Presentation is loading. Please wait.

Presentation is loading. Please wait.

Office of Marine Prediction, Japan Meteorological Agency

Similar presentations


Presentation on theme: "Office of Marine Prediction, Japan Meteorological Agency"— Presentation transcript:

1 Sea Surface Temperature monitoring for the western North Pacific at JMA
Office of Marine Prediction, Japan Meteorological Agency Toshiyuki SAKURAI*, Mika KIMURA, Akiko SHOJI, Hiromu KOBAYASHI and Yoshiaki KANNO * 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

2 Introduction Office of Marine Prediction (OMP) is responsible for providing oceanographic information in the western North Pacific. Sea Surface Temperature (SST) analysis/forecast is one of the key OMP’s products. SST information is used for Fisheries and other marine industries (ship routing, tourism, etc.) Recreation Monitoring oceanographic environment (Coral bleaching, etc. ) Ocean data assimilation system and boundary conditions of weather forecasting system 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

3 Flow of Data and Information
observation analysis forecast Provision GTS Internet Mail/FAX Users (Ships, governmental agencies, private companies, general public) numerical model 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

4 Satellites’ SST data JMA/MSC JAXA NOAA/NESDIS OMP/JMA L3 SST
【LEO】 Low Earth Orbit Satellite 【GEO】 Geostationary Satellite JMA/MSC NOAA/NESDIS JAXA GAC L1B Data AVHRR/NOAA-18, -19, Metop-A ACSPO L3U SST VIIRS/sNPP L2 SST AMSR2/GCOM-W1 WindSat/Coriolis L3 SST Himawari-8, MTSAT-2 L3 SST (from HRPT)   AVHRR/NOAA-18, -19 Metop-A OMP/JMA 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

5 Himawari-8 L3 SST JMA’s Meteorological Satellite Center (MSC) produces Himawari-8 L3 SST data. Same SST retrieval algorithm as used by JAXA based on a quasi-physical algorithm (Kurihara et al. 2016) Cloud mask based on JMA’s Fundamental Cloud Product for Himawari-8 Hourly, 0.02 horizontal resolution (0.04 for MTSAT-2) Coverage: 60S – 60N, 80E – 160W Himawari-8 MTSAT-2 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

6 Himawari-8 SST Validation (Monthly Statistics)
Match-up of satellite and buoy and Argo float SSTs with time differences within 1.25 hours and spatial distances of less than 10km Bias: -0.3~-0.4°C, RMSE: 0.7~0.8°C for Himawari-8 The result shows significant improvement over those from MTSAT-2. MTSAT-2 Himawari-8 Himawari-8 MTSAT-2 *These statistics values are provided by MSC/JMA 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

7 SST analysis products of OMP
Himawari-8 SST imagery (Operational) Daytime and nighttime composite for the seas adjacent to Japan MGDSST (Operational) Global, 0.25 resolution, Daily OI based analysis using LEO data Data period: 1982/1~ HIMSST (Experimental) Western North Pacific, 0.1 resolution, Daily OI based analysis using LEO and GEO data Data Period: 2015/10~ (w/ Himawari-8) 2013/6~2016/3 (w/ MTSAT-2) 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

8 (1) Himawari-8 SST imagery
Daytime (9-20h LT) and nighttime (21-8h LT) composite. AHI/Himawari-8 doubles the spatial resolution of MTSAT-2, and it’s SST imagery enables to detect paths of ocean surface currents and ocean fronts clearly. The SST imagery is useful for fishery and marine industry. Kuroshio Path Warm water intrusion Daytime composite 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

9 (1) Himawari-8 SST imagery
High frequency observations of Himawari-8 is the biggest advantage and enable to reduce cloudy area, compared to LEO. Himawari-8 SST imagery has been in public via the JMA website since 25th May 2016. URL VIIRS/sNPP 1 day composite Himawar-8 Daytime composite Cloudy area 2016/6/27 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

10 (2) MGDSST MGDSST (Merged satellite and in-situ data Global Daily Sea Surface Temperature) OMP/JMA’s operational SST products Global, 0.25  x 0.25  grid resolution, daily Input: AVHRR (NOAA-18, 19, MetOp-A), AMSR2, Windsat, in situ Biases of satellites’ data are corrected using in situ SSTs Scale decomposed space-time Optimal interpolation MGDSST is used as a lower boundary condition of NWP models and used as observation data for ocean assimilation system. MGDSST analysis contributes to the GHRSST Multi-Product Ensemble (GMPE) system (Martin et al, 2012) as one of input data. 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

11 (2) MGDSST MGDSST is used for SST maps on JMA’s website. (left)
NEAR-GOOS regional real time data base provides SST maps and GPV (text format) of MGDSST (right) 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

12 (3) HIMSST (regional SST)
HIMSST (High-resolution MGDSST) Experimental product western North Pacific, 1/10 resolution, Daily Input: AVHRR, AMSR2, Himawari-8, in-situ HIMSST MGDSST HIMSST w/ Himawari-8 The new regional SST analysis (left) shows sharper SST gradients. 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

13 Spatial-temporal scale decomposition
Short wavelength Middle wavelength Long wavelength Long period   AHI AVHRR AHI AVHRR Microwave AVHRR Microwave 53 Limit due to spatial resolution Period (day) Not used AVHRR Microwave Short period AHI AVHRR Microwave AHI 27 AHI Not used Not used 10 2 Limit due to temporal resolution 22 55 143 580 Wavelength (km) HIMSST utilizes the shorter and smaller scale components, which are not used in MGDSST. 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

14 Validation of HIMSST BIAS RMSD # of data -0.052 0.806 50787 -0.035
Target region: NW Pacific Period: Oct ~ Mar. 2016 Statistics Match-up of analyzed SSTs and in-situ SSTs (ship and buoy) , which are averaged over 1 x 1 grid. No systematic bias is recognized. Bias and RMSD of the HIMSST w/ Himawari-8 are smaller than those of MGDSST. SST Scatter diagram BIAS RMSD # of data HIMSST w/ Himawari-8 -0.052 0.806 50787 MTSAT-2 -0.035 0.820 49856 MGDSST -0.071 0.819 50791 COBE-SST* -0.067 0.879 50463 * COBE-SST is a SST product based on in-situ observation and is produced by JMA’s Climate Prediction Division. 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

15 Summary Himawari-8 SST data have higher spatial resolution and higher precision than MTSAT-2 SST data. Himawari-8 observes more frequently than those of LEO. OMP/JMA has two operational and one experimental SST products. - Himawari-8 SST imagery - MGDSST (global, 0.25 , OI-based analysis) - HIMSST (the western North Pacific, 0.1 , OI-based analysis) High frequency observations are the biggest advantage of Himawari-8 SST, and this enable to reduce cloudy area in Himawari-8 SST imagery compared to LEO. HIMSST (experimental product) is produced with a combination of GEO and LEO SST. It utilizes the shorter and smaller scale components derived from Himawari-8 SST, which are not used in MGDSST. 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

16 Thank you for your attention
2nd Japan-Australia GEO-LEO Application Tokyo, 2016

17 Himawari-8 SST Validation (March 2016)
Match-up of satellite and buoy SSTs with time differences within 1.25 hours and spatial distances of less than 10km from March 1 to 31 (until March 24 for MTSAT-2). Himawari-8 SST data were superior to those of MTSAT-2. Himawari-8 MTSAT-2 #: BIAS: RMSE: STDE: #: BIAS: RMSE: STDE: 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

18 Himawari-8 SST Validation (March 2016)
Large negative bias for satellite zenith angles exceeding 70 degrees and for night time Bias vs. local time Bias vs. satellite zenith angle 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

19 SST cooling response after typhoon passage
2016/2/2海洋気象技術検討会 SST cooling response after typhoon passage HIMSST exhibits clearer cooling response than MGDSST. 18 Oct. 2015 23 Oct. 2015 HIMSST HIMSST 18 Oct. 8/24 8/25 23 Oct. MGDSST MGDSST ― Typhoon track CHAMPI   Typhoon position on 18 and 23 Oct. 2015 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

20 Regional SST analysis with MTSAT/Himawari-8 SST
western North Pacific, 1/10 resolution, Daily Input : AVHRR, AMSR2, MTSAT, in-situ It utilizes shorter/smaller scale components, which are not used in MGDSST under development MGDSST Regional SST w/ MTSAT 2nd Japan-Australia GEO-LEO Application Tokyo, 2016

21 (3) HIMSST (regional SST) gradient comparison with MGDSST
w/ Himawari-8 The new regional SST analysis (right) shows sharper SST gradients than those of MGDSST because of its higher grid resolution and the use of short wavelength components from Himawari-8. 2nd Japan-Australia GEO-LEO Application Tokyo, 2016


Download ppt "Office of Marine Prediction, Japan Meteorological Agency"

Similar presentations


Ads by Google