Statistical evaluation of Sorbonne University Atmospheric Forecasting System (SUAFS) at Dubai Desert Conservation Reserve (DDCR) (phase 1)
Project Summary
Sorbonne University Atmospheric Forecasting System (SUAFS) is an integrated combination of advanced models able to provide detailed weather desert forecasts on various temporal scales over the Arabian Peninsula and the greater UAE areas.. The following predictions currently available by SUAFS (http://forecast.psuad.ac.ae) are:
- Weather forecast for 72 hrs ahead (3 days).
- Desert dust forecast for 72 hrs ahead (3 days).
SUAFS is running operationally since July 2018 and an annual archive of weather and desert dust forecasts are almost available for evaluation.
This research proposal offers a unique opportunity to initiate the framework for an important collaboration with DDCR in terms of environmental warnings and monitoring. Thus, the main aim of the Phase-1 project is to validate SUAFS forecasts against the atmospheric measurements obtained from DDCR meteorological stations. The entire records and data of the main atmospheric parameters from the network of the meteorological stations of DDCR, as well as the neighbored available stations, will be utilized for comparison with SUAFS forecasts. The extracting statistical scores will be helpful to identify the forecast errors of the system and their propagation into the period of the simulation. The results of this research work will lead SUAFS forecasts to achieve greater accuracy and reliability especially over the area of interest.
These results will be the base of the Phase 2, who focus on the air quality monitoring and dust forecasting dedicating for the DDCR. All the results will be at DDCR and SUAD disposal and a report with all the data will be submitted. Phase-1 will also act as a preparatory phase for attracting additional funding to launch the demanding Phase-2.
Research Objectives
- Process and homogenization of the atmospheric measurements from the DDCR meteorological stations network.
- Statistical evaluation of SUAFS.
- Achievement of greater accuracy and reliability for the weather forecasting conditions in DDCR broader area.
Methodology
In Phase-1, the primitive records (bulk data) obtained from the meteorological stations will be processed in order to be implemented for the SUAFS assessment. Atmospheric forecasts of specific parameters such as near surface temperature, wind speed, humidity, mean sea level pressure and precipitation will be also processed for direct comparison against in-situ records of the stations.
The evaluation of the SUAFS weather forecasts will be made in the DDCR area where the available surface stations have been installed. Additionally, surface observations from neighborhood networks will be used to verify and compare categorical model forecasts of the 10-m wind field, 2-m air temperature and sea level pressure and the accumulated precipitation.
The evaluation methodology will be based on the point-to-point comparison between model-generated variables and observations. For the variables of wind speed, air temperature, humidity and sea level pressure the scores produced are the standard mean error (BIAS) and the root mean square error (RMSE). The verification scores used for the precipitation will be derived using the contingency table approach. This is a two-dimensional matrix where each element counts the number of occurrences in which the gauge measurements and the model forecasts exceeded or failed to reach a certain threshold for a given forecast period. Thus, the forecast skill can be measured by evaluating the bias score (BS) and the equitable threat score (ETS).
Expected Outcomes
- SUAFS statistical scores (BIAS, RMSE, ETS) in the forecast window of 06-72hours
- Final report or publication in scientific journal or in conference proceedings. Presentation or one-day workshop in DDCR or SUAD to present the results.
