Although much work has been conducted worldwide to discover, rescue and digitise historical weather observations, there remains a lack of access to weather observations from many regions of the world. One such region is Africa which has a scarcity of historical observations available from the pre satellite era which is preventing important research taking place on past extreme weather events and the climate of the continent. The need for this data becomes more urgent when we consider that the workforce in African countries is disproportionately employed in climate-exposed sectors with over 60% employed in the agricultural sector where crops are completely dependent on rainfall. The ACMAD (African Centre of Meteorological Applications for Development) collection offers an opportunity to improve the temporal and spatial data available across the continent of Africa with data available in some countries as far back as the late nineteenth century.          

An inventory has been created of the collection and shows that within it a large amount of unique data exists that has not been previously inventoried by other sources. This includes numerous weather stations that have never been inventoried before. With an estimated four million images within the collection, numerous methods are required to digitise the unique data. The team in ICARUS (Irish Climate Analysis Research UnitS), Maynooth University, launched a project called CliDAR-Africa, where we assisted second year undergraduate geography students to digitise unique data from stations in Madagascar and Guinea. Discussion on the use of AI as a tool to transcribe data have also taken place. Quality issues with the images within the collection also exist and the team have been developing a citizen science project where inferior quality images can be identified. Once identified it is hoped that the quality of these images can be improved at a later date.

In my presentation I will discuss the various issues with the ACMAD collection the ICARUS team are attempting to solve, the results following the inventorying of the collection, efforts to digitise data from Madagascar, proposed projects involving Citizen Science and AI, and finally the important need to rescue African meteorological data in order to improve our knowledge of climate change and extreme weather events in Africa.  

How to cite: Healion, K., Noone, S., and Thorne, P.: Rescuing historical data from the ACMAD collection- The importance for climate research in Africa., EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-342, https://doi.org/10.5194/ems2024-342, 2024.

Using hourly surface wind speed data from 155 stations in Canada, this study first developed a homogenized monthly mean wind speed dataset for the period of 1953-2023, which was then used to characterize observed changes in surface wind speed in Canada. The hourly data were first quality controlled and adjusted for non-standard anemometer heights before being used to calculate monthly mean wind speed series. To identify artificial discontinuities, the monthly mean wind speed series were subject to a semi-automated comprehensive data homogenization procedure, which uses a combination of station metadata and multiple statistical tests with and without using reference series. Reference series used include up to four best significantly-correlated neighbour stations’ data series, the ensemble mean series of monthly wind speed taken from the Twentieth Century Reanalysis version 3 (20CRv3), and monthly mean geostraphic wind speeds derived from homogenized surface pressure data. The results from the automated procedure were then reviewed manually using metadata and visual inspection of the multiphase regression fits with expert judgement. As a result, all the 155 data series were identified to have one or more artificial discontinuities, which were diminished by quantile matching adjustments. Anemometer height change, station joining, relocation, instrument changes/problems were found to be the main causes of data inhomogeneities. The homogenized dataset for 1953-2023 shows wind stilling in region from northern British Columbia (BC) to southern Yukon-Northwest Territories and from southern Prairies to Quebec-Labrador, which was matched with wind strengthening in the region from southern-central BC to the Rocky Mountains, and in Newfoundland and the high Arctics. The trend pattern of in-situ wind speed data bear substantial similarity to that of both the  ERA5 and 20CRv3 reanalysis wind speed data.

How to cite: Wang, X., Feng, Y., Isaac, V., Vincent, L., and Hartwell, M.: On Monthly Mean Surface Wind Speed Data Homogenization and Trend Assessment, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-72, https://doi.org/10.5194/ems2024-72, 2024.

The homogeneity of wind speed and wind gust data is very sensitive to changes in the instrumentation or the environment of the sensor. Time series of wind observations over Catalonia have been analysed, and in spite of the relatively short history of automated observations, several significant inhomogeneity biases were detected and removed.

A dense network of observing stations and documentation of station histories (metadata) helped the homogenization. Only time series with observation periods of at least 10 years were used, resulting in the homogenization of 209 time series for both wind speed and wind gust. The data was originated from three sources, i.e. the primary network of the Catalan Meteorological Service (SMC), the agrometeorological observation network of the SMC and the network of the Spanish Meteorological Agency (AEMET) in Catalonia.

The data underwent basic quality control before homogenization. Homogenization was performed with ACMANT, known to be the most accurate homogenization method based on currently available method comparison test results. The used new version ACMANTv5.2 can take the benefit of metadata either in automatic or interactive mode. Prior to homogenization, a specific examination was performed, in which residual standard deviations of two inhomogeneity models were compared, i.e., the additive model and multiplicative model. The additive model showed clear advantages for wind gust homogenization, while the suitability of the two models appeared to be similar for the homogenization of mean wind speed data. Metadata indicated several technical changes during the observation periods of 24 years on average. Most stations experienced changes in sensors and data logger multiple times, while a few time series were affected by station relocations or changes in the sensor height.

The homogenization process revealed that changes in sensors and data logger often did not cause perceptible inhomogeneities. On the other hand, two network-wide changes in data loggers occurred, in 2005 in the red agrometeorological network and in 2007 in the primary network of Meteocat, when the technical changes impacted significantly and almost synchronously the homogeneity of the data. Since all relative homogenization methods, including ACMANT, presume that inhomogeneities are station-specific, metadata played a crucial role in detecting these problems. The final homogenization was performed in a way that neighbour series affected by the same type inhomogeneity bias as that of the candidate series were excluded from the homogenization.

How to cite: Domonkos, P., Prohom, M., and Cunillera, J.: Homogenization of Daily Wind Speed and Wind Gust Time Series in the Era of Automatic Observations in Catalonia, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-231, https://doi.org/10.5194/ems2024-231, 2024.

In essence the theme of homogenization can be divided into two subgroups, such as monthly and daily data series homogenization. These subjects are in strong connection with each other of course, for example the monthly results can be used for the homogenization of daily data. The earlier versions of our method MASH (Multiple Analysis of Series for Homogenization; Szentimrey) were developed for homogenization of the daily and monthly data series in the mean i.e. the first order moment. The software MASH was developed as an interactive automatic, artificial intelligence (AI) system that simulates the human intelligence and mimics the human analysis on the basis of advanced mathematics. The new version MASHv4.01 is able to homogenize also the standard deviation i.e. the second order moment. We remark if the data are normally distributed (e.g. mean temperature) then the homogenization of mean and standard deviation is sufficient since in case of normal distribution if the first two moments are homogenous then the higher order moments are also homogeneous.

In our presentation, we present the application of the MASH4 software on real climate data. From 1 January 1901 to 31 December 2023, the daily mean temperature data series of 27 stations in Hungary were homogenized in both first and second moments. We present our results based on verification statistics. The developed automatic verification procedure of MASH makes it possible to test the homogenized time series systematically in order to clear up the uncertainty. The basic concept of the verification procedure is that confidence in the homogenized series may be increased by the joint comparative examination of the original and the homogenized series systems. The comparison is based on such adequate questions, which can also be formulated mathematically, consequently, a programmed statistical test procedure can be obtained to evaluate the quality of the homogenized series. The questions are related to the estimated inhomogeneity of the series before and after homogenization the comparison of the modification measure of the series with the estimated inhomogeneity of the original series, the representativity of the given station network in respect of the homogenization, the relation between the estimated inhomogeneities and the metadata.

The research presented in the article was carried out within the framework of the Széchenyi Plan Plus program with the support of the RRF 2.3.1 21 2022 00008 project.

How to cite: Izsák, B. and Szentimrey, T.: Homogenization in mean and standard deviation (MASHv4.01), EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-406, https://doi.org/10.5194/ems2024-406, 2024.

The longest period of meteorological measurements in Hungary is available for the temperature records. A substantial expansion of the number of data series used to homogenize temperatures is currently under way, extending back to the second half of the 19th century. The main motivation of the study is the fact that a long-term database based on high-quality measurements is essential to better understand the regional climate and its changes.

To improve the understanding of climate and its changes requires temporally and spatially representative climate databases. However, measurement conditions change frequently: the relocation of stations, instrument changes, changes in measurement time, changes in environmental conditions can all cause inhomogeneities in the data series, and therefore homogenization is needed.

For homogenization of data series, quality control and filling in the missing values the MASH (Multiple Analysis of Series for Homogenization) procedure (MASHv3.03 software) is used at the Climate Research Department of the HungaroMet Hungarian Meteorological Service. Nowadays, temperature is measured in many more places than, for example, 100-150 years ago, so the homogenization should consist of several steps (i.e. 3 or more MASH systems). For example, we currently use temperature data from 34 stations from 1901 and 55 stations from 1951. These station networks are further extended. Inhomogeneities are estimated using monthly data series. Monthly, seasonal and annual inhomogeneities are harmonized in all MASH systems. Then, we create temporally representative data series using the MASH homogenization procedure.

However, weather stations are not evenly distributed, the station network consists of both densely and sparsely covered subregions. In order to estimate the values of meteorological variables at points where no measurements are available, a spatial interpolation method must be used. Our gridded climate datasets are generated using the MISH (Meteorological Interpolation based on Surface Homogenized Data Basis) method (MISHv1.03 software). The use of the MISH interpolation results in a spatially representative climate database.

In this presentation, the new mean temperature station systems used for homogenization are described together with the most important verification statistics of the homogenization of mean temperature data series, and finally, the gridded spatial means (national averages for Hungary) are analyzed from the mid-19th century to the present.

Acknowledgements:

The research presented was carried out within the framework of the Széchenyi Plan Plus program with the support RRF 2.3.1 21 2022 00008 project.

How to cite: Szentes, O., Lakatos, M., and Pongrácz, R.: Homogenized and gridded mean temperature data series in Hungary from the mid-19th century, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-410, https://doi.org/10.5194/ems2024-410, 2024.

The Global Atmosphere Watch (GAW) Programme of the World Meteorological Organization coordinates a worldwide network of hundreds of ground-based in-situ monitoring stations that provide reliable scientific data on the chemical composition of the atmosphere. In the framework of the GAW Programme, the Quality Assurance/Scientific Activity Centre at Empa has developed an interactive dashboard based on data science to support station operators in timely detecting issues in their in-situ measurements of various trace gases.

The application (GAW-qc), currently in beta testing, makes use of a mixture of purely data-driven and hybrid anomaly detection techniques. It exploits historical measurements made at the target station as well as the archive of gridded numerical forecasts by the Copernicus Atmosphere Monitoring Service (CAMS). The accuracy of the latter for the specific site is improved through machine learning using various predictors, including meteorological parameters and aerosol concentrations.

GAW-qc allows station operators to upload their latest measurements, visualize the data with different temporal aggregations, and easily detect anomalous values using just their internet browser. By combining the information gathered from the dashboard with logbook entries and local expertise, they can effectively flag problematic measurements and even detect instrumental issues that would remain unnoticed otherwise. First case studies indicate that this process can indeed facilitate the detection of malfunctionings in the analytical setup and reduce the ingestion of erroneous data into the international data repositories. Moreover, it has the potential to shorten data gaps if applied timely. Therefore, it may become a game-changer towards reliable, comparable and traceable world-wide datasets in the field of air quality and greenhouse gases. The software is freely available through a GitHub repository and can be adapted to analyze other atmospheric variables.

How to cite: Brugnara, Y., Steinbacher, M., and Emmenegger, L.: GAW-qc: A data science-based dashboard for quality control of atmospheric composition measurements, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-367, https://doi.org/10.5194/ems2024-367, 2024.

Climate normal periods are a common way to describe climate and its changes. With the period 1991 -2020, the first independent climate period after the frequently used reference period of 1961-1990 is available. While, according to the recommendation of WMO, the period of 1961-1990 should still be used as a reference for the long-term climate development, the newer period provides the information for the current climate and is going to be used in many applications regarding also different sectors as agriculture, infrastructure and energy.

The stations for which climate normal period values for 1991-2020 have been calculated were selected depending on their:

• inclusion in regular international data exchange
• use in climate products and services provided by GeoSphere Austria
• inclusion in former homogenization activities
• use in former national climate normal period datasets
• the current state of the measurements of the stations
• completeness in the period 1991-2020.

The homogenization of the selected stations data was done for the daily mean temperature, the daily maximum temperature, the daily minimum temperature and daily precipitation and for an as long period as possible (if possible back to 1961). This was done in order to compare the most current climate period to the 1961-1990 period. For the calculation of both climate normal periods the same criteria of completeness have been used. The method used for homogenization was ACMANT. Finally, around 170 stations where selected for most of the calculated parameters (significantly less stations have been taken into account for snow-depth, radiation and sunshine duration). The homogenization was evaluated by the analyses of breaks and adjustments, by comparison with parallel data as well as by the effect on trends and climate normal values. Additionally the results for the 1991-2020 climate normal period where compared to the results from neighboring countries.

The presentation will cover aspects of the homogenization, its evaluation and the effects of climate change.

How to cite: Chimani, B. and Tilg, A.-M.: Austria data for climate normal period 1991-2020, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-831, https://doi.org/10.5194/ems2024-831, 2024.

It is essential to have validated and trusted records of past climate extremes.

These records are used by planners and policymakers to help them make informed decisions regarding many different sectors - from construction projects to health budgets, from environmental legislation to infrastructure planning, for example.

They are also used to tune and improve climate models, leading to more reliable future projections in a changing climate. Assessing and improving on the abilities of climate models to reproduce these (by definition) rare events, provides a stronger basis from which better informed mitigation and adaptation measures against such potential future climate extremes can be taken.

In this research, we present the comprehensive re-evaluation process of the Irish national maximum air temperature and the records for the months of June, July and August.

For records prior to 1961, we use newly digitised historical climate data from the Met Éireann archives and integrate advanced 20CRv3 sparse-input reanalysis data, station metadata, historical newspaper articles, and contemporaneous references from the examined timeframes. For more recent records, we use climate data from the Met Éireann database and integrate the most recent ECMWF reanalyses products, along with all relevant metadata for our analysis. We also employ time series methods and extreme value theory to help us assess the veracity of the records.

The result of this study will be a list of validated maximum air temperature summer monthly records for Ireland. This process will then be applied to other months and other climate variables in future work.

This research underscores the significance of data rescue efforts in advancing our understanding of past climate extremes, and advocates for continued digitisation and analysis of historical climate data and metadata. By refining national air temperature records through the integration of historical data and advanced reanalysis techniques, the research contributes to a more comprehensive understanding of climate dynamics.

How to cite: O'Sullivan, J., Curley, M., Kelly, C., and McGovern, J.: Reassessing Ireland’s maximum air temperature record value and the monthly maximum air temperature records for June, July and August, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-877, https://doi.org/10.5194/ems2024-877, 2024.

Reliable observational data are essential for robust climate analyses, especially long-term trends.

HOMPRA Europe 2 (HOMogenized PRecipitation Analysis of European in-situ data) is a gridded monthly precipitation dataset based on homogenized time-series. The carefully selected database of more than 5500 stations is a subset of the time series collected by the Global Precipitation Climatology Centre (GPCC). The subset is characterized by few missing values < 20% and intensive quality control.

Compared to its predecessor, the new HOMPRA Europe 2 product covers a longer period from 1951-2015 and is syncronized with the more recent data of the GPCC Full Data Monthly product, so that the data can be extended to the present without a break, although a homogeneous time series cannot be guaranteed for these more recent data.

The actual homogenization process consists of multiple steps: In the first step, for each station, comparable time series are selected. The decision is based on the correlation and Ward's method of minimum variance performed on the deterministic first derivative. For the artificial breakpoint detection, natural variability and natural trends are temporarily removed using the previously selected neighboring time series. This ensures that only artificial changes can be detected. The actual detection applied to annual values is based on the segmentation algorithm of Caussinus and Mestre (2004). In the final step, the breaks detected in the monthly data are corrected using multiple linear regression (Mestre, 2003).

An additional random variation of the neighboring stations shows the robustness of the correction of the individual time series.

Finally, the actual HOMPRA Europe 2 product is created by interpolating the homogenized series onto a 1° grid using the modified spheremap interpolation schemes in operation at the GPCC (Becker et al., 2013 and Schamm et al., 2014).

How to cite: Rustemeier, E., Finger, P., Ziese, M., and Schirmeister, Z.: HOMPRA Europe 2 – An update of a gridded precipitation data set from European homogenized time series, EMS Annual Meeting 2024, Barcelona, Spain, 1–6 Sep 2024, EMS2024-954, https://doi.org/10.5194/ems2024-954, 2024.