Arrival and Peak Abundance of Barn Swallows Hirundo rustica in Three Regions of South Africa in Relation to Climate Indices, Deduced from Bird Atlas Data

1

Reviewer comment

Response

Old text

New text

1

I suggest that the authors improve Figure 1A by using a better cartographic background. The current one is very fuzzy, with even large-scale geographic features (e.g. shape of continents, shape of Caspian Sea etc.) poorly visible. Please use a map with clear features. Figure 1B is fine.

We used better quality background for Figure 1A, as suggested.

2

Line 12: the words 'of arrival' are duplicated, please remove.

Removed

of arrival of arrival

of arrival

3

Line 26: should be 'The first is a methodological one'.

Suggestion implemented

The first is methodological.

The first is a methodological one.

4

Line 82: what do you mean by 'along the eastern borders of Poland and Ukraine'? Eastern borders of Ukraine are located rather far from the eastern borders of Poland. Ukraine is quite a large country, the largest one which is located entirely in Europe.

We meant the border between Poland and Ukraine. The paragraph to which this comment refers has been rewritten, and this has been corrected.

with the border between Central and Eastern Europe at c. 20˚–30˚E, along the eastern borders of Poland and Ukraine (Spina et al. 2022).

with the border between Central and Eastern Europe set along the border between Poland and Ukraine, i.e. at c. 20˚–30˚E (Spina et al. 2022).

5

Line 521: please remove the dot before the word 'midsummer'.

That section of text was moved at the suggestion of another referee

6

Line 588: should be 'The dataset is too limited'.

Suggestion implemented

The size of the dataset is too limited

The dataset is too limited

7

Line 683: please insert a blank between 'MOI1' and 'in'.

Corrected

MOI1in May–August

MOI1 in May–August

8

Lines 690-691: better 'to utilize food resources'. You do not really utilize availability.

Agreed

to utilize food availability

to utilize food resources

2

Reviewer comment

Response

Old text

New text

Major

1

The authors claimed that the model was selected through a forward stepwise procedure. What are the specific criteria for selecting the models? In other words, why were these models retained and ranked in Table 3–8?

We explored the data using a combination of the RSEARCH procedure and the STEP directive in Genstat. The inratio and outratio criteria were both set to the default values of 1.

2

If I understand correctly, the authors constructed models using each of the 98 climate factors as independent variables, then performed forward stepwise regression. How could the authors avoid potential Type I errors when conducting thus multiple regressions?In addition, the authors should at least provide their initial models to demonstrate the effectiveness of their model selection.

This is a penetrating question, and we value the opportunity to comment on it. In our analyses, we were brutally aware of this issue, so we sidestepped it. The concept of Type I errors only is pertinent when we do formal hypothesis testing. In our methods we explained that we were going to use multiple regression as a tool of Exploratory Data Analysis (EDA), with the meaning of EDA developed by that extra-ordinary pioneer statistician John Tukey. We calculated and used AIC and "percentage varience explained " as mere summary statistics (in the same way as the mean, median and standard deviation are summary statistics). With a large number of potential explanatory variables, and a short dataset, our model selection could not be performed in the classical analytical framework.  We had to operate with a descriptive and exploratory framework. We made this clear in our description: "the statistical results have to be treated as exploratory rather than as confirmatory." What astonished us was that the results obtained made impressive biological sense in model after model. With integrity, we strongly feel that we have found some amazing insights which need to be shared with the research community.

3

Line 76–82:Authors should use clear and concise expressions.eg., it should be clear that 100% of the Barn Swallows breeding in East Europe overwinter in Southern Africa, but not the other way around.

We rewrote this paragraph to make it clearer.

Barn Swallows which stay in the non-breeding season in South Africa are not a uniform pan-mixture of populations from the entire breeding range (Burman et al. 2018). Neither do all the birds staying in one area of the non-breeding range come from a single part of the breeding range, as Szép et al. (2006) tried to imply. The connectivity between the breeding grounds and non-breeding grounds, based on ring recoveries, has been estimated for Barn Swallows as moderate, and the transition probabilities, i.e. the likelihood of the birds moving from one area to another, for Barn Swallows staying in southern Africa have been estimated at 100% for Eastern Europe, 91.1% for Northwestern Europe, 75% for Northern Europe, and 11.8% for Central Europe, with the border between Central and Eastern Europe at c. 20˚–30˚E, along the eastern borders of Poland and Ukraine (Spina et al. 2022). For South Africa, the percentage splits between swallows from the western, central and eastern Palearctic (cut lines at 10°E and 60°E; Figure 1A) differed between regions of the country (Figure 1B), and were respectively estimated to be 16:80:4 in Greater Gauteng, 0:46:54 in the Greater Durban area, and 34:35:31 in Greater Cape Town area (Burman et al. 2018).

Barn Swallows which stay in the non-breeding season in South Africa are a mixture of populations from the entire breeding range from Western Europe to the western part of Asia, but this mixture is not uniform and the proportions in which birds from different regions of Europe and western Asia are represented vary across South Africa (Burman et al. 2018). The percentage of Barn Swallows staying in South Africa splits between swallows coming from the western, central and eastern Palearctic (cut lines at 10°E and 60°E; Figure 1A), and differed between regions of the country (Figure 1B), and were respectively estimated to be 16:80:4 in Greater Gauteng, 0:46:54 in the Greater Durban area, and 34:35:31 in Greater Cape Town area (Burman et al. 2018).Neither do all the birds staying in one area of the non-breeding range come from a single part of the breeding range, as Szép et al. (2006) tried to imply. The connectivity between the breeding grounds and non-breeding grounds, based on ring recoveries, has been estimated for Barn Swallows as moderate, and the transition probabilities, i.e. the likelihood of the birds moving from one area to another (Spina et al. 2022). For Barn Swallows breeding in Eastern Europe their likelihood of staying in southern Africa was estimated at 100% , for those from northwestern Europe at 91.1% , for Northern Europe at 75%, and at 11.8% for Central Europe, with the border between Central and Eastern Europe set along the border between Poland and Ukraine, i.e. at c. 20˚–30˚E (Spina et al. 2022).

4

Line 145–199: The authors emphasise the reasons for using the reporting rate to measure the migration process of Barn Swallowsin the Methods section. If this method has been proven effective, then this part should be included in the Introduction section.

This description of reporting rates clearly belongs in Methods, and not the Introduction. Otherwise the Introduction would need a long detoor into the concept of reporting rates. We prefer to leave the ordering as it is. No change

5

Line 361: The author should clearly indicate what the 98 climate variables are, in a table or a supplementary table.

This is now done, the table with all the considered climate variables (corrected to 84) was included in the Appendix as Table A1.

6

Figure 1: As far as I know, the Barn Swallows migrating to Africa utilise more migratory flyways than shown in the Figure, such as the West Asia-East Africa migratory flyway.

We added the West Asia-East Africa migration flyway of Barn Swallows, as suggested

See new Figure 1 in revised paper.

7

Figure 4 & 5: Fitting lines should be added to the figures.

Lines were added to the 4 regions (out of six) which were significant, we updated the figures.

See new Figures 4 and 5 in revised paper

8

Lines 243–337: I suggest organising these descriptions of the climate indices into a table or a box

We moved shortened descriptions of the climate indices into Table 1. Instead of the longer descriptions of each large-scale index, we added a paragraph explaining in which form and from where these indices were obtained. We replaced the links to data sources for each index with the reference number. For now, for convenience, we moved these links at the end of the reference list. In the final version of the paper these links will be placed in relevant places in the reference lists, which will be arranged according to the journal requirements.

See the new Table 1 and the following paragraph on climate indices in the revised paper

9

Lines 388–390: Please provide the standard approaches used to avoid multicollinearity.

We described these standard approaches and provided the relevant reference which recommends these appraoches.

In choosing models for presentation in the Results, we were sensitive to the issue of multicollinearity, and avoided it using standard approaches

In choosing models for presentation in the Results, we were sensitive to the issue of multicollinearity, and avoided it using standard approaches, i.e. we avoided including highly correlated variables (at |r| > 0.7) in one model, and we monitored the Variance Inflation Factors in each model, which should not exceed VIF=10 (Dormann et al, 2013) .

10

Table 3: If the swallows in the Greater Gauteng region do not come from Western Europe, how could the climate factors of Sahel West affect their migration?

The proportion of the Barn Swallows from Western Europe which stay in the Greater Gauteng region were estimated from ringing recoveries at 16% (Burman 2018), and the transition probability of birds from northwestern Europe to migrate to South Africa was estimated at  91.1% (Spina et al. 2022). These proportions were provided in the second paragraph of the Introduction (lines  77-90 in the current msc). We adjusted the Westrn Flyway at Figure 1A to better reflect this direction of arrivals. Thus, among the birds arriving in Greater Gauteng, there is some proportion of Barn Swallows that come from Western Europe, likely along the Western Flyway. This would explain the effect of temperature and precipitation in the Western Sahel on Barn Swallows arriving in Greater Gauteng we found.

See new version of Figure 1A

Minor

1

Line 210: The di do not increase -> The di does not increase

In mathematical English, the d_i are plural, because d_i refers to all the d_i and not just to one of them. If we referred to, say, only d_1, then, in mathematical English I would write, for example, "d_1 is smaller than or equal to d_2"

No change

2

Line 221: We did not calculate the anomalies over the full 215-day period from 1 July to 31 January. We did the arithmetic for the periods in which arrivals are evident in each region ... -> We did not calculate the anomalies over the full 215-day period from 1 July to 31 January; we did the arithmetic for the periods in which arrivals are evident in each region ...

Yes. The reason why we did not use a semi-colon in the first place was to avoid a one sentence paragraph!

Full stop removed

Semi-colon inserted. Words remain the same!

3

Line 496: the arrival iof Barn Swallows -> the arrival of Barn Swallows

Thank you. It surprises me that errors like this sneak in! This is in the caption to Table 4

the arrival iof Barn Swallows

the arrival of Barn Swallows

4

Table 3: Is Prec short for Precipitation?

Yes. This has been searched for and corrected wherever it occurred. Six times in Tables 2, 3 and 4

See Tables 2, 3 and 4

5

What is the difference between season in Table 3 and year in Table 4?

Thank you for picking up the inconsistency. It should be Year throughout

This has been corrected in Tables 3 and 5

See Tables  3 and 5

6

Table 7: Why are the year and annual anomalies negatively correlated yet not significantly related?

Year was the only explanatory variable used in the Model 1 in Table 7. But this model the model explained only 21.9% of variation in the annual model, and was therefore  considered not biologically meaningful (nbm). In all tables, we have removed the indication of the not significant effect (ns),  and replaced it by the abbreviation nbm (not biologically meaningful) because the model selection was traditional concepts of significance.

3

Reviewer comment

Response

Old text

New text

1

The one major comment I have is that I highly recommend that you consider adding discussion of the potential effects of climate change to your well-described migration patterns of Barn Swallows.  That would make this an exceptional manuscript.

We followed this suggestion, and we added a paragraph on the possible effects of the climate change on  Barn Swallow migrations.

In the revised version of the paper we have added a new section.  4.5. The expected effect of the climate change on the migration patterns of Barn Swallows between Eurasia and South Africa

The population size of Barn Swallow breeding in Europe has shown a moderate decline since 1990s (PECMBS 2025). This decline has been attributed to changes in agricultural practices, including increased use of pesticides and decrease in livestock farming, and the related decline of their aerial insect prey, on which Barn Swallows depend at all stages of their migratory life (Grüebler et al. 2010, Orłowski & Karg 2012, Spiller and Dettmers 2019, McClenaghan et al. 2019b). Breeding success and adult survival rates of migratory bird feeding on similar aerial insectivores tend to be lower in years with high NAO values and dry conditions at their non-breeding grounds (Sillett et al. 2000; Stokke et al. 2005; Nebel et al. 2010). High summer NAO is related with hot and dry conditions in northwestern Europe and the western Sahel (Table 1). Such conditions might reduce survival of juveniles due to their heat stress, as reported during heat waves (Toronto Wildlife Centre 2023). Hot and dry weather might limit insect abundance at stopover and staging sites in arid areas of Europe and Africa (Peach, Baillie & Underhill, 1991; Zwarts et al. 2009), thus reduce survival of Barn Swallows during the non-breeding season. Increasing temperatures across Africa and Europe, with more frequent extreme heat waves (NOAA 2025), caused by climate change, thus do not bode well for Barn Swallows. In addition, two El Niño events, in a short sequence, in 2015/2016 and 2023/2024 , reflected with extreme negative SOI, which caused the following summers at both hemispheres the hottest on record (World Meteorological Organisation 2024), and brought catastrophic drought to South Africa in 2016 (Benkenstein et al. 2017), likely has not benefitted Barn Swallows. However, analysis of ringing recoveries of the species showed that Barn Swallows visiting the region of Africa south of ca 9˚S, including South Africa, have shortened their migration distance over 1912–2008 (Ambrosini et al. 2012). If this tendency continues, we can expect fewer Barn Swallows reaching the southernmost destination in South Africa, especially those prone to drought, as the Greater Gauteng area. However, IOD in November-February has also increased since 1980s and was mostly positive in the last five years, which has been related with wet conditions in East Africa (Remisiewicz & Underhill 2022, 2025). The climate change also involves increased precipitation and “greening” in the Sahel since the 1980s, which has positive impact on primary productivity and insect abundance (Wang et al. 2011, Biasutti et al. 2019).  Thus, an increased precipitation when Barn Swallows visit West and East Africa, combined with possible shortening of their migration routes into Africa (Ambrosini et al. 2012), which should reduce costs and risks of migration and enable them to opportunistically use suitable sites, should at least partly counteract the other negative effects of climate change.

2

Page 2 – line 74:  I appreciate the citation to Burman et al. 2018 here but “pan-mixture” is not a common term used in ornithology.  It would be helpful if you provided a brief definition of pan-mixture at this point.

As a result of comments on the paragraph by other referees, we reworded this paragraph. We expanded the description of the "pan-mixture" pattern, avoiding use of this term. See the lines 74-82 of the revised msc.

Barn Swallows which stay in the non-breeding season in South Africa are not a uniform pan-mixture of populations from the entire breeding range (Burman et al. 2018).

3

Page 2 – lines 80-86:  I am not sure this detail is necessary here.  Perhaps it could be summarized and supported with citations.

This is also in the paragraph which was rewritten. See the lines 74-82 of the revised msc.

4

Page 2 – line 88:  What do you base this anticipation on?

We replace the word "anticipate" with "expect". We based our expectations on the mixture of populations of Barn Swallows arriving in South Africa, which we explained in the previous paragraph (lines 74-82 of the revised paper).

5

Page 3 – lines 96-113: Again, I am not sure this detail is necessary here.  Perhaps it could be summarized and supported with citations.

We think that the visual illustration presented in Figure 1 is novel, and needs to be interpreted in the caption to the figure. If the interpretation of the figure gets relegated to citations, its value is lost.The figure should be self-explanatory, which will be lost if the explanations are moved in the text.  The  figure caption will finally get shorter, as the names in references will be all changed to the numbers.  We could probably shorten the figure caption with adding a visual legend below the map at Figure 1A.

Minor changes to shorten the figure caption.

6

Page 4 – lines 117-121:  I think most of this section can be deleted.

We deleted the second sentence in this section.

There is no bird observatory in southern Africa generating bird ringing data comparable to the long-term monitoring datasets generated across Europe and North America (e.g. Dunn 2016, EUFLYNET 2023, Underhill 2025). We have used the data collected at one of these, the Bukowo ringing station, northern Poland; to relate daily catches of migrants with large-scale climate indices (Remisiewicz & Underhill 2020, 2022a, b). To do a comparable analysis in South Africa, we therefore used data from the Second Southern African Bird Atlas Project (SABAP2) (Underhill 2016, Underhill et al. 2017, Brooks et al. 2022, Lee et al. 2022).

There is no bird observatory in southern Africa generating bird ringing data comparable to the long-term monitoring datasets generated across Europe and North America (e.g. Dunn 2016, EUFLYNET 2023, Underhill 2025). To do a comparable analysis in South Africa, we therefore used data from the Second Southern African Bird Atlas Project (SABAP2) (Underhill 2016, Underhill et al. 2017, Brooks et al. 2022, Lee et al. 2022).

7

Page 9 – lines 304-308:  This is a complicated sentence with a complicated message.  Perhaps it could be broken up and simplified.

We split the long sentence into two sentences.

We downloaded both temperature and precipitation as daily values, recalculated as anomalies in relation to the 1981-2022 baseline, based on reanalysed spatial data based on observations from weather models and weather stations, provided by the Climate Explorer by the World Meteorological Organisation Climate through the KNMI Climate Explorer facility (http://climexp.knmi.nl).

We downloaded both temperature and precipitation as daily values; these were provided by the Climate Explorer by the World Meteorological Organisation Climate through the KNMI Climate Explorer facility (http://climexp.knmi.nl). Climate Explorer rerecalculates the daily values as anomalies in relation to the 1981-2022 baseline, based on reanalysed spatial data based on observations from weather models and weather stations.

8

Page 9 – lines 334-37:  Unless there is some interpretation of these observations relative to the development of the specific index, it may be better to locate this in Results.  That is, it seems superfluous to the specific index being developed.

We think that this is useful and important background to understanding the Result, and we would prefer to leave it here.

9

Page 10 – lines 338-359:  This is a very good explanation of how the exploratory variables were selected.

Thank you

10

Page 11 – lines 377-378:  I agree with choosing biologically meaningful models but it would be helpful to provide some insight on how you did this.

The short answer : "We used our training as biologists, and our knowledge on biology and migrations of Barn Swallows"

We aimed to choose biologically meaningful models

We aimed to choose biologically meaningful models, guided by our cumulative eight decades of experience.

11

Page 11 – line 390:  It would be helpful to provide some explanation of the “standard approaches”.

We have added references to the standard approaches for detecting multicollinearity

In choosing models for presentation in the Results, we were sensitive to the issue of multicollinearity, and avoided it using standard approaches

In choosing models for presentation in the Results, we were sensitive to the issue of multicollinearity, and avoided it using standard approaches, i.e. we avoided including highly correlated variables (at |r| > 0.7) in one model, and we monitored the Variance Inflation Factors in each model, which should not exceed VIF=10 (Dormann et al, 2013) .

12

Page 20 – line 601:  I suggest that NDVI be briefly introduced with a supporting citation.

We explained the abbreviation and provided the relevant citation.

High rainfall and low temperatures at the African winter quarters, which are related with high NDVI that reflects high primary production

High rainfall and low temperatures at the African winter quarters, which are related with high Normalized Difference Vegetation Index (NDVI), which reflects high primary production (Gessesse & Melesse 2019, Wen et al. 2012), facilitated early arrival of Barn Swallows to their breeding colonies in Italy, likely through positive effect on their condition (Saino et al. 2004).

13

Page 22 – line 721:  Check the wording in this line (…none or for…).

wording error

and none or for western Asia

and none for western Asia

14

Page 23 – lines 760-764:  This is another complicated sentence that may be modified and simplified.

Yes! Rewritten. In the text alongside, the whole paragraph is included, otherwise the re-write does not make sense!

This analysis made use of bird atlas data. It is essentially an extension and development of the concepts developed in Underhill et al (1992) to describe bird migration using bird atlas data, using the concept of reporting rates pioneered over the period 1982–86 by the bird atlas of in the Western Cape (Hockey et al. 1989) and adopted by the First Southern African Bird Atlas Project (SABAP1) (Harrison et al. 1997, Harrison et al. 2008).

This analysis made use of bird atlas data. It uses the concept of reporting rates pioneered over the period 1982–86 by the bird atlas of in the Western Cape (Hockey et al. 1989) and adopted by the First Southern African Bird Atlas Project (SABAP1) (Harrison et al. 1997, Harrison et al. 2008). We extend and develop the methods used in Underhill et al. (1992) to describe bird migration by using the bird atlas reporting rates on an annual basis.

15

Page 24 – line 795:  This is an incomplete sentence.

Removed