**5. Conclusions**

Based on the practices of a number of countries, review of the literature, and evaluation of the survey on use of snow observations in the modelling environment, conclusions can be drawn on: (i) the status and future evolution of conventional snow observations from national networks and satellite products for DA and model validation, including availability, error characteristics and reliability, towards an improved usage of conventional snow observations from national networks for data assimilation and model validation; (ii) the review of the methods to combine remote sensing and conventional snow observations with modeling results for user applications; and (iii) snow observations errors for data assimilation and modelling systems.

The results of the survey show that the measurement networks, instruments, and techniques are exploited well by existing DA systems and used in model environments for NWP, hydrology, or special snow models. The survey reveals that there is a fit between the snow macro-physical variables required for DA and those provided by the measurement environment, since snow depth, snow presence, snow density and SWE are the most measured variables. It is also important to take into consideration that in many cases these variables are measured with different instruments and techniques, in particular snow depth and SWE. On the other hand, developments in DA systems are necessary to exploit the evolving capabilities of the observing systems, and vice versa. The increasing automation of the measurements requires enhanced data managemen<sup>t</sup> in the DA system (quality control, consistency). There is a requirement for remotely sensed snow depth or SWE observations from satellite, to provide snow data in regions with sparse measurement networks, but this necessitates developments in instrument technology (e.g., automatic measurement of snow microstructural properties) and also development of DA systems in order to make use of such observations. There are also concrete future plans on using enhanced snow observations for all model environments, in particular for NWP with DA, hydrology, and reanalysis. Data availability and resources to integrate the data in the model environment are the current barriers and limitations for the use of new or upcoming snow data sources independent of the model environment used.

The further outcomes from the survey support new, innovative and upgraded observing strategies; enhanced usage of snow data for scientific research and applications; a broader overview and easier access to existing snow measurements and snow model data for the benefit of different applications, such as NWP models, hydrological, climatology and climate change research. Further support of these aims is provided by the related COST HarmoSnow activities of the parallel survey on snow measurements and the training school on data assimilation in Europe [179]. The monitoring of floods, droughts, snow avalanches and hydropower production could benefit from improved real-time snow measurements for assimilation into operational prediction models to improve hydrological, meteorological and climate forecasting while a further integration and harmonization of the European snow network into global networks (e.g., WMO GCW) supports the strengthening WMO and EUMETSAT activities on snow observations. The main scientific impact will emerge from improved snow and weather products via better knowledge of snow properties and their evolution. It will induce a lasting structural improvement of the interaction between participating communities, thus very relevant for the Intergovernmental Panel on Climate Change (IPCC) and Copernicus (Global Monitoring for Environment and Security). Policy and decision makers at all levels from local safety to global environment policy will benefit from improved knowledge on current and future snow cover and climate conditions.

**Supplementary Materials:** COST ESSEM 1404 working group 3 survey: Questionnaire and results are available at http://www.harmosnow.eu/index.php?page=WG3.

**Author Contributions:** Conceptualization, J.H., A.S.S., P.d.R., S.P., D.C.F, C.D.M., R.A.M., M.L., M.D., G.P., V.P., D.V.-S, A.N.A.; Methodology, J.H., A.S.S., P.d.R., S.P., D.C.F, C.D.M., R.A.M., M.L., M.D., G.P., V.P., D.V.-S, A.N.A.; Writing (original draft preparation), A.S.S., J.H., P.d.R., S.P., D.C.F, C.D.M., R.A.M., M.L., M.D., G.P., V.P., D.V.-S, A.N.A.; Writing (review and editing), J.H., A.S.S., P.d.R., S.P., D.C.F, C.D.M., R.A.M., M.L., M.D., G.P., V.P., D.V.-S, A.N.A.

**Funding:** This research is part of the COST Action ES1404 activity funded by the COST programme.

**Acknowledgments:** We thank all of the participants of the COST Action ES1404 for their support and fruitful comments to the results. We also thank all of the respondents to the survey who provided the material for this study. We are grateful to Sylvain Joffre, who had a key role in planning and initiating the COST Action ES1404.

**Conflicts of Interest:** The authors declare no conflict of interest.
