|
||||||||||||||||||||||||||||||||||||||||||
|
Analyses & Numerical Prediction Analysis and Prediction Operations Bulletin No. 57
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The operational tropical region Limited Area Prediction System known as TLAPS_PT375 was upgraded to a new version in NMOC Melbourne on 11 December 2002. The BMRC Model Development Group, led by Dr. Kamal Puri, and the Data Assimilation Group, led by Dr Bill Bourke developed this upgraded version. It incorporates a number of minor model changes including:
The upgraded version underwent a parallel operational trial in NMOC during October to December 2002 after a long period of initial testing in BMRC since late 2001. Objective assessments were carried out against analysis and observations. Skill scores indicate that the updated TLAPS_PT375 generally performed better than the previous version throughout the 48-hour forecast period over the Australian region and very similar to the current version over the tropics.
The configuration of TLAPS_PT375 has not changed from that described in Analysis and Prediction Operation Bulletin No 48 (APOB48) of 22 September 1999.
Analyses are performed at time T-6 and T0 hours. First guess for these analyses are GASP +6 and +12 hour forecasts, with any pre-existing tropical cyclone circulation in the GASP forecasts being removed and replaced with an idealised vortex at the observed location. GASP also provides the boundary conditions for the model forecasts. The pre-processor collates and converts GASP data from pressure levels onto sigma levels prior to input into the analysis and prediction parts of the system.
Manual input is included by the insertion of tropical cyclone bogus observations supplied by the Darwin Regional Specialised Meteorological Centre (RSMC) when appropriate.
The soil moisture-wetness is set at the start of the run using results from the Pescod daily soil moisture scheme over Australia and the output of the previous model run at -12 hours to current run. There is now no reset to fortnightly climatology over any part of the domain unless the previous run was more than 24 hours old and then the reset is only outside Australia. In addition the modelled sub-surface soil moisture and soil temperature are set prior to the model run to be derived from soil moisture-wetness and from previous run data respectively. In the parlance of the algorithm, sub-surface soil moisture has a cold start and the soil temperature a warm start. If the previous model run is more than 24 hours old then the soil temperature also has a cold start.
Analysis program changes include:
The overall tropical cyclone bogus scheme as described in APOB48 (and detailed in Davidson & Weber) remains the same. Changes to the scheme to produce a more realistic vortex include:
Forecast model changes include:
The general operational configuration of the TLAPS system remains essentially as described in APOB48, e.g. the model domain, resolution, vertical level structure etc have not been changed and readers should refer to that report for details, noting the following points:
NMOC conducted an assessment of TLAPS_PT375 in a parallel trial during October - December 2002 prior to operational implementation. In addition to the conventional verification targets, such as the mean sea level pressure (MSLP) and upper winds (WIND), surface weather parameters such as the screen temperature and 10m winds were also verified. The results from these two different targeted groups and a brief comparison of rainfall distribution are discussed below.
1. S1 Skill Score, Bias and Root Mean Square error for MSLP and WIND
Table 1 below presents the mean S1 skill score, bias and root mean square (RMS) error of the previous TLAPS_PT375 (1999) and the new TLAPS_PT375 (2002) for the period 21 October to 7 December 2002 over the Australian and Australian Tropics verification region. The new version generally performed better than the old model throughout the 48-hour forecasting period over the Australian region, but there was mixed results over the Australian Tropics.
| Table 1. | Mean S1 Skill Score, Bias and RMS Error for MSLP and WIND comparing the previous TLAPS_PT375 model and the new version. | ||
|---|---|---|---|
| Region: | 1. Australian Region | 2. Australian Tropics(25N - 25S, 100E – 160E ) | |
| Total Period: | 21 Oct - 07 Dec 2002 using 00 UTC runs (48 cases) | ||
| Verifying Analyses: | Self | ||
| Change: | absolute value of TLAPS_PT375 (1999) - absolute value of TLAPS_PT375 (2002). A positive improvement implies that the new version is skilled better than the old version. | ||
| Australian Region | Australian Tropics | |||||||
|---|---|---|---|---|---|---|---|---|
| Forecast period | Verification statistic | Field | 1999 | 2002 | Change | 1999 | 2002 | Change |
| +24HRS | S1 Skill Score | MSLP | 21 | 21 | 0 | 30 | 32 | -2 |
| BIAS | MSLP | 0.6 | 0.5 | 0.1 | 0.4 | 0.2 | 0.2 | |
| 850 WIND | 0.1 | 0.0 | 0.1 | 0.1 | 0.1 | 0.0 | ||
| 500 WIND | 0.1 | 0.1 | 0.0 | 0.3 | 0.1 | 0.2 | ||
| 200 WIND | 0.3 | 0.2 | 0.1 | 0.3 | 0.4 | -0.1 | ||
| RMS | MSLP | 1.5 | 1.3 | 0.2 | 0.8 | 0.8 | 0.0 | |
| 850 WIND | 3.3 | 3.2 | 0.1 | 2.5 | 2.7 | -0.2 | ||
| 500 WIND | 4.4 | 4.4 | 0.0 | 3.2 | 3.5 | -0.3 | ||
| 200 WIND | 5.6 |
5.8 | -0.2 | 4.5 | 5.0 | -0.5 | ||
| +48HRS | S1 Skill Score | MSLP | 30 | 30 | 0 | 38 | 40 | -2 |
| BIAS | MSLP | 0.7 | 0.5 | 0.2 | 0.5 | 0.2 | 0.3 | |
| 850 WIND | 0.1 | 0.1 | 0.0 | 0.2 | 0.4 | -0.2 | ||
| 500 WIND | 0.1 | 0.1 | 0.0 | 0.6 | 0.5 | 0.1 | ||
| 200 WIND | 0.4 | 0.3 | 0.1 | 0.6 | 0.9 | -0.3 | ||
| RMS | MSLP | 2.1 | 2.0 | 0.1 | 1.2 | 1.1 | 0.1 | |
| 850 WIND | 4.3 | 4.3 | 0.0 | 3.4 | 3.6 | -0.2 | ||
| 500 WIND | 6.3 | 6.4 | -0.1 | 4.1 | 4.4 | -0.3 | ||
| 200 WIND | 8.4 | 8.6 | -0.2 | 5.8 | 6.3 | -0.5 | ||
Units for bias and RMS are hPa for MSLP and m.s-1 for WIND. S1 skill score is non-dimensional.
2. Surface Weather Parameter Verification against the Aviation METAR Observations
A new verification display system that highlights the statistical characteristics of surface weather elements has been implemented in NMOC. The system, developed by Kevin Tory of the BMRC Model Development Group, verifies weather elements such as surface (2m, or screen level) temperature and dewpoint and surface (10m) wind direction and speed against METAR observational data. RMS errors and biases are calculated at regular intervals out to a specified forecast period using all available Australian METARS (approximately 297).
|
|
| Figure 2. Screen (2m) dry bulb temperature comparison between the previous TLAPS_PT375 and the new TLAPS_PT375 errors averaged over all (approx. 297) available METAR sites and over all hourly model forecast periods out to +48 hrs for a) Bias and b) Root Mean Square (RMS) Error |
|
|
| Figure 3. 10m Wind Speed comparison between the previous TLAPS_PT375 and the new TLAPS_PT375 errors averaged over all (approx. 297) available METAR sites and over all hourly model forecast periods out to +48 hrs for a) Bias and b) Root Mean Square (RMS) Error |
Figures 2 and 3 show the time series plots of the mean bias and RMS errors averaged over all METAR sites and over all hourly model forecast periods out to +48 hrs. Results are shown for the period 17 November to 8 December. It can be seen that, overall, there is a clear and consistent improvement in the new version of TLAPS_PT375 (TLAPS_TEST) over the previous version (TLAPS) in predicting surface temperature and wind.
Figures 4 and 5 show 2m temperatures and wind speed forecasts respectively, for 48-hour period commencing at 1200 UTC on 26 November 2002 for (a) Alice Springs, (b) Townsville and (c) Darwin. Overall the new version shows improvement over the old version especially in improved forecast wind strength and timing.
3. Rainfall Verification
Figure 6 shows a comparison of the 24-hour rainfall analysis from NMOC up to 0000 UTC 6 January 2003 with the 24, 48 and 72h rainfall from the new version of TLAPS_PT375 for the same period. The meteorological situation was a developing monsoon low off the northeast Northern Territory coast and a weak frontal system moving across Western Australia. The analysis shows falls of 50-150mm near the monsoon low with light falls over Western Australia, Queensland and the southeast coast of Australia. All three forecasts period captured the situation very well. Overall the new version tends to increase the amount of rainfall, which has been observed to be more realistic for tropical convection.
4a)
4b)
4c)
Figure 4. Screen (2m) dry bulb temperature comparison of forecasts against METAR observations from (a) Alice Springs, (b) Townsville and (c) Darwin for the previous version (tlaps) and the new version (tlaps_test).
5a)
5b)
5c)
Figure 5. 10m wind speed comparison of forecasts against METAR observations from (a) Alice Springs, (b) Townsville and (c) Darwin for the previous version (tlaps) and the new version (tlaps_test).
Figure 6a: Verifying rainfall over Australia for 24 hours to 0000 UTC Monday 6 January 2003 from NMOC analysis |
Figure 6b: TLAPS_PT375 rainfall forecast for 24 hours up to 0000 UTC Monday 6 January 2003 |
Figure 6c: TLAPS_PT375 rainfall forecast for 24-48 hours up to 0000 UTC Monday 6 January 2003 |
Figure 6d: TLAPS_PT375 rainfall forecast for 48-72 hours up to 0000 UTC Monday 6 January 2003 |
Figure 6. 24-hour rainfall analysis and forecasts for the period up to 0000 UTC 6 January 2003. All totals are in mm.
Apart from the extended 72-hour forecast range, this upgrade should be essentially transparent to most users of TLAPS_PT375 products. The only significant product modification is that the cloud amount fields produced from the model are now expressed as percentages (0-100%) rather than the fractional units (0.0-1.0) used previously. This change was made to be consistent with the cloud fields produced by the global GASP system.
A limited number of extra charts are available on DIFACS to display the new +72 hr forecast available from the system. The relevant DIFACS slots are:
| 0340: | Mean Sea Level Pressure/Thickness +72 hr | |
| 0341: | 850 hPa Wind Barb / Height (Australian Tropics) +72 | |
| 0342: | 700 hPa Wind Barb / Height (Australian Tropics) +72 | |
| 0343: | 500 hPa Wind Barb / Height (Australian Tropics) +72 | |
| 0344: | 250 hPa Wind Barb / Height (Australian Tropics) +72 | |
| 0345: | 200 hPa Wind Barb / Height (Australian Tropics) +72 |
Data available in the real-time database is essentially the same as was described in APOB48 except for the addition of the extra forecast periods out to +72 hours (+54, +60, +66, +72 hours for the low-resolution 0.75 degree 120x160 grids, and 1-hourly from +49 to +72 hours for the full-resolution 0.375 degree 240x360 single level field grids).
TLAPS_PT375fields can be accessed from the real-time database (rtdb2) by using MCIDAS ADDE. The KENNY visualization program is also being modified to allow ADDE access and this method will give access to all TLAPS_PT375 data.
Data from TLAPS_PT375 will continue to be archived on SAM-FS with the same name format as before.
It is planned during 2003 that the TLAPS and TC-LAPS models will be combined, with the TLAPS model providing the large-scale environment for the high resolution TC-LAPS model. This will include extending the domain of TLAPS to 143°W, covering most of Fiji RSMC’s tropical cyclone region. This version is currently being tested in BMRC.
With the combined model it is expected that the new bulk explicit microphysics changes to the model will be introduced. This will improve the realism of the representation of moisture within the model by introducing fields of cloud and rainwater, ice crystals, snow and graupel. Details of this scheme are available in Dare.
There are also plans to soon commence ingesting TLAPS_PT375 model output into the MARS archive system.
Davidson, N.E. and Weber, H.C. "The BMRC High-Resolution Tropical Cyclone Prediction System: TC-LAPS" Mon.Wea.Rev,. 128, 1245-1265, 2000.
Dare, R., "Bulk Explicit Microphysics in the LAPS models", BMRC Research Report No. 90, 95-98, 2002.
