Climate normals have been set up at Météo-France to characterize climate trends and to situate a day in its climatic context. First, these normals have been calculated for more than a thousand reference observations and have been recently updated with the 1991-2020 reference as stated by the World Meteorological Organization.

In order to assess more accurate temperature climatology of France, a high-resolution temperature climatology over France has been performed, based on a spatialization of extreme daily temperatures (called ANASTASIA) on a 1 km regular grid using a regression-kriging method. ANASTASIA data are available 1947 onwards.

Classic temperatures climatological indices are computed such as number of days above or below temperature thresholds, mean temperatures, etc. Less common indices are computed like the first/last date of temperature threshold or the climatological equivalent date of an observed temperature for a given day.

Those climatology variables are of great interest in daily climate monitoring in order to fully understand exceptional meteorological situations such as heatwaves. That is why real-time products are also produced, sometimes with an indication of the geographical area impacted as a percentage of France total area, and latter with the number of people impacted.

Records on extreme temperatures and the year of the record, earliest/latest date of temperature threshold are also computed in order to complete the climatological dataset.

ANASTASIA product is of great interest and allows climatologists to have a better spatial representation of temperature patterns. However, the method used in ANASTASIA does not reproduce well inversion temperatures for example and may lead to great errors in such situations in mountain areas as the product is highly dependent on the observation network.

How to cite: Sorel, M., Tiriau, M., and Van Hyfte, S.: A high-resolution temperature climatology over France using a spatialization of daily temperatures extremes from 1947 to present, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-177, https://doi.org/10.5194/ems2024-177, 2024.

CLIMADAT-hub is a two-year project within the framework of H.F.R.I call “Basic research Financing (Horizontal support of all Sciences)” under the National Recovery and Resilience Plan “Greece 2.0” funded by the European Union – NextGenerationEU. The project aims at bridging the gap between the available climatic information and the information required for assessing climate risks at the local scale by creating high resolution observational gridded datasets, as well as statistically downscaled seasonal forecasts and climate change projections for Greece.

Regarding the observational gridded datasets, the primary goal is to construct a state-of-the art, 1km high resolution gridded dataset for daily temperature, precipitation, relative humidity and wind speed for the period 1981-2021. Starting with temperature and precipitation, long term daily values have been collected from various meteorological networks and databases for a large number of locations in Greece. These raw daily data underwent quality control and homogenization. To obtain the daily gridded values for air temperatures (maximum, minimum, mean) and precipitation, at a spatial resolution of 1km the following methods have been examined: i) combination of classical geo statistical methods such as Thin Plate Splines and Kriging as used in the early versions of E-OBS and Iberia01, ii) Regression-Kriging, a spatial prediction technique commonly used in geostatistics that combines a regression of the dependent variable (e.g., temperature) on auxiliary/predictive variables (e.g., elevation, distance from shoreline) with kriging of the regression residuals (similar to EOBSv17 and afterwards), iii) ensemble machine learning, an approach to modeling, where instead of using a single best learner, multiple strong learners are used and, consequently, are combined into a single union, iv) hybrid method, where the available observations are blended with the Weather Research and Forecasting (WRF) model to produce the high resolution observed gridded datasets through gridding and bias adjustment techniques.

We report the results from evaluating all the created gridded datasets for temperature and precipitation against withheld station data to determine the best performing approach for each variable. Future work will extend these methodologies to include the remaining variables.

How to cite: Varotsos, K. V., Karali, A., Kitsara, G., Lemesios, G., Patlakas, P., Hatzaki, M., Tenentes, V., Katavoutas, G., Sarantopoulos, A., Koutroulis, A. G., Grillakis, M. G., Psiloglou, B., and Giannakopoulos, C.: High resolution observational daily gridded dataset for Greece: The CLIMADAT-hub project, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-643, https://doi.org/10.5194/ems2024-643, 2024.

At present, most operational climate grid datasets are available with a temporal resolution of one day. This can be limiting in applications that rely on an explicit representation of the diurnal cycle, such as when modelling temperature-dependent environmental processes (e.g. transpiration). In this study we propose a methodology for the construction of hourly datasets of surface air temperature and present results from its application over Switzerland. The method uses a spatio-temporal (ST) statistical model. Unlike classical interpolation, the ST approach exploits data from a period of times simultaneously, and therefore allows to formally represent the connection of spatial and temporal variations. This is particularly desirable in complex terrain, where the diurnal cycle of temperature has marked topographic imprints. As ST model we use a dynamic linear model (DLM), which is a conceptual extension of kriging with external drift (KED), and offers high flexibility to be configured to the specifics of a region. In our application, for example, the DLM incorporates model components for cold-air pooling, basin-scale inversions, and the lake- and valley effects.

We present results from an application of the method to several multi-day episodes with specific weather conditions and from a continuous application over several months. The results illustrate that the method is capable to represent complex spatio-temporal variations, including the build-up and decay of an inversion and distinct diurnal variations in valleys, the flatland and at lake shores. Comparison of the spatio-temporal DLM to a spatial-only KED shows added value of DLM in terms of enhanced temporal continuity of trend coefficients. However, cross-validation statistics are not significantly different between the two approaches. Further experiments suggest that this is due to the dense station coverage in Switzerland. In regions / times where data is sparser, or when model complexity is substantially increased, spatio-temporal modelling is expected to provide a measurable improvement over spatial-only modelling. Our presentation will illustrate the results of our application in films of the temperature evolution. We will also introduce the extensions, both methodological and technical, implemented for the continuous multi-month application, and quantify the predictive performance of the method in this setting.

Our study provides valuable insight on extending a familiar interpolation concept (KED) into a spatio-temporal model, suitable to produce sub-daily temperature grid datasets.

How to cite: Frey, L. and Frei, C.: Gridding of Hourly Surface Air Temperature – Application of a Spatio-Temporal Statistical Model in Complex Terrain, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-905, https://doi.org/10.5194/ems2024-905, 2024.

Today, the volume of data monitoring atmospheric conditions near the Earth’s surface is vast and continually expanding. Among the most rapidly developing sources are observational networks composed of stations managed by citizens. From the perspective of national meteorological services, this citizen-generated data presents an opportunity to enhance the existing networks of traditional weather stations operated by public institutions. For variables like temperature and precipitation, the presence of a dense network that delivers data at hourly or finer sampling rates enables the detailed reconstruction of weather phenomena occurring between the microscale and the mesoscale (i.e. from hundreds of kilometres down to 1-2 kilometres or less). The specific applications of the work we present are: i) development of quality control tests; ii) development of post-processing of observational and gridded datasets, aiming at providing enhanced gridded datasets reconstructing atmospheric variables near the surface.

In previous studies, we analyzed hourly precipitation data collected by a network of citizen-operated stations in Finland, Norway, and Sweden from September 2019 to October 2022. These observations were gathered using Netatmo weather stations, which are commercially available and installed in private homes for various purposes, including home automation. We compared this crowdsourced data with reference observations from WMO-compliant stations managed by the national weather services of the three countries. Our findings indicate that while reference observations consistently fall within the empirical distribution of the crowdsourced data, there are signs that intense precipitation events may be underestimated by crowdsourced data. Further investigation into the spatial variability of crowdsourced precipitation revealed significant deviations between measurements from locations as close as 1 to 5 km apart, with differences reaching up to 50% of the mean hourly precipitation in the area. This variation is partially attributed to the suboptimal siting and exposure of some Netatmo stations. However, it also provides an estimate of the inherent variability of precipitation over short distances.

In this study, we expand our research to include hourly temperature data collected by the same network of crowdsourced stations. We aim to address several research questions concerning the representativeness errors of citizen observations compared to WMO-compliant observations of hourly temperature. Specifically, we investigate whether systematic deviations exist between the two data sources, whether these deviations vary by season, and how accurately a normal distribution can model these deviations. Additionally, we will explore the spread of this normal distribution. Our research also seeks to quantify the typical variability in space of the deviations between crowdsourced data and reference observations, aiming to provide a more comprehensive understanding of representativeness errors.

How to cite: Båserud, L., Holm Høgemo, E., Seierstad, I. A., Lussana, C., Neuville, A., and Nipen, T. N.: Exploratory analysis of hourly observations of temperature measured by citizen observations over Scandinavia, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-515, https://doi.org/10.5194/ems2024-515, 2024.

Context. For monitoring, analysing and forecasting the impact of weather and climate change on society, we see an increasing need for high-quality, high-resolution gridded weather and climate services. The official networks of the National Meteorological and Hydrological Services (NMHS) sample part of the land-use excluding, for example, urban areas, which motivated us to include third-party data which does sample urban areas.  We have provided the results in real-time as well as historical gridded data services.

Approach. By including a statistical, simplified model of the observation error in our gridding approach (i.e. multi-fidelity regression Kriging), we did not include any advanced quality control (QC) of the crowd-sourced data. This might sound surprising in the context of weather and climate data,but is consistent with approaches in other scientific disciplines (e.g. marine engineering, aerospace engineering).

Results. In this study, we investigate the quantitative effect of state-of-the-art quality control on the accuracy of gridded services, by analysing hourly temperature observations in The Netherlands for the year 2023. Our results indicate that quality control of crowd-sourced weather data can potentially increase the accuracy of straightforward – and commonly used – nearest-neighbour approximation, but generally deteriorates the accuracy of more advanced gridded services. This finding indicates that using strict QC procedures to turn crowd-sourced data into a dataset with similar fidelity as the NMHS-sourced data is not the way to go. Therefore, we do indeed continue to blend first-party data, crowd-sourced data and land-use data without applying any advanced quality control to the crowd-sourced data.

Ecosystem. We emphasize that crowd-sourced weather data only improves services when it is blended with a high-quality network of NMHS data and possibly with land-use data. In other words, the availability of crowd-sourced data does not remove the need for high-quality NMHS observation networks. Therefore, we do not intend to present crowd-sourced data as a stand-alone product; rather, it lives in an integrated ecosystem.

How to cite: de Baar, J. H. S. and van der Schrier, G.: Gridding crowd-sourced weather data: is quality control required?, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-664, https://doi.org/10.5194/ems2024-664, 2024.

Reliable near-surface wind speed  (NSWS; ~10 m above ground level) data is crucial for assessing the impact of wind changes on various socio-economic and environmental sectors, such as wind energy production or risk assessment. Meteorological stations provide local and realistic observations, but their spatial coverage is limited. Although this limitation can be overcome by using classical geostatistical interpolation methods, the reliability of their results is questionable, especially in regions with complex topography. This has motivated the use of reanalyses or dynamical downscaling of simulations as gridded NSWS products that contain local to regional wind data. However, their uncertainties in reproducing observed trends and their coarse resolutions raise doubts about their reliability for reproducing local NSWS. The use of classical interpolation products is even riskier in regions such as the Valencian Community (Eastern Iberian Peninsula, Spain), a region where both local winds (sea breezes) and extreme winds (westerlies “ponientes” or convective wind gusts “downbursts”) occur at local scales (~3km), impacting the tourism activities and wildfires propagation fatalities, among others.

Here, we propose a deep neural network based on partial convolutions as a more reliable spatial interpolation method, capable of learning the wind speed pattern across the Valencian Community observed in a dense observational network.  Observed NSWS from a citizen weather network of up to ~600 stations from the Valencian Association of Meteorology (AVAMET) were used after homogenization, resulting in a high-resolution (3-km) wind speed product. This offers a new tool for both climate and marine research in the framework of the ThinkInAzul project.

How to cite: Plaza, N. P., Martinez-Roig, M., Azorin-Molina, C., Andres-Martin, M., Navarro, J., Gonzalez-Rouco, J. F., Garcia-Bustamante, E., Guijarro, J. A., Piroz, A. A. S., Chen, D., McVicar, T. R., Zeng, Z., and Vicente-Serrano, S. M.: Applying citizen weather data and AI for developing a high-resolution wind speed monitor in the Valencia region (Spain), EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-424, https://doi.org/10.5194/ems2024-424, 2024.

Surface air temperature (t2m) data are essential for understanding climate dynamics and assessing the impacts of climate change. Reanalysis products, which combine observations with retrospective short-range weather forecasts, can provide consistent and comprehensive datasets. ERA5 represents the state-of-the-art in global reanalyses and supplies initial and boundary conditions for higher-resolution regional reanalyses designed to capture finer-scale atmospheric processes. However, these products require validation, especially in complex terrains like Italy.

This study analyzes the capability of different reanalysis products to reproduce t2m fields over Italy during the 1991-2020 period. The analyses encompass ERA5, ERA5-Land, the MEteorological Reanalysis Italian DAtaset (MERIDA), the Copernicus European Regional ReAnalysis (CERRA), and the Very High-Resolution dynamical downscaling of ERA5 REAnalysis over ITaly (VHR-REA_IT).

The validation we conduct pertains to both the spatial distribution of 30-year seasonal and annual normal values and the daily anomaly records. Each reanalysis is compared with observations projected onto its respective grid positions and elevations, overcoming any model bias resulting from an inaccurate representation of the real topography.

Key findings reveal that normal values in reanalyses closely match observational values, with deviations typically below 1°C. However, in the Alps, winter cold biases sometimes exceed 3°C and show a relation with the elevation. Similar deviations occur in the Apennines, Sicily, Sardinia. VHR-REA_IT, on the contrary, presents a warm summer bias of about +3 °C on average over the Po valley. Daily anomalies generally exhibit lower errors, with MERIDA showing the highest accuracy. Moreover, when aggregating daily anomalies to annual time scales, the errors in the anomaly records rapidly decrease to less than 0.5 °C.

The results of this study empower reanalysis users across multiple sectors to gain a more profound insight into the capabilities and constraints of different reanalysis products. This knowledge, in turn, enables them to make well-informed choices when incorporating these products into their research and practical applications.

How to cite: Cavalleri, F., Viterbo, F., Brunetti, M., Bonanno, R., Manara, V., Lussana, C., Lacavalla, M., and Maugeri, M.: Inter-comparison and validation of high-resolution surface air temperature reanalysis fields over Italy, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-67, https://doi.org/10.5194/ems2024-67, 2024.

The Met Office has been generating gridded climate datasets from UK in situ observations for over 25 years. The current version of these gridded data is known as the HadUK-Grid dataset – it includes monthly, seasonal, annual and long-term average values for 11 variables at 1km resolution. Three of these variables are also available at daily resolution. These datasets have a variety of applications including climate monitoring, model calibration and model validation.

Although the software, data sources, file formats, grid resolution and distribution methods have all changed since we first started producing these datasets, the techniques used to create the grids from in situ observations have remained largely the same. The observations are first detrended, either by converting the values to anomalies from the long-term average, or by using regression analysis to model the dependency on geospatial variables such as terrain elevation or proximity to the coast. The anomaly values or regression residuals are then interpolated to the target grid points using inverse-distance weighted averaging.

Here we present a review of the uncertainties in our gridded data. The aims of the analysis are threefold – to update the information we provide to users regarding the quality of our gridded products, to better understand the strengths and weaknesses of our gridding methods, and to investigate the efficiency of the quality control tests in our gridding software.

A leave-one-out cross-validation analysis has been carried out for the majority of our archive of gridded data. Time series graphs will be presented showing how the RMS error varies through the data record for each variable. These graphs show clear trends, seasonal cycles and outliers. Case studies have been used to understand the causes of some of these features.

Based on this analysis we have investigated aspects of our gridding techniques and quality control methods which have the potential to be improved. Preliminary results are presented which show how different gridding methods affect the cross-validation results and we draw some initial conclusions regarding the impact on gridding uncertainties.

How to cite: Hollis, D., Kendon, M., and Carlisle, E.: Investigating the uncertainty in gridded in situ climate data for the UK, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-1062, https://doi.org/10.5194/ems2024-1062, 2024.

We improve the representation of relative humidity fields over Scandinavia by combining outputs from our numerical weather model (MEPS) with real-time weather observations from citizen-managed stations. The method has to work operationally as a part of the automatic forecast at Yr.no, so both computational speed and accuracy are important evaluation metrics.  

The method of choice is Optimal Interpolation (OI), a Bayesian statistical analysis technique. In this approach, the model output serves as the prior information, which is updated by incorporating the more accurate and precise observations from real-time data. OI operates under the assumption that both model predictions and observations have associated uncertainties. By weighing these uncertainties, OI adjusts the model output towards the observations, aiming to minimise the overall error in the final data product. 

A challenge with real-time observations from citizen-managed stations is that they cannot always be trusted, for various reasons. Therefore, we need a method of quality assurance of the observations. A second challenge is to incorporate OI in a practical way in the codebase. For the first challenge we use the titanlib library, developed at the Norwegian Meteorological Institute (MET Norway). This library deals with automatic quality control of weather observations. For the second challenge we use the library gridpp, also developed at MET Norway. 

The quality of the reconstructed relative humidity fields is then evaluated by comparing them against a set of independent observations, which are known to be trusted and accurate. An example would be the weather station at Blindern, outside the headquarters of MET.

The software are openly available on:

Titanlib: https://github.com/metno/titanlib

GridPP: https://github.com/metno/gridpp

How to cite: Ersland, O., Lussana, C., Seierstad, I. A., and Nipen, T. N.: Spatial analysis of 2m relative humidity over Scandinavia, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-770, https://doi.org/10.5194/ems2024-770, 2024.

Gridded data is highly appreciated for many climate related applications due to their spatial information compared to single station locations. The quality and additional value of a gridded dataset is depending on the method itself and on the amount of additional information, e.g. topographical data, used for the interpolation. Furthermore, there is an ongoing discussion on the implications of using gridded datasets of primary climate parameters like temperature or precipitation to derive gridded datasets of climate indices. Within the project SDGHUB (https://www.sdghub.at/), we are exploring the influence of different approaches on the final gridded climate index dataset. 

The focus is on the frequently used climate index of hot days (days with a maximum air temperature above or equal to 30 °C). In the first approach the hot days were computed from the gridded climate dataset SPARTACUS (Hiebl and Frei, 2016), while in the second approach the hot days were directly interpolated considering station values. SPARTACUS is a national gridded dataset of Austria, available on a daily basis with a 1 km-spatial resolution via the DataHub of GeoSphere Austria (https://data.hub.geosphere.at/). It covers the period from 1961 onwards and includes the parameters of maximum, minimum and mean air temperature and rain amount.  

For the second approach the efficacy of several geostatistical interpolation methods, including Kriging with External Drift (KED), Ordinary Kriging (OK), and Regression Kriging (RK), with a particular emphasis on their ability to represent the spatial variability of hot days was explored. As KED allows to include external variables to improve the interpolation results, we decided to proceed with that method. Furthermore, we tested different data transformation methods and variogram models to improve further the results.  As for SPARTACUS, the direct interpolation of hot days was done on a 1 km-scale for Austria.

To evaluate the performance of the two respective approaches a comparison with independent station data was done.

The presentation will include information about the two approaches, details about the direct interpolation of hot days and provide insights into the evaluation and differences of the two datasets.

References

Hiebl J, Frei C (2016) Daily temperature grids for Austria since 1961 – concept, creation and applicability. Theor Appl Climatol 124:161–178. https://doi.org/10.1007/s00704-015-1411-4

Acknowledgement

The project SDGHUB is funded by the Austrian Federal Ministry for Climate Action, Environment, Energy, Mobility and Technology (BMK) via the ICT of the Future Program - FFG No 892212.

How to cite: Tilg, A.-M., Habibi, M., Chimani, B., and Greilinger, M.: Gridded datasets of climate indices: comparison of two approaches, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-567, https://doi.org/10.5194/ems2024-567, 2024.

Observations from local networks of surface stations, managed by local (regional) weather services, are collected in a nation-wide network, providing hourly observations in near real time, accessible through the MyDewetra Database, for the Italian national Civil Protection Department. Such national network provides detailed information on meteorological fields near the surface; it is inhomogeneous though, and affected by Gross Errors and Large Representativeness Errors of various origin. CIMA has undertaken an effort to operationally and automatically check such information for quality, in order to provide a dataset suitable for various uses, such as model verification, data assimilation, meteorological analysis of past events and climatological characterization of recent years and present time.

An important part of the work is devoted to ensure quality of basic station metadata, such as geographical coordinates and orographic elevation above mean sea level. A station archive for each variable is created and monthly checked and updated.

The main checks on observed meteorological variables are based on the idea of spatial consistency. The Spatial Consistency Test (SCT) represents a fine check based on Optimal Interpolation and leave-one-out Cross Validation. This and other tests have been recently included in the open-source library Titan (https://github.com/metno/titanlib).

The Titan library test functions are routinely used at CIMA to provide quality control of precipitation raingauge observations. Such quality control checks for precipitation are presently supervised by a meteorologist, aware of the current weather state.

For temperature data the SCT has actually been rewritten to ensure a detailed analysis of check behaviour on the Italian dataset and is monthly used to provide reliable data for model forecast verification.

Plans are in place to:

• extend the quality control to other meteorological variables, such as relative humidity;

• test the inclusion of existing high resolution observational network managed by private amateur citizens;

• test the impact of such quality-controlled local observations in data assimilation experiments for short-range forecasts;

• compare SCT results with those obtained by Titan library functions, also with the possible scope of contributing to that open-source software test and development.

How to cite: Uboldi, F., Oberto, E., Pellegrini, U., Lagasio, M., and Milelli, M.: Automatic quality control for a nation-wide high-resolution network, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-261, https://doi.org/10.5194/ems2024-261, 2024.

Context. Over the past decades, we have seen important developments in climatological observational gridded data sets. One development is the increasing number of stations included in the gridding process. For any gridded data set, one might ask oneself: (a) How sensitive is the gridded data set to the number of stations? (b) How accurate is the gridded data set? (c) What is the potential improvement of the gridded data set if we add more stations, and does this potential improvement depend on the locally existing station density, the local terrain, etc.?

Approach. We can learn from similar questions in engineering, specifically in the discipline of computational fluid dynamics (CFD). In this field of research, questions (a) and (b) are addressed in a standardized process of ‘verification’ (a) and ‘validation’ (b). The results of a verification and validation study are usually reported in terms of a grid convergence index (GCI) and (cross-)validation results. This approach was standardized by Roache in his work Verification and Validation in Computational Science and Engineering (Hermosa Publishers, 1998). We aim to apply the same procedure to gridded data sets. In this way, our approach is a way of interdisciplinary exchange of methodology.

In addition, we analyze the effects of adding stations by tracking their type of location, terrain, etc. (part of which might also be used as covariates during gridding) when we quantify their effect on the gridded data set. In this way, we can train a simple machine learning model of how sensitive the gridded data set is to inclusion of stations with specific characteristics. The aim is to identify which type of stations, based on their characteristics, would be the most valuable addition to the data set. The last step (c), we name ‘vacancy-profiling’.

Application. As a first study, we apply this approach to the E-OBS gridded data set for daily mean wind speed. This is an interesting data set, since the network density has increased significantly over the years, because the gridding process includes the use of covariate information which gives more details in the verification and validation processes.

How to cite: de Baar, J. H. S., van den Besselaar, E. J. M., and van der Schrier, G.: Verification, Validation and Vacancy-Profiling (VVV-P) as an assessment of quality and network growth potential of gridded climate data sets, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-667, https://doi.org/10.5194/ems2024-667, 2024.