Atmospheric circulation reconstructions over the Earth ( <scp>ACRE</scp> ) 2025: climate data recovery

On 15-16 September 2025, Singapore Management University hosted the annual Atmospheric Circulation Reconstructions over the Earth (ACRE, www.met-acre.net/) workshop, ’ACRE 2025: Climate Data Recovery’, chaired by Fiona Williamson, Rob Allan and Praveen Teleti. The 2-day event took place at the College of Integrative Studies and saw around 30 participants from multi-disciplinary backgrounds all working in meteorological data rescue (DaRe, www.community.wmo.int/en/data-rescue-projects-and-initiatives-dare) sharing their latest work and discoveries. Speakers were a mix of climate scientists and historians, archivists and historical climatologists from across the globe. ACRE has been operating for 18 years under Professor Rob Allan, formerly of the UK Meteorological Office Hadley Centre. Over these two decades, ACRE has evolved as a leader in the data rescue space, facilitating the recovery of millions of data points for historic weather for global data repositories, including the International Surface Pressure Database (ISPD, www.reanalyses.org/observations/international-surface-pressure-databank), the Twentieth-Century Reanalysis (www.psl.noaa.gov/data/20thC_Rean/) and ERA-CLIM (www.ecmwf.int/en/research/projects/era-clim). The workshop began with a keynote by Rob Allan, who gave an update on current projects and partners under the ACRE umbrella. He stressed the importance of continuing data rescue efforts, open discussion and bringing together disparate projects from across the globe. The overview that follows is not in order of presentation. Most sessions featured speakers working on recovering data, tools for recovering data and uses of recovered data in climate reconstructions. Cary J. Mock (University of South Carolina) began proceedings with an overview of his intensive work locating and reconstructing northwestern Pacific historical typhoon records from ship log books over a span of several hundred years, before Elaine Kuan-Hui Lin (Academica Sinica) demonstrated how the recent REACHES Database (www.reaches.rcec.sinica.edu.tw/) could be used to reconstruct historical tropical cyclone landfall on the China coastal region. Several of the talks profiled ways to get weather data directly from archival sources that might not be obvious. Jun Matsumoto (Tokyo Metropolitan University), for example, spoke about extracting climate knowledge from diaries, which included numerical data but also narrative information, and Qing Pei (Hong Kong Polytechnic University) explored how climate information can be extracted from qualitative sources such as 500 years of northeast Chinese history using content analysis, followed by temporal and statistical analysis. Economic historian Luisa Piccinno (University of Genoa) gave a fascinating introduction to how historical maritime accidents can provide insights into sources of weather information, in a study of risk and accident reporting at sea. Togo Tsukahara (Kobe University) considered the value of nineteenth-century Dutch and Japanese naval archives for the extraction of data for climate analysis, and Masumi Zaiki (Seikei University) took a deep dive into analogue weather information recorded by automatic meteorological recording instruments in the early twentieth century. Hisayuki Kubota (Hokkaido University) provided insight into natural climate variability within the context of anthropogenic climate change in Japan, as well as new sources of weather and typhoon observations for Japan and East Asia. Looking at changes in magnitude and frequency of droughts (since 1877) and large daily rainstorms (since 1906) in Sabah (Malaysian Borneo), Rory Walshe (Swansea University) explored changes in drought and rainfall events unearthed by a collaborative team of Malaysian and British scholars. The research demonstrated significant changes since 2000 and some possible linkages to ENSO. However, different meteorological events and locations have different patterns across eastern Malaysia, making it hard to generalise. Also looking at trends, Carla Mateus (Maynooth University) gave an overview of cold and hot spells in Ireland, garnered from over 750 historical collections, including weather diaries, meteorological publications, newspapers: all of which are available in an open access database containing instrumental data from 1885 and documentary evidence from the 1600s (Mateus, 2021). Some of this data collection involved student–scientist service-learning collaboration to enable fast, efficient data keying (Mateus et al., 2021). Several other presentations also focused on how to manage or improve the arduous task of extracting and digitising weather information from the millions of available archival sources. Praveen Teleti (Singapore Management University) spoke about a citizen science project hosted by Zooniverse – Monsoon Voyages (www.zooniverse.org/projects/p-teleti/monsoon-voyages) – which seeks to recover weather data points from nineteenth-century British Navy ship logs and Veronica Manara (University of Milan) profiled the Cli-DaRe project to recover Italian meteorological records, also through citizen science (Manara et al., 2025). Philip Brohan (UK Meteorological Office) however put forward a convincing case for increasing the speed and decreasing the cost at which weather information can be extracted and digitised by using machine learning (www.brohan.org/Proxy_20CR/), rather than citizen science, for some data types, which he light-heartedly termed ‘Robot Rainfall Rescue’. Several participants presented on current, or just completed, data rescue projects. Martin Stendel (Danish Meteorological Institute) spoke about the ROPEWALK project (www.en.rigsarkivet.dk/about-us/collaborations-and-projects/ropewalk-turning-historical-weather-logs-into-climate-data/), which recovers weather observations from Danish ship journals and logbooks stored in the Danish National Archive. This especially includes trade ships that travelled between Denmark, Norway, Iceland and Greenland, and is funded by the Danish Meteorological Institute and the Danish National Archives. Alessandro Ceppi (DICA, Politecnico do Milano) gave a presentation on some incredible work rescuing and digitising climate data from countries which have hitherto had little attention, including Ethiopia, Libya, Somalia, Albania and Egypt. In total, more than 40000 images of weather observations were scanned to produce datasets which are now available publicly (www.drive.google.com/drive/u/0/folders/1d8toQW6qDSodvZ4eRIgV3iq0ffLFdWk6). Also looking at more data sparse regions (or regions enveloped in the ‘fog of ignorance’, signalling where limited data recovery work has been undertaken), Monichoth Im So (Department of Meteorology, Cambodia) gave an excellent keynote on the hard copy data that his agency has stored from the French colonial era and how difficult it can be to translate this paper-based information into usable formats. Urip Haryoko (BMKG) raised similar concerns in profiling the Indonesian meteorological service’s data rescue efforts (www.didah.org/) from 2018, which began with a shed-like structure piled to the ceiling with paper, and ended with a large quantity of this now imaged and digitised. Nur Sherizan Binti Darus (MetMalaysia) profiled their ongoing recent efforts to digitise and manage historical climate records as a scientific asset and part of Malaysian national heritage. Similar to the Cambodian experience, Darus noted challenges inherent in working with data in obsolete formats, especially microfilm, which is not easily converted to modern digital systems. An interesting component of the workshop demonstrated how climate scientists and services used this data once it had been extracted. Gil Compo (University of Colorado) showed how data rescue activities had consistently improved the NOAA-CIRES-DOE Twentieth Century Reanalysis project, a four-dimensional global atmospheric dataset of weather spanning 1806 to 2015. Later, Compo gave an invaluable primer on how data rescuers can submit their data to have it processed and assimilated into global public access databases (www.climate.copernicus.eu/global-land-and-marine-observations-database). Dale Barker (Centre for Climate Research Singapore) introduced a new project for Maritime Continent Regional Reanalysis (MCARR) run jointly with A* Agency for Science, Technology and Research Singapore which will result in improved regional reanalysis for the 2000s. A meeting highlight was a workshop by Catherine Ross, chief archivist at the UK Meteorological Office, who sought to gather information from the scientists in the audience to better inform the archival community about what resources would be helpful to them in data rescue activities. For example, as consumers of historic data, how could archive catalogues and databases be improved to direct users to relevant information? This was an important session in understanding the pain points or constraints in the process of DaRE. Indeed, many of the projects profiled over the course of the 2 days showed how data recovery is not simply the process of finding numerical data from obvious meteorological logs, but can come from information contained more broadly in the archives of societies, such as diaries, medical records, botanical records and so on - which are not immediately obvious to the researcher. The workshop ended as it began, with a wrap-up and thank you by ACRE lead Rob Allan (Figure 1). There remains much work to do on data rescue, not least because the historical archive is scattered and funds tend to focus on outcomes, rather than inputs. That is, focus is often on the uses of data in modelling, not on the hunt for and extraction of data in the first instance. Much remains hidden in hard copy records, which tend to be lost over time or to be stored in unfavourable conditions resulting in loss of integrity. Efforts improve as technology improves, but this technology is not evenly accessible across countries and organisations. Data rescue is not just a job. The people involved are personally invested and passionate about the subject and much of the work is driven by volunteer activities, personal research projects and individual efforts to keep the momentum going. Do join us in making climate data accessible and freely available by joining the ACRE community, or in helping in any of the extant citizen science digitisation projects mentioned above, such as Monsoon Voyages. This conference was funded by a grant awarded by the Ministry of Education, Singapore, under MOET2EP40123-0021: Building a Deeper Long-Term Climate Understanding of Rainfall Trends in Singapore, West Malaysia and Northern Sumatra.