmers, A.J., C.S. Velden, 2010: Objectively Determining the Rotational Center of Tropical Cyclones in Passive Microwave Satellite Imagery. Journal of Applied Meteorology and Climatology, 49, Introduction  Only west Atlantic has routine hurricane hunter aircraft for finding storm centers  Satellite data used subjectively to find centers across the globe  Objective center fixing in real-time highly desirable  CIMSS ARCHER method (Wimmers et al. 2010) estimates center from microwave imagery by fitting spiral patterns  Alternate objective method will use machine learning techniques with multi-spectral data from S-NPP  ATMS and VIIRS Center Fix Method  Features from AMSU (ATMS) fields input to machine learning to provide microwave-only center estimate  AVHRR (VIIRS) data will refine microwave center estimate using image processing and machine learning Microwave Center Method  Select area around extrapolated position  Area size based on average error between the extrapolated position and the true center  Used ~0.5 degrees in either direction from the extrapolated position  Each grid cell in the selected area represents a single row of data:  Pressure  Laplacian of pressure  Distance from min pressure in selected area  Value indicating if the cell contains the true storm center  Perform Linear Discriminant Analysis (LDA), and Quadratic Discriminant Analysis (QDA)  40,125 grid cells selected for training and divided into two classes  Class1: No Storm Center Present  Class 2: Storm Center is Present  LDA and QDA, after training, provide a function for each class indicating how probable a new grid cell is to belong to each class  Grid cell most probable to be the storm center is found for each satellite image (Fig. 2)  The distance between the selected grid cell and the true storm center position is measured for verification of the algorithm (Fig. 3) Machine Learning Techniques for Tropical Cyclone Center Fixing using S-NPP Robert DeMaria 1, Charles Anderson 2 1 NOAA/NESDIS Regional and Mesoscale Meteorology Branch (RAMMB), Ft. Collins, CO 2 Department of Computer Science, Colorado State University (CSU), Ft. Collins, CO Data  Initial development with AMSU from POES as ATMS proxy  AMSU statistical retrievals used to provide T and moisture profiles  Hydrostatic and nonlinear balance constraints provide geopotential height (Z) and wind field (Bessho et al. 2006)  Standard levels 1000 to 100 hPa  Data includes 2,021 Atlantic TC cases from 2006 to 2011  Extrapolated position available at time of satellite image creation  “Best track” data used as truth for training/testing  Fig. 1 shows example 700 hPa Z field from AMSU  Visible and IR window AVHRR data collected for AMSU cases  VIIRS Day/Night and IR I05 bands collected for ATMS cases Figure 1. Hurricane Katia 04 Sept UTC Preliminary Results  Mean Error Using LDA: ~0.45 degrees  Mean Error Using QDA: ~0.40 degrees  11% improvement using QDA Figure 2. Sample Probability Field With Maximum Probability Marked Figure 3. Distance errors of the center estimates for a testing dataset Figure 4. MIRS wind and geopotential height retrievals and VIIRS visible image for Hurricane Leslie 09 Sept UTC Figure 5. MIRS wind and geopotential height retrievals and VIIRS visible image for Hurricane Leslie 02 Sept UTC