Thousands of Handwritten Records Recovered From Archives Reveal The True Scale of Warming in The Congo : ScienceAlert (opens in a new tab)

In plain English

Rescue and digitization of paper meteorological records from the Democratic Republic of the Congo: two field campaigns imaged 9,885 paper sheets from 37 weather stations and—via transcription—produced a daily observational dataset of temperature and precipitation exceeding 1 million observations.

Key findings

• Field campaigns imaged 9,885 paper meteorological sheets from 37 weather stations in the Democratic Republic of the Congo, creating an imaged archival resource.
• Transcription of the imaged sheets produced a digitized daily observational dataset exceeding one million temperature and precipitation observations.

In plain English

The authors digitized 9,885 paper sheets from local archives (two field campaigns) and used an improved MeteoSaver software (v1.1) to transcribe the images into structured daily station time series. The transcription output, after quality control, comprises daily maximum, minimum, and average temperatures, precipitation, and dry- and wet-bulb temperature observations measured at 06:00, 15:00, and 18:00 local time, producing over one million observations from 37 stations in the Democratic Republic of the Congo suitable for subsequent climate trend and extremes analyses.

Key findings

• Application of an improved MeteoSaver software (v1.1) transcribed imaged paper sheets into structured daily station time series.
• Digitization effort produced over one million temperature and precipitation observations from 37 weather stations based on 9,885 imaged sheets collected during two field campaigns.

In plain English

The authors report a distinct quality-control step applied to transcribed daily station observations prior to climate trend and extremes analysis. Quality control is described as occurring after digitization and transcription (using MeteoSaver v1.1) and before analysis; specific QC rules or procedures are not detailed in the abstract. After QC the authors derived daily time series (max/min/avg temperature, precipitation, dry/wet bulb temps at 06:00, 15:00, 18:00) for 37 stations, of which 21 stations had sufficient temperature data for trend analysis and 3 stations had sufficient precipitation records for event analysis.

Key findings

• A quality-control step was applied to the transcribed observations and yielded daily station time series used for climate trend and extremes analyses; after QC there were 37 station series overall with 21 stations meeting sufficiency criteria for temperature trend analysis and 3 for precipitation event analysis.

In plain English

Using rescued and digitized daily station temperature records, the authors compute decadal frequencies of hot and cold extremes and report an increase in hot extremes and a decrease in cold extremes across successive decades; the most recent decade shows nearly twice as many hot days per year compared to the earliest decade.

Key findings

• Frequency of hot extremes increased and frequency of cold extremes decreased across successive decades in the rescued station dataset.Nearly 2× increase in hot days per year in the most recent decade versus the earliest decade; decrease in cold days reported but magnitude not specified.

In plain English

The paper reports station-based analyses of precipitation extremes using rescued daily weather records. Precipitation-change assessment was restricted to three stations with sufficient precipitation data; at two of these stations the frequency of heavy-precipitation events increased, while at the third no substantial change was detected. The authors note that limited spatial coverage due to data availability prevents drawing broader conclusions for the Democratic Republic of the Congo.

Key findings

• Analysis of precipitation at three stations with sufficient data indicates an increased frequency of heavy precipitation events at two stations and no substantial change at the third.