| 21 Sep 2026
Revisiting the daily weather information in the Guan Tingfen Diary (Haining, China, 1815–1865)
Abstract. Historical weather diaries provide valuable high-resolution information for investigating past weather and climate variability before instrumental observations. Here, we digitize and systematically categorize weather- and climate-related information from the Guan Tingfen Diary, a nearly continuous daily record from Haining, eastern China, spanning 1815–1865. The diary documents precipitation and sky conditions, together with wind, perceptions on cold/warm, hydrological conditions, unusual phenomena, and societal responses to weather and climate. Descriptive records were converted into standardized categories while retaining their original information as much as possible. Comparisons with other climate datasets show that the diary captures meaningful climatic variability. The high temporal resolution of the diary provides detailed insights into the intra-seasonal evolution of extreme climate events that cannot be resolved by conventional seasonal reconstructions. The diary also contains unusual weather and atmospheric phenomena that may provide additional evidence for investigating the climatic impacts of major volcanic eruptions. These results highlight the value of the Guan Tingfen Diary as a high-resolution documentary climate record and its potential for investigating historical climate variability and extremes and reconstructing daily weather in the early nineteenth century.
The manuscript “Revisiting the daily weather information in the Guan Tingfen Diary (Haining, China, 1815–1865)” presents the digitization, categorization, and analysis of weather and climate information contained in the Guan Tingfen Diary, providing a valuable historical climate dataset consisting of a nearly continuous daily documentary record from eastern China spanning 1815–1865.
The manuscript is generally well written, clearly structured, and based on an impressive effort of data recovery and processing. The resulting dataset has considerable potential value for historical climatology and documentary climate research. Before publication in Climate of the Past, however, I believe several aspects could be further strengthened to enhance the scientific impact and robustness of the study. In particular, the manuscript would benefit from a clearer discussion of its novelty relative to previous studies, a more quantitative evaluation of the derived indices, and a more cautious interpretation of the volcanic signal.
i) Novelty and added value of the dataset
The manuscript notes that the Guan Tingfen Diary has already been used in previous studies for precipitation and temperature reconstructions, while the present work revisits the source through systematic digitization and categorization. Although the value of this effort is evident, the manuscript could more clearly highlight the additional scientific advances enabled by the present study beyond the creation of a digital database.
In particular, it would be helpful if the authors could elaborate on: which variables or categories have been digitized and analysed here for the first time; what new climatic applications or research opportunities are made possible by this work; how the present dataset and its associated analyses differ from, and extend, previous reconstructions derived from the same diary.
A clearer discussion of these aspects would help readers better appreciate the contribution of the study.
ii) Quantitative evaluation of the derived indices
The manuscript includes several comparisons with instrumental or reanalysis-based products, which are useful and generally encouraging. However, most of these comparisons remain qualitative.
For example, rainfall variability is compared visually with RAP and 20CRv3 datasets, and wind-direction distributions are compared with ERA5 climatology. Where feasible, the manuscript could be strengthened by including simple quantitative measures of agreement, such as rank correlations (e.g., Spearman coefficients), particularly in cases where more conventional correlation metrics may not be appropriate.
In addition, providing uncertainty estimates or confidence intervals, when possible, would help readers assess the robustness of the reported relationships.
iii) Interpretation of volcanic signals
Section 5.1 addresses an interesting topic and explores potentially valuable evidence of volcanic impacts in documentary records. The discussion of red moons, unusual solar coloration, and anomalous summer cooling following major eruptions is certainly intriguing.
At the same time, some caution may be warranted in the interpretation of these results. The number of available observations is relatively limited, and the superposed epoch analysis is based on a small sample of eruption events. Moreover, some of the reported phenomena may also arise from factors other than volcanic forcing.
For these reasons, I would encourage the authors to adopt slightly more cautious language throughout this section and to emphasize the exploratory nature of some of the findings. If feasible, additional information on uncertainty, such as confidence intervals or significance estimates for the superposed epoch analysis, would further strengthen this part of the manuscript. Likewise, the figure showing the volcanic composite response could benefit from the inclusion of uncertainty bands.
Overall assessment
Overall, this manuscript presents a valuable and potentially important documentary climate dataset. The digitization effort is substantial, the data appear carefully processed, and the examples of drought and cold-event reconstruction effectively illustrate the scientific value of high-resolution documentary records.
With a clearer demonstration of novelty, additional quantitative validation, and a somewhat more cautious discussion of the volcanic interpretation, the manuscript would make a stronger contribution to the literature.
I therefore recommend major revision, while emphasizing that the suggested revisions are aimed primarily at strengthening the presentation and interpretation of an already valuable dataset and analysis before the manuscript can be considered for publication in Climate of the Past.