Context-Dependence of Urban Forest Vegetation Invasion Level and Alien Species’ Ecological Success
Round 1
Reviewer 1 Report
General comments
The manuscript presents an interesting study regarding forest invasion by alien species in the area of Poznań in Poland.
I’m concerned with the number of hypotheses that are claimed to be investigated. Testing specific ecological hypotheses needs specific experimental or field designs and data collection approaches, while this studies uses data derived from previous research projects (that therefore would have a sampling design appropriate for the previous specific aims). I do not think the study can test all these hypotheses but rather can use these hypotheses to explain the observed results. In fact, for example, some variables can indicate more than one hypothesis (e.g., from your table 1: EIV for environmental heterogeneity and habitat filtering). I suggest focusing on your main question and result: number and cover of alien species to change with vegetation types. It would be interesting to see if similar results are maintained if only neophytes are used.
Furthermore, I think that variables may be correlated and this could be stated in the discussion. For example, ecological indicators (i.e. soil fertility and moisture) may play a role in the presence of different associations. Therefore, not really making it possible to distinguish between habitat filtering and the effect of specific tree species. In addition, variability of these ecological indicators may be more relevant in testing environmental heterogeneity rather than FDis. I also think that certain associations are more linked to disturbed conditions (e.g. Chelidonio-Robinietum, Salicetum-albae, Populetum albae) and this may have played a role to.
I also suggest to compare these results with literature reporting studies and comparison between alien and native species in forests within urban areas (please see specific comment). Finally, I suggest modifying the title. I would not to use a question (“Is urban forest vegetation invisibility context-dependent?”), but rather a main and clear message from your results.
Specific comments
L14: There are a number of studies investigating this aspect (see special issue in Biological invasions “Non-native species in urban environments: Patterns, processes, impacts and challenges” and references in the different articles). I suggest deleting this sentence. I would rather highlight that a number of reasons have been highlighted but rarely with reference to the invasion ecology hypotheses.
L29: “invasibility”
L44: “In cities, aliens species..”
L63-65: This has been also pointed out in Gaertner et al. (2017).
Gaertner et al. 2017. Non-native species in urban environments: patterns, processes, impacts and challenges. Biological Invasions 19: 3461-3469.
L69: Do you mean you have examined the possible change in invasibility between forest vegetation types?
L86-87: It should be clear from the main text what are and what differentiates “ecological lands” and “non-protected urban forest”? Is it simply protected vs. non-protected land?
Figure 1: Location of study area (a) should be smaller than the other maps (b,c,d).
Figure 1 caption: Poznań is the study area while the four areas can be described as study sites. I imagine the dots are the study plots to which different syntaxa/associations are assigned.
L118: Do authors mean that they assigned a vegetation association to each plot?
L114-118: Authors should state that early emerging native species might have been overlooked based on data selection.
L128-131: Are all square plots? Or there were circular plots?
L131-132: What about looking for all species together?
L136-137: I suggest moving this at the beginning of the results section.
- I suggest to add the database or list of species used for nomenclature and for differentiating between native and alien species.
L140: I suggest inserting here the sentence regarding the use of R (L171).
L140-171: Please mention the packages of R that have been used (in addition to the function).
L140-171: What transformation method was used with the Braun-Blanquet scales?
L147: EIV acronym is not introduced in the text.
Table 1: group columns for “8 classes”
L204: Did you mean “alien species cover”?
Table 3: “No. of plots” as description of the second column.
Table 4: Are frequencies reported as % of the total?
Table 4: “Chaenomeles japonica– P – cas*”, please move the asterisks just after the species name to be consistent with the reporting of archeophytes.
- It is someway strange to see that Robinia pseudoacaciais not reported in the table, even though it is a characteristic species of the Chelidonio-Robinietum!
- Discussion: It would be important to have a paragraph comparing the general results of the study with what observed in the literature focusing on urban forests and alien species. The origin of the forests may be an important factor not discussed in the manuscript. Spontaneous vegetation can host a high number of alien species. For example, Sitzia et al. (2016) reported for secondary forests higher number of alien species than those recorded by the authors. Higher number of alien species are also reported at higher latitudes such as in the case of Riga and Kaunas (Marozas et al. 2015). A number of other examples are available in the literature, but these are particularly related to this study as they consider different forest types and context or origin of forest stands.
Marozas et al. 2015. Comparison of neophyte communities of Robinia pseudoacacia L. and Acer negundo L. in the eastern Baltic Sea region cities of Riga and Kaunas. Urban Forestry & Urban Greening 14, 826-834.
Sitzia et al. 2016 Novel woodland patches in a small historical Mediterranean city: Padova, Northern Italy. Urban Ecosystems 19: 475-487.
L277-282: It seems to me that it should be part of another paragraph separated from the focus on Impatiens parviflora. Furthermore, it is not related to the discussion of the results.
L310: Better “formed” rather than “comprised”.
L326-327: But how did authors deal with propagule pressure?
L330-331: Not clear what you mean with this sentence.
L333-334: I have read your results in a different way. No Robinia pseudoacacia is in Chelidonio-Robinietumplots (Table 4), so it is difficult (not to say wrong) to state “but just produced their own offspring”. Furthermore, the predicted number of alien species is statistically higher in Chelidonio-Robinietumplots compared to all other associations (figure 2c). Therefore, one could suggest that invasional meltdown may be a likely process. Please better clarify all this.
L338-340: It could be driven by the effects of trees as “ecosystem engineers” but, authors cannot state this based on their results. It could be, for sure, even the contrary. Nutrients in the soil, as many other factors, may have played a role in the establishment of certain tree species.
L531, citation 44: the title of the chapter is missing.
Author Response
Reviewer 1
The manuscript presents an interesting study regarding forest invasion by alien species in the area of Poznań in Poland.
I’m concerned with the number of hypotheses that are claimed to be investigated. Testing specific ecological hypotheses needs specific experimental or field designs and data collection approaches, while this studies uses data derived from previous research projects (that therefore would have a sampling design appropriate for the previous specific aims). I do not think the study can test all these hypotheses but rather can use these hypotheses to explain the observed results. In fact, for example, some variables can indicate more than one hypothesis (e.g., from your table 1: EIV for environmental heterogeneity and habitat filtering). I suggest focusing on your main question and result: number and cover of alien species to change with vegetation types.
Reply: We agree with the Reviewer that our study cannot handle all hypotheses in table 1. The main intention of the study was, as it was mentioned by the Reviewer, to assess impact of variables. For that reason we decided to follow the suggestion and change aims into drivers of alien species richness and cover and use mentioned hypotheses to justify choice of predictors. We changed the aims and hypotheses and we moved column with factor name as a first in Table 1, which is justification of factors selection.
It would be interesting to see if similar results are maintained if only neophytes are used.
Reply: As both Reviewers concerned about lack of delimitation between archeophytes and neophytes, we decided to check the models using neophytes only. We ran the analysis excluding archeophytes, which resulted in similar variables importance and similar response curves This may result from low abundance of archeophytes (only in one vegetation type median cover exceeded 10%, but only in one study plot, Armoracia rusticiana). In three of eight vegetation types studied archeophytes were absent (Carici elongatae-Alnetum, Carpinetum betuli and Querco-Ulmetum minoris) and in other types they constituted in average from 1.2 ±0.7% (Fraxino-Alnetum) to 2.6±1.1% (Salicetum albae) of all species cover. Thus, we decided to exclude archaeophytes, due to their low overall frequency and abundance and focus on neophytes. We added that information in methods section, to acknowledge what we studied and why we exclude that group. As Reviewer 2 also raised the concerns about plot sizes, thus we added plot size into new models, to exclude these effects. We checked the effect of plot size within each vegetation type and we found significant effects, which could influence results of our study. For that reason we decided to include plot size in the new models, to separate size effects from studied biological effects. However, in new models we found significant influence of plot size on alien species cover. Thus, we decided to exclude the smallest plots (i.e. with an area of 50 m2 and lower, as well as the largest plots (i.e. with area >200 m2). This reduction of sample size caused that we had only three plots representing Querco-Ulmetum, which is insufficient sample size for that vegetation type. Thus, we excluded these three plots. This dataset reduction lead to provide more robust dataset with n=120, less biased by plot size effects. We also included plot size in random forest models to ensure that differences between 100 and 200 m2 plots will be separated from other factors.
Furthermore, I think that variables may be correlated and this could be stated in the discussion. For example, ecological indicators (i.e. soil fertility and moisture) may play a role in the presence of different associations. Therefore, not really making it possible to distinguish between habitat filtering and the effect of specific tree species. In addition, variability of these ecological indicators may be more relevant in testing environmental heterogeneity rather than FDis. I also think that certain associations are more linked to disturbed conditions (e.g. Chelidonio-Robinietum, Salicetum-albae, Populetum albae) and this may have played a role to.
Reply: Although in methods section we stated that ‘Random forest analysis is suitable to analyze cover and count data due to its robustness to non-normal distributions and collinearity of predictors’, we agree with the Reviewer that correlation between variables need to mentioned in the discussion, as it may not influence the model (in a technical meaning) but may blur the interpretations. Thus we added a specific paragraph dedicated to study limitations.
I also suggest to compare these results with literature reporting studies and comparison between alien and native species in forests within urban areas (please see specific comment).
Reply: According to the suggestion we added a new paragraph to the discussion to compare our results with more literature, especially those studies which were suggested by the Reviewer.
Finally, I suggest modifying the title. I would not to use a question (“Is urban forest vegetation invisibility context-dependent?”), but rather a main and clear message from your results.
Reply: According to the suggestions of both Reviewers we decided to change the title to both (as suggested by Reviewer 1) be a clear message and (as suggested by Reviewer 2) be more conceptualized – refer to invasion level (expressed by alien species richness) and their ecological success (alien species cover). We entitled the revised version ‘Context-dependence of urban forest vegetation invasion level and alien species ecological success’
Specific comments
L14: There are a number of studies investigating this aspect (see special issue in Biological invasions “Non-native species in urban environments: Patterns, processes, impacts and challenges” and references in the different articles). I suggest deleting this sentence. I would rather highlight that a number of reasons have been highlighted but rarely with reference to the invasion ecology hypotheses.
Reply: Corrected according to the suggestions.
L29: “invasibility”
Reply: We checked the manuscript thoroughly and replaced each ‘invisib*’ into ‘invasiv*’
L44: “In cities, aliens species..”
Reply: We added a coma, according to the suggestion
L63-65: This has been also pointed out in Gaertner et al. (2017).
Gaertner et al. 2017. Non-native species in urban environments: patterns, processes, impacts and challenges. Biological Invasions 19: 3461-3469.
Reply: Thank you for nice hint, we added the citation.
L69: Do you mean you have examined the possible change in invasibility between forest vegetation types?
Reply: As according to the Reviewer 2 we changed scope from invasibility to level of invasion we changed the aims and we deleted the mentioned sentence
L86-87: It should be clear from the main text what are and what differentiates “ecological lands” and “non-protected urban forest”? Is it simply protected vs. non-protected land?
Reply: We added in the parentheses a brief definition of ecological land (areas under low-restrictive nature conservation), as this is a simple information about conservation status. We also refer to our study with wider description of the idea of ecological lands, as this nature conservation form is not common.
Figure 1: Location of study area (a) should be smaller than the other maps (b,c,d).
Figure 1 caption: Poznań is the study area while the four areas can be described as study sites. I imagine the dots are the study plots to which different syntaxa/associations are assigned.
Reply: We corrected the figure and captions according to the suggestions We also added the sentence that dots represents vegetation types assigned to each study plot.
L118: Do authors mean that they assigned a vegetation association to each plot?
Reply: We wrote the sentence not precisely. We mentioned that firstly we conducted a field reconnaissance to assess which vegetation types are most frequent – rough estimation of abundance of each type in the study sites. We reworded this sentence for being better understandable.
L114-118: Authors should state that early emerging native species might have been overlooked based on data selection.
Reply: According to the suggestion we added a sentence “For that reason we might overlook early emerging native species” in the mentioned lines.
L128-131: Are all square plots? Or there were circular plots?
Reply: We added an information that plots were square or rectangular.
L131-132: What about looking for all species together?
Reply: According to the Reviewer 2 suggestion we checked potential plot size effects and we included plot size in the models as a variable. This allowed to exclude plot size-depndent effects from the new models.
L136-137: I suggest moving this at the beginning of the results section.
- I suggest to add the database or list of species used for nomenclature and for differentiating between native and alien species.
Reply: We added the information about source of nomenclature and alien species status. We decided to not move an information about species number as this one, separated sentence does not fit to any chapter of results.
L140: I suggest inserting here the sentence regarding the use of R (L171).
Reply: Moved according to the Reviewer’s suggestion.
L140-171: Please mention the packages of R that have been used (in addition to the function).
Reply: We have already mentioned package names using R code notation package::function_name() – this is the most convenient way to provide all information necessary to use a function, even without loading the library (if the library is not already installed, R returns error with the message that there are no package ‘package’, which informs which package should be installed).
L140-171: What transformation method was used with the Braun-Blanquet scales?
Reply: We did not use any transformation, as we aimed to model cover of alien species.
L147: EIV acronym is not introduced in the text.
Reply: We explained the acronym in the text in first mention.
Table 1: group columns for “8 classes”
Reply: Grouped, according to the suggestion
L204: Did you mean “alien species cover”?
Reply: Yes, we corrected it according to the suggestion.
Table 3: “No. of plots” as description of the second column.
Reply: Corrected according to the suggestion. However, according to the Reviewer 2 suggestion, we moved this column into plots characteristics table.
Table 4: Are frequencies reported as % of the total?
Reply: Are % of all plots within each vegetation type. We added this explanation to the caption.
Table 4: “Chaenomeles japonica– P – cas*”, please move the asterisks just after the species name to be consistent with the reporting of archeophytes.
Reply: Corrected.
- It is someway strange to see that Robinia pseudoacacia is not reported in the table, even though it is a characteristic species of the Chelidonio-Robinietum!
Reply: Yes, it was a mistake connected with tables selection and formatting. We corrected this and now R. pseudoacacia is present.
- Discussion: It would be important to have a paragraph comparing the general results of the study with what observed in the literature focusing on urban forests and alien species. The origin of the forests may be an important factor not discussed in the manuscript. Spontaneous vegetation can host a high number of alien species. For example, Sitzia et al. (2016) reported for secondary forests higher number of alien species than those recorded by the authors. Higher number of alien species are also reported at higher latitudes such as in the case of Riga and Kaunas (Marozas et al. 2015). A number of other examples are available in the literature, but these are particularly related to this study as they consider different forest types and context or origin of forest stands.
Marozas et al. 2015. Comparison of neophyte communities of Robinia pseudoacacia L. and Acer negundo L. in the eastern Baltic Sea region cities of Riga and Kaunas. Urban Forestry & Urban Greening 14, 826-834.
Sitzia et al. 2016 Novel woodland patches in a small historical Mediterranean city: Padova, Northern Italy. Urban Ecosystems 19: 475-487.
Reply: We added the paragraph with general description of results with relation to other studies, as suggested by the Reviewer.
L277-282: It seems to me that it should be part of another paragraph separated from the focus on Impatiens parviflora. Furthermore, it is not related to the discussion of the results.
Reply: According to the suggestion we deleted this paragraph, as not related to the discussion of the results.
L310: Better “formed” rather than “comprised”.
Reply: Corrected according to the suggestion
L326-327: But how did authors deal with propagule pressure?
Reply: We agree that this sentence does not fit thus we deleted it.
L330-331: Not clear what you mean with this sentence.
Reply: We reworded the sentence.
L333-334: I have read your results in a different way. No Robinia pseudoacacia is in Chelidonio-Robinietumplots (Table 4), so it is difficult (not to say wrong) to state “but just produced their own offspring”. Furthermore, the predicted number of alien species is statistically higher in Chelidonio-Robinietumplots compared to all other associations (figure 2c). Therefore, one could suggest that invasional meltdown may be a likely process. Please better clarify all this.
Reply: As we found regeneration of R. pseudoacacia in Chelidonio-Robinietum (corrected table 4 due to missed rows), now sentence about producing the offspring is correct. However, we agree with Reviewer that we can confirm invasional meltdown. We corrected it in the mentioned lines.
L338-340: It could be driven by the effects of trees as “ecosystem engineers” but, authors cannot state this based on their results. It could be, for sure, even the contrary. Nutrients in the soil, as many other factors, may have played a role in the establishment of certain tree species.
Reply: We agree with the Reviewer and we used softer vocabulary and added the alternative provided by the Reviewer.
L531, citation 44: the title of the chapter is missing.
Reply: The title of chapter is “Poznań”, as chapters of that book are entitled by cities names.
We are thankful to the Reviewer for a thorough and valuable comments to the earlier draft of our manuscript.
Reviewer 2 Report
This manuscript analyses plant alien richness and cover and their determinants across eight urban forest vegetation types within one city. Authors tested quite a wide range of factors that are supposed to have an influence on the magnitude of invasion in sampled plots and that were related to several community invasibility hypotheses. Forest communities in urban areas have not been studied very intensively yet and submitted manuscript therefore provides valuable insight into the invasion patterns in urban forests. The text is generally well and concisely written. However, I have some major concerns about the framing, data processing and discussion of the results. Below I offer some suggestions that may help to clarify its contributions.
Major concerns:
The focus of the paper, also mentioned in the manuscript title, is to find determinants of urban forest vegetation invasibility. However, in the analyses, alien richness as the number of alien species in vegetation plots or maybe alien cover (it is not specified anywhere) is used as a measure of the forest invasibility. I think this is not a convenient measure. Alien richness is co-determined by invasibility and propagule pressure as is also stated in the Introduction. Thus high number of aliens in a community can be a consequence of high invasibility of that community despite lower propagule pressure as well as lower invasibility overwhelmed by high input of alien propagules into the community, i.e. high propagule pressure. Alien richness therefore should be better described using the term level of invasion. Please see the overview in Pyšek et al. (2010) concerning the relation between invasibility and level of invasion in different temperate plant community types (Table 6.1) including information about related studies. To assess the community invasibility per se, the confounding effect of the differences in propagule pressure among communities has to be removed. An example of such analysis can be found in the study of Chytrý et al. (2008). Authors in the submitted manuscript should therefore decide, whether to use the term level of invasion for the alien richness that is more appropriate and clarify the difference between actual level of invasion found and invasibility in the discussion, or whether to recalculate the analyses via removing the confounding effects of propagule pressure first and then test factors related to community invasibility on adjusted alien richness.
Alien richness and alien cover and their implications should be more clearly differentiated in the aims and discussion of the paper. For example in Aims on l. 68 authors stated that they would like to examine how much may be the variance in ecological success of aliens explained by each hypothesis. Here you refer specifically to alien richness or alien cover as the explained variable? Alien cover from this point of view should be related to the alien impact or ecological success rather than to community invasibility.
Chytrý, M., Jarošík, V., Pyšek, P., Hájek, O., Knollová, I., Tichý, L. & Danihelka, J. (2008) Separating habitat invasibility by alien plants from the actual level of invasion. Ecology, 89, 1541–1553.
Pyšek P., Chytrý, M. & Jarošík, V. (2010) Habitats and land use as determinants of plant invasions in the temperate zone of Europe. Bioinvasions and globalization: Ecology, Economics, Management and Policy (ed. by Perrings, Ch., Mooney, H. and M. Williamson), pp. 66–79, Oxford University Press, Oxford.
I have some doubts about analysing archaeophytes and neophytes together, because archaeophytes are well known to have different habitat preferences compared to neophytes. I would expect that archaeophytes generally preferring drier and warmer conditions than neophytes can more sharply respond to the moisture gradient than neophytes. Could this somehow influence your results for different forest vegetation types? Would be your results the same with these two alien species groups analysed separately? I miss some information about the possible effect of such alien delimitation in the discussion.
Can you please add somewhere to the Methods how many plots for each of eight forest vegetation types were selected? What was the plot size range for each of distinguished forest vegetation types? Was the effect of the plot size on alien and total species richness non-significant also within each of eight forest vegetation types?
Minor points:
Check for invisible/invasible throughout the text
l. 169-171 ANOVA tests were used to compare alien richness and cover among forest vegetation types, how was the normality of data treated?
Author Response
Reviewer 2
This manuscript analyses plant alien richness and cover and their determinants across eight urban forest vegetation types within one city. Authors tested quite a wide range of factors that are supposed to have an influence on the magnitude of invasion in sampled plots and that were related to several community invasibility hypotheses. Forest communities in urban areas have not been studied very intensively yet and submitted manuscript therefore provides valuable insight into the invasion patterns in urban forests. The text is generally well and concisely written. However, I have some major concerns about the framing, data processing and discussion of the results. Below I offer some suggestions that may help to clarify its contributions.
Major concerns:
The focus of the paper, also mentioned in the manuscript title, is to find determinants of urban forest vegetation invasibility. However, in the analyses, alien richness as the number of alien species in vegetation plots or maybe alien cover (it is not specified anywhere) is used as a measure of the forest invasibility. I think this is not a convenient measure. Alien richness is co-determined by invasibility and propagule pressure as is also stated in the Introduction. Thus high number of aliens in a community can be a consequence of high invasibility of that community despite lower propagule pressure as well as lower invasibility overwhelmed by high input of alien propagules into the community, i.e. high propagule pressure. Alien richness therefore should be better described using the term level of invasion. Please see the overview in Pyšek et al. (2010) concerning the relation between invasibility and level of invasion in different temperate plant community types (Table 6.1) including information about related studies. To assess the community invasibility per se, the confounding effect of the differences in propagule pressure among communities has to be removed. An example of such analysis can be found in the study of Chytrý et al. (2008). Authors in the submitted manuscript should therefore decide, whether to use the term level of invasion for the alien richness that is more appropriate and clarify the difference between actual level of invasion found and invasibility in the discussion, or whether to recalculate the analyses via removing the effects of propagule pressure first and then test factors related to community invasibility on adjusted alien richness. confounding
Alien richness and alien cover and their implications should be more clearly differentiated in the aims and discussion of the paper. For example in Aims on l. 68 authors stated that they would like to examine how much may be the variance in ecological success of aliens explained by each hypothesis. Here you refer specifically to alien richness or alien cover as the explained variable? Alien cover from this point of view should be related to the alien impact or ecological success rather than to community invasibility.
Chytrý, M., Jarošík, V., Pyšek, P., Hájek, O., Knollová, I., Tichý, L. & Danihelka, J. (2008) Separating habitat invasibility by alien plants from the actual level of invasion. Ecology, 89, 1541–1553.
Pyšek P., Chytrý, M. & Jarošík, V. (2010) Habitats and land use as determinants of plant invasions in the temperate zone of Europe. Bioinvasions and globalization: Ecology, Economics, Management and Policy (ed. by Perrings, Ch., Mooney, H. and M. Williamson), pp. 66–79, Oxford University Press, Oxford.
Reply: We are thankful to the Reviewer for that conceptualization. We clearly mentioned in abstract, introduction and aims that we tested both level of invasion and alien species ecological success. We decided to not remove cofounding confounding effects of propagule pressure following Chytrý et al. (2008), as our dataset covers mostly plots with high level of variables approximating propagule pressure, due to neighboring estates, roads and rivers. For that reason design of our study provides relatively short gradient of propagule pressure. Instead, we changed invasibility into ‘level of invasion’. We also distinguished between ‘level of invasion’ and ‘ecological success’ in introduction, aims and conclusions. According to the suggestions of both Reviewers we decided to change the title to both (as suggested by Reviewer 1) be a clear message and (as suggested by Reviewer 2) be more conceptualized – refer to invasion level (expressed by alien species richness) and their ecological success (alien species cover). We entitled the revised version ‘Context-dependence of urban forest vegetation invasion level and alien species ecological success’.
I have some doubts about analysing archaeophytes and neophytes together, because archaeophytes are well known to have different habitat preferences compared to neophytes. I would expect that archaeophytes generally preferring drier and warmer conditions than neophytes can more sharply respond to the moisture gradient than neophytes. Could this somehow influence your results for different forest vegetation types? Would be your results the same with these two alien species groups analysed separately? I miss some information about the possible effect of such alien delimitation in the discussion.
Reply: We agree that this delimitation would influence the results. We ran the analysis excluding archeophytes, which resulted in similar variables importance and similar response curves This may result from low abundance of archeophytes (only in one vegetation type median cover exceeded 10%, but only in one study plot, Armoracia rusticiana). In three of eight vegetation types studied archeophytes were absent (Carici elongatae-Alnetum, Carpinetum betuli and Querco-Ulmetum minoris) and in other types they constituted in average from 1.2 ±0.7% (Fraxino-Alnetum) to 2.6±1.1% (Salicetum albae) of all species cover. Thus, we decided to exclude archeophytes, due to their low overall frequency and abundance and focus on neophytes. We added that information in methods section, to acknowledge what we studied and why we exclude that group. We also mention it in the discussion.
Can you please add somewhere to the Methods how many plots for each of eight forest vegetation types were selected? What was the plot size range for each of distinguished forest vegetation types? Was the effect of the plot size on alien and total species richness non-significant also within each of eight forest vegetation types?
Reply: As suggested by the Reviewer we checked the effect of plot size within each vegetation type and we found significant effects, which could influence results of our study. For that reason we decided to include plot size in the new models, to separate size effects from studied biological effects. However, in new models we found significant influence of plot size on alien species cover. Thus, we decided to exclude the smallest plots (i.e. with an area of 50 m2 and lower, as well as the largest plots (i.e. with area >200 m2). This reduction of sample size caused that we had only three plots representing Querco-Ulmetum, which is insufficient sample size for that vegetation type. Thus, we excluded these three plots. This dataset reduction lead to provide more robust dataset with n=120, less biased by plot size effects. We also included plot size in random forest models to ensure that differences between 100 and 200 m2 plots will be separated from other factors. We also added the table with plot sizes (min, max, mean and SE). Sample size (no. of plots per each vegetation plot) was provided in previous version of the manuscript, in Table 3. However, we agree with the Reviewer that it would better fit to the descriptive table in Methods section, thus we moved this information to the new table.
Minor points:
Check for invisible/invasible throughout the text
Reply: We checked the manuscript thoroughly and replaced each ‘invisib*’ into ‘invasiv*’
l. 169-171 ANOVA tests were used to compare alien richness and cover among forest vegetation types, how was the normality of data treated?
Reply: As our data had non-normal distribution (according to Shapiro-Wilk’s tests) we decided to use Poisson regression and zero-inflated beta regression instead. We added the description to Methods section and we changed tables. However, changes in interpretations were low, thus we did not have to change results description.
We are thankful to the Reviewer for a thorough and valuable comments to the earlier draft of our manuscript.
Round 2
Reviewer 1 Report
Dear Authors,
I thank you for addressing all the issues that have been raised.
Minor english corrections are needed and together with other small changes to ensure consistency throughout the text (e.g. "including 6 archaeophytes..., 24 neophytes.."; "three ecological lands (i.e. areas under low-restrictive)"). In the caption of table 4, please do point out that species are those recorded in the herb layer (this is important, because if not, one should expect to find robinia in 100% of the Chelidonion-Robinietum plots!).
Good luck with your future research
Author Response
Reviewer 1
I thank you for addressing all the issues that have been raised.
Minor english corrections are needed and together with other small changes to ensure consistency throughout the text (e.g. "including 6 archaeophytes..., 24 neophytes.."; "three ecological lands (i.e. areas under low-restrictive)"). In the caption of table 4, please do point out that species are those recorded in the herb layer (this is important, because if not, one should expect to find robinia in 100% of the Chelidonion-Robinietum plots!).
Good luck with your future research
Reply: We are happy that our revisions were appreciated by the Reviewer. We thoroughly revised language issues within the manuscript. In case of numbers we decided to follow the pattern suggested in majority of language guidelines: in case of numbers without units lower than 11 we spelled them in full, and in case of number higher than ten – we wrote numbers. We also pointed out that alien species in table 4 are recorded in herb layer. We are thankful for a good wishes.
We are thankful to the Reviewer for a thorough and valuable comments to the earlier draft of our manuscript.
Reviewer 2 Report
After a thorough reading of the manuscript I can say that I am personally satisfied with the changes made. Authors properly addressed all concerns raised. I appreciate the clarification of concepts and terms used in the title and Introduction and the amendment of the analyses that enhanced the quality of the ms. I believe that revised manuscript brings valuable contribution to the understanding of urban forests’ levels of invasion patterns and their underlying mechanisms. I attached several additional remarks below, however all of them being of minor importance.
Minor points
Figure 1. variants a), b), c), and d) used in the maps should be also distinguished and described in the figure caption
l. 106-134 Vegetation type Querco-Ulmetum is no longer assessed in the analyses and therefore its description in this paragraph has become excessive and should be deleted.
l. 398 Jarošík et al. – not given by citation number
l. 414 wrong word order used, change to e.g. different responses to environmental factors compared to neophytes
Author Response
Reviewer 2
After a thorough reading of the manuscript I can say that I am personally satisfied with the changes made. Authors properly addressed all concerns raised. I appreciate the clarification of concepts and terms used in the title and Introduction and the amendment of the analyses that enhanced the quality of the ms. I believe that revised manuscript brings valuable contribution to the understanding of urban forests’ levels of invasion patterns and their underlying mechanisms. I attached several additional remarks below, however all of them being of minor importance.
Reply: We are happy that our revisions were appreciated by the Reviewer.
Minor points
Figure 1. variants a), b), c), and d) used in the maps should be also distinguished and described in the figure caption
Reply: According to the suggestion we added captions for each figure variant (a-d) separately.
l. 106-134 Vegetation type Querco-Ulmetum is no longer assessed in the analyses and therefore its description in this paragraph has become excessive and should be deleted.
Reply: According to the suggestion we removed parts refering to Querco-Ulmetum
l. 398 Jarošík et al. – not given by citation number
Reply: We added a lacking number to the mentioned line
l. 414 wrong word order used, change to e.g. different responses to environmental factors compared to neophytes
Reply: Changed according to the suggestion
We are thankful to the Reviewer for a thorough and valuable comments to the earlier draft of our manuscript.
