1. Australian VLab Centre of Excellence National Himawari-8 Training Campaign Forecaster use of Rapid Scan Data: Part A

2. Part 1a: Instructions Now that you have downloaded the PowerPoint file, please read the Instructions in Parts 1a (this slide) and familiarise yourself with the Learning Outcomes in slide 1b (next slide) Print out the Worksheet in Part 2a of this file (if applicable) Examine the Pre-activity Resources in Part 2b of this file (if applicable). Download the appropriate accompanying Video Recording (.wmv file). The recording is typically of 3-13 minute duration. Commence listening to the Recording. Pause the Recording whenever you want to annotate notes on the Worksheet. You may wish to examine the slides in Part 3 of this file in Slideshow mode when you stop the Recording. Note that Rapid Scan imagery in the animations embedded in the PowerPoint slides is often clearer than in the Recording. Towards the end of the recording, recommended answers for the exercises are sometimes given.

3. Part 1b: Learning Outcomes At the end of this exercise you will: Have a basic knowledge of some tools, specifically Alerts, that the Operational Forecaster can use to effectively use 10 minute rapid scan satellite data within their busy forecasting routine. Have a basic understanding of the details of existing Alerts for thunderstorm detection and monitoring. Be familiar with a forecast funnel procedure to interrogate 10 minute rapid scan satellite data more effectively. Have a better understanding of how the new generation of satellite data may best be delivered to the Operational Forecaster. Note – corresponding WMO-1083 Capabilities and BOM Enabling Skills are given on the link "Learning Outcomes" on the National Himawari-8 Training Campaign homepage.

4. Part 2a: Worksheet for the exercise There is no "worksheet" associated with this activity Part 2b: Pre-activity resources There are no "pre-activity resources" associated with this activity

5. Part 3: Forecasters have a huge amount of information available to them, they have to be selective in what they choose Effective use of Rapid Scan satellite data: Timely Data Access of appropriate products for the current situation. Able to use / interpret Rapid Scan data systematically and effectively The importance of Alerts

6. Part 3: Forecaster use of Himawari 8/9 10 minute, 16 channel data. Reducing the data volume. 10 minute / 16 channel data only over the region of interest. This reduces data volume Similarity to the "nesting" of regional NWP models within the global models. (Google Earth analogy) Forecasters would like a "roving" window of high resolution also Coarse resolution / Limited channels / Half hourly data Victoria Full resolution /All channels / 24 hours of half hourly images, the last two hours 10 minute data Ability to extract 10 minute data blocks over the past 24 hours

7. Part 3: How to use and interpret Rapid Scan satellite data effectively Fast versus slow moving versus stationary features (eg. Jets, shear, clouds anchored to topography) Rotating features (lows etc.) Lifecycle of features Commencement, persisting, dissipating. Rapidly developing versus slowly developing. Rapid Scan data used to quickly highlight important features Highlighting mesoscale features (convection, cloud banding, smoke plumes etc.) Broadscale Case Study - Japan

8. Part 3: Severe Storm Alert Algorithm (Bedka et al. 2011) Mean BT of anvil cloud ≤ 225 K (8km from overshooting top) Overshooting tops BT ≤ 215 K Overshooting tops (OT) must be: ≥ 6.5 K colder than the mean BT of the surrounding anvil cloud colder than NWP model tropopause temperatures.

9. Part 3: Severe Storm Alert Algorithm (forwarded by G.Ferent, from Bedka et al. 2011) Here is an algorithm for automatic alerting and detection for overshooting convective tops likely to have severe weather associations (based on AVHRR examination of 450 storms associated with the inverted V signature) In particular overshooting tops (OTs) are associated with brightness temperatures (BTs) ≤ 215 K, and ≥ 6.5 K colder than the mean BT of the surrounding anvil cloud, which must be associated with a BT ≤ 225 K; and OTs must be associated with BTs colder than NWP model tropopause temperatures. The mean anvil BT is computed by sampling the anvil BT at an 8km radius from the OT minimum location in 16 directions. (most OT’s are less than 15 km in diameter)

10. These files were provided by Himawari-6 (MTSAT-1R) Rapid Scan Observations. These were performed for the sake of aviation users. Japanese Meteorological Agency 1 2 34 Part 3: RDCA convective cloud detection algorithm (JMA) July10th 2011

11. No.Diagnostic ParametersMain objective 1*VISR To detect optical thick cloud (mainly for **Pre-detection) 2Difference between maximum and minimum of VISR To detect a roughness in developing cloud 3Standard deviation of VISR 4 Difference between maximum and minimum of 10.8μm ***BT 5Standard deviation of 10.8μm BT 6Difference between 10.8μm and 12μm BT To exclude optically thin cloud (cirrus) (mainly for Pre-detection) 7Difference between 6.8μm and 10.8μm BTTo detect the potential to develop 8 Slope index (relation between 10.8μm BT and effective radius of cloud top estimated from 3.8μm) To evaluate cloud microphysical structure Time / trend parameters (cloud motion is considered) 9 Time differential of maximum of VISR To evaluate vertically developing trend of developing cloud 10 Time differential of averaged VISR 11 Time differential of minimum of 10.8μm BT 12 Time differential of averaged 10.8μm BT 13 Pinpoint fall down of 10.8μm BT Part 3: Rapidly Developing Cumulus Areas product

12. Part 3: “Lead time” for Lightning (5 minute rapid scan) 12 Average: about 20 minutes Detection before the lightning occurrence Delay 86 samples in the summer 2011 Slide taken from “Status of Japanese follow-on Geostationary Meteorological Satellites HIMAWARI-8/9”, Toshiyuki KURINO Meteorological Satellite Center (MSC), JMA; Joint Australian - Japanese Himawari-8/9 Symposium 22 May 2013, Canberra, Australia

13. Part 3: Current RADAR Alerts compared to thunderstorm detection algorithms Severe Storm Algorithm thresholds (Bedka et al.) RDCA method images courtesy JMA and BOM