The article presents a recently discovered series of meteorological measurements made in Wrocław (SW Poland) in 1773–81 comprising daily observations of air temperature and atmospheric pressure, precipitation, wind force and direction.. The analysis focuses on air temperature extending the regular temperature series for Wrocław back to 1773. the scientific methods and assumptions valid and clearly outlined? The paper is clearly structured and the figures are of high quality. As a result, the Wroclaw series will be the longest temperature series in Poland.

 The article should be published with minor revisions.

1. From a climate history perspective, one would have liked to see a more in-depth analysis of the German-language source. It is likely that the Wroclaw observations were inspired by the example of the Bern Economic Society, founded in 1759. In 1760, the society implemented a meteorological observation programme. Specifically, this involved measurements of air pressure, air temperature using mercury thermometers with Rèaumur scale and precipitation measurements with rain gauges manufactured at the expense of the Society. Results were published in the journal of the society, including detailed information regarding the instruments used. (see Christian Pfister, Agrarkonjunktur und Witterungsverlauf im westlichen Schweizer Mittelland 1755–1797. Bern 1975. Geographica Bernensia, Reihe G, 2, p.24-26,) BORIS Portal funktioniert erst wieder ab 18.00).
2. Line 22 specify newspaper “published every other day”
3. Line 246 and 247 give the year

Dear authors,

Please find my comments attached.

In this paper, the authors describe a newly recovered series of temperature observations from a site in Wrocław covering the period 1773–1781. Such series are extremely valuable because—despite the inherent limitations of early instrumental observations—they provide data at a much higher temporal resolution than is achievable with most proxy records. In particular, they capture the full seasonal cycle, whereas many proxy series are restricted to specific parts of the year (for example, growing-season temperatures).

The period studied is also of particular interest, as it coincides with a time when the competing influences of solar variability and volcanic forcing remain poorly constrained, and which broadly corresponds to a phase of relatively cooler average temperatures.

The temperature observations presented appear to be highly reliable and to have been recorded using a responsive instrument. I was pleased to see the comparison with temperature estimates from the ModE-RA reanalysis, and I fully agree with the authors that assimilation of these data into palaeoreanalysis products would further enhance their value, particularly if combined with the construction of a longer regional temperature series.

The paper primarily consists of a description of the data and an evaluation of the observations against other available series. As such, it represents an incremental contribution; nevertheless, it is a valuable one and fits well within the remit of the journal.

Specific comments

Line 21: “atmospheric” is not needed before “precipitation”.

Line 44: The definition of the aims of historical climatology is too limited. For example, Brázdil et al. (2005, p. 366, doi:10.1007/s10584-005-5924-1) describe three aims of historical climatology:

(i) reconstructing temporal and spatial patterns of weather and climate, as well as climate-related natural disasters, for the period prior to the establishment of national meteorological networks (mainly over the last millennium);

(ii) investigating the vulnerability of past societies and economies to climate variability, climate extremes, and natural disasters;

(iii) exploring past discourses and social representations of climate.

Line 160: It would be useful to place the link to the data in a more prominent location, for example in a dedicated section at the end of the paper titled “Data Availability”, with a brief description of the repository.

Line 142: Change “exposition” to “exposure”.

Figure 4: The morning and Tmin curves are shown in the same colour; please revise the figure to allow clear visual distinction.

Figure 9: Please invert the sign of the difference so that, for example, the grey bars represent positive values.

Around page 15: Please include a time series (1773–1781) showing the difference between the ModE-RA series representative of Wrocław and the newly recovered temperature series. This could be included as a supplementary figure if space is limited.

Line 404: Please explain why the reconstructed temperature series should be expected to outperform the palaeoreanalysis.

The paper presents a newly discovered, historical meteorological record from Wroclaw in Poland, covering the years 1773-1781. It is a relevant addition to the existing record in Wroclaw, which began in 1791. Because the gap between the two records can be filled with data from nearby Zagan (1781-1792), a long record spanning more than 250 years ca be generated. 

The paper is describes the record and some analyses. It is relevant and within the scope of the journal. However, I have some suggestion that I would like the authors to take in to account.

Quality Control: The authors should elaborate more on how they perform the quality control. A number of standard checks are available: Range checks, checks of large steps, comparing morning, noon and evening measurements, checking for constant values. The series could and should be compared with those of Berlin and Prag at a daily scale. After 1779, outliers could even be compared against Warsaw, although the distance is large. 

Is the series homogenised? I don't really see this described. But there is the sentence "The correction of historical mean monthly temperature data in Wrocław was necessary because the Prague series was homogenised, which means that the historical data were also corrected to be comparable to present-day data." So how were the data "corrected", what was the reference, which tests were applied? I think the authors should refrain from making statements about the temperature difference to more recent periods. 

The paper says a lot about temperature, though very desceiptive, but very little about the other measurements I think pressure would be very valuable, and also precipitation (which was more rarely measured) and humidity would deserve more description. Conversely, I don't see much value in tabulating means.

The comparison with ModE-RA is best performed on the scale of monthly-to-decadal variability. Longer term trends and changes of longer periods in this data set are from model simulations and not from observations.  

Minor:

L. 28: I think comparisons over long time periods are difficult since the series is not homogenised.

In the Introduction, perhaps give a somewhat broader context on enlightenment science activities in Poland ("Patriotische Gesellschaft in Schlesien").

L. 156: Weighted averages: How calculated?