by Bettina Matti, EFAS Dissemination Centre
The 14th EFAS Annual Meeting took place in Stresa, Italy between the 21 and 22 of May 2019, and the participants were invited to answer the yearly survey regarding the satisfaction of the EFAS performance in general, the service and the products. A link for a web-based survey was made available to all EFAS partners. 36 responses were received, compared to 43 in 2018 and 22 in 2017.It is noted however that in 2018 the survey was sent to all partners; the response rate for 2019 is therefore very positive. The survey was, as in previous years, anonymous.
The survey responses are summarised according to the following categories:
- Overall satisfaction
- Skill, performance and trust
- EFAS services
- EFAS products and added value
- Satisfaction with the EFAS centres
- Future developments
No major changes in overall satisfaction were reported in comparison to the previous two years. It is noted however that satisfaction in overall performance during the past 12 months (question 3) has decreased, with four rating the performance as “low” and fewer rating it as very high. This may be related to the decrease in satisfaction relating to flash flood notifications (question 12).
Skill, performance and trust
In general, most questions showed a slight increase in satisfaction compared with previous years. Notable changes are that model skill is not used, and that more partners are taking the EFAS notifications more seriously when they are received.
In general there are no significant changes between the responses received for 2019 and in previous years. A notable change is that fewer organisations are using the EFAS web services, and that fewer partners are reading the EFAS Bulletins. All users are positive towards the new EFAS-IS, with 28 responding with “high” or “very high”.
A wide range of topics for future EFAS webinars were lifted. Many of these have previously been covered by the EFAS DISS, implying that previous webinars need to be re-advertised to partners (see “Possible improvements”). Topics requested were:
- The EFAS interface
- Nowcasting and Ericha
- Forecast criteria and improved understanding (probabilities, flash flood thresholds, uncertainties, ensemble post-processing, runoff generation during extreme rainfall)
- Calibration and model verification
- New EFAS products
- How to work with MARS and access data
- Hydrological modelling and LISFLOOD
- Seasonal prediction
- Impact forecasting
EFAS products and added value
In general, most questions show either a very small increase or decrease but no major change. An encouraging trend is seen in added value from both formal and informal notifications. A prominent negative trend is seen in the added value of flash floods (question 12), relating presumably to the change in notification criteria during 2019. This may also correspond to the increase in negative answers to question 3.
In terms of feedback, an increase in the number of positive responses regarding the feedback process is observed. Partners commented that the new process enables “more flexibility” and that the “survey is simplified”, but that “sometimes it is difficult to provide an answer”. In addition, further automation of the process was suggested, although no details provided.
Satisfaction with each of the EFAS centres has also increased from last year (this question was first asked 2018).
A number of questions were asked relating to future developments in either the EFAS products or the availability of EFAS data. In addition, partners were asked whether they were aware that the possibility exists to activate EMS Mapping for satellite imagery support in cases of flooding. Encouragingly, more than 70% are aware of this service.
A discussion on making EFAS data more open was also held at the 14th Annual Meeting, following on from the same discussion at the previous meeting. The new proposal suggests only presenting certain initial condition layers, rather than all datasets, and is summarised in Figure 21. Previously, 20% (8 people) replied that data should not be made more open, compared to 6% this year (2 people). 50% replied “yes” (18 people), compared to 37% (15 people) previously. Approximately half of the partners remain uncertain, as last year.
Lastly, participants were asked what future services they would like to see included in EFAS. The results were relatively equal:
- Continue increasing the skill of existing products (score = 3.28)
- More weather forecasts in addition to ECMWF, DWD and COSMO-LEPS (e.g. GFS) (score = 3.03)
- Coastal flood forecasts (score = 2.45)
Partners were also able to provide specific suggestions for future services. Responses were:
- Inclusion of additional hydrological models
- Storm surge forecasts
- Flash flood and informal notification skill indicators
- Dam influence in flood notifications
- Improvements to seasonal forecasting
- Inclusion of the point rainfall post-processed ensemble forecasts from ECMWF, with probabilities of precipitation type
- Cross-cutting products using data from other Copernicus services