Review Reports - Changes in the Occurrence of Late Spring Frost in Poland

Round 1

Reviewer 1 Report

The paper describes trends in the occurrence of late frost in Poland over the last 60 years using station temperature data from 6 locations. The authors use two sets of indices (soil and air frost) and measure occurrence first in day-of-year and then relative to the beginning of the growing season. They find a clear anticipation of the date of the last frost, but only small and mostly non-significant trends were found when expressing this relative to the start of the growing season. They then describe the change in probabilities.

The paper is rather straight forward; I found it interesting, although results are not all that surprising. I have some minor revisions, which I list below. My only major recommendation is to weight down the analysis pf probabilities which now occupies 11 Figures (each with 2 panels), which is however based on a relatively small sample (6 stations, 30 years).

L. 24-25. This definition is unusual in that it requires average temperature to be positive and in that it does not state the measurement height (2 m?, 5 cm?). Is there a reference? Later in the paper I do not see the requirement for daily average temperature to be positive.
The first two paragraphs of the introduction should also be referenced.
l. 78: "mild frost - Tmin <0°C"? Is this correct or should it be Tmin <0°C and Tmin >= -1°C ?
l. 83: representative / adequate -> rephrase
l. 84-87: This is confusing as the growing season is not yet defined. Please rephrase, e.g., that you do not just want the last day-of-year with frost, but also want to express that day-of-year relative to the beginning of the growing season in that specific years. For that you need a definition of growing season etc.
At that point it wold be good to include a clear terminology, or use of symbols.
l. 98: And so/then -> rephrase
l. 104: Not sure what you mean hear, it sounds rather complicated but all you do is to look not just at the mean of 1961-1990 but also at the distribution. I don't think that more needs to be said. Or did I get this wrong?
l. 116: What are "Air frosts"? (not defined above)
l. 135: It would be interesting to also plot the trends in the start of the growing season.
Concerning the discussion, I recommend not to put too much emphasis on the probability distributions, as they are based on a very limited sample, but to stick more to the discussion of the central tendency.

Figures 7 and 8: Are the captions confounded (day-of-year versus day since growing season start)? How can the day-of-year be negative?

Literature: This might perhaps be relevant: https://doi.org/10.3390/atmos11040391

Author Response

We would like to thank the reviewer for all their comments. We considered all of them and tried to react in the revised version of our paper. We are confident that these actions have helped to improve our paper, especially in terms of clarity of presentation.

Comments to the author/ Author answers

My only major recommendation is to weight down the analysis pf probabilities which now occupies 11 Figures (each with 2 panels), which is however based on a relatively small sample (6 stations, 30 years).

We tried to present all types of curves, hence such a large number of figures. However, we agree that some curves may result only from the specific location of a particular weather station. Therefore, we finally reduced the number of figures

24-25. This definition is unusual in that it requires average temperature to be positive and in that it does not state the measurement height (2 m?, 5 cm?). Is there a reference? Later in the paper I do not see the requirement for daily average temperature to be positive.

The first two paragraphs of the introduction should also be referenced.

There was an error in the definition, a mental shortcut. Obviously, the key variable for frost is the minimum temperature. Its occurrence defines the appearance of frost. Nevertheless, it can happen even if the average daily temperature is above 0 and / or the maximum temperature is positive. Some sentences were re-written.

78: "mild frost - Tmin <0°C"? Is this correct or should it be Tmin <0°C and Tmin >= -1°C ?

We did not consider the temperature ranges, but the cases of exceeding certain temperature threshold values. The selected thresholds refer to the divisions into the severity of frost used by other authors. The relevant paragraph of the text was re-written.

83: representative / adequate -> rephrase It was done.
84-87: This is confusing as the growing season is not yet defined. Please rephrase, e.g., that you do not just want the last day-of-year with frost, but also want to express that day-of-year relative to the beginning of the growing season in that specific years. For that you need a definition of growing season etc.

At that point it wold be good to include a clear terminology, or use of symbols.

In our opinion, the order of description assumed in this chapter seems to be correct. First, we list the time series of the data to be used, and then we define the variables under study.

98: And so/then -> rephrase It was done.
104: Not sure what you mean hear, it sounds rather complicated but all you do is to look not just at the mean of 1961-1990 but also at the distribution. I don't think that more needs to be said. Or did I get this wrong?

It may sound a bit complicated, but we decided on such a methodology and wanted to introduce it briefly.

116: What are "Air frosts"? (not defined above)

„Air frosts” concern frosts at a height of 2m. The appropriate entry in the Data section was added.

135: It would be interesting to also plot the trends in the start of the growing season.

True, this is an interesting point. However, in this paper, the beginning of the growing season serves only as a reference point for determining the date of the last spring frost. Discussing the characteristics of the growing season, including trend detection for the beginning of the growing season, goes beyond the research questions of this study.

Concerning the discussion, I recommend not to put too much emphasis on the probability distributions, as they are based on a very limited sample, but to stick more to the discussion of the central tendency.

In the Discussion section, we added statements of a more general nature.

Figures 7 and 8: Are the captions confounded (day-of-year versus day since growing season start)? How can the day-of-year be negative?

Incorrect figure descriptions/titles were corrected.

Literature: This might perhaps be relevant: https://doi.org/10.3390/atmos11040391.

Thank you for deliver an interesting paper. It was mentioned in the Introduction section.

Reviewer 2 Report

The research in the article "Changes in the occurrence of late spring frost in Poland" concerns the change in the occurrence of late spring frosts in six different spots in Poland. The assessment of the risk of late frosts is an important issue, especially for the Tmin <-4 ° C threshold. This type of spring frost causes the greatest economic losses.

The assessment took into account changes in the length of the growing season, which is important and essential to assess the real risk of losses caused by spring frosts in agriculture.

1.1 Introduction

The statements “In general, at the beginning of vegetation, the frost resistance of plants is very low” (line 31) are imprecise. I would propose to refine this description, e.g. the sensitivity of plants to spring frost increases with their development and the most critical moment is their flowering phase.

1.2 The occurrence of frosts is most often caused by the influx of Arctic air through a cloudless sky. I would suggest to complete this information.

The choice of methodology is detailed and correct.

3.1 The results are presented in a clear and legible way, they enable comparison and evaluation of change trends. There are some inaccuracies in the description of the results (lines 152-153) with the values given in Table 4.

3.2. In the chapter "Changes in the probability of late spring frosts occurrence", the presented results are difficult to trace, because only selected stations are presented. The charts require an unambiguous description, e.g. (line195) Figure 6. The probability curves of the last spring frost event for consecutive day in the calendar year…… Should the description read The probability curves of the last spring frost event for consecutive day of vegetation ?

the authors ought to present the results of all analyzed stations in a more transparent manner.

- Normalize (unify?) charts (e.g. scales)

- Vary the description of charts (e.g. day number or growing day number),

- For different thermal thresholds, use e.g. different colors

- Moreover, the work requires checking the compliance of the drawings with their descriptions.

Such changes would enable readers to compare the results more easily.

4. References need correction (line 350-370. Position numbers 20-29, (moved by 1).