Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study is useful because it predicts the spread of the invasive insect Pagiophloeus tsushimanus. However, there are a few comments.

1. Keywords must not repeat the words from the title, because keywords are necessary for search systems only.
2. The insect's full binomial name with authority must be mentioned – Pagiophloeus tsushimanus Morimoto, 1982, host plant name with authority – Cinnamomum camphora (L.) J.Presl
3. Cinnamomum camphora grows in tropical and subtropical climates. As temperatures rise, the climate will become drier almost everywhere, making it less suitable for growing this plant. Even in suitable climates for C. camphora in northern regions, landowners will not allocate land suitable for traditional crops to this tree.
4. The insect's response to climate change in terms of distribution and harmfulness depends on its life cycle characteristics. Therefore, a brief description of the P. tsushimanus life cycle should be included in the Introduction or in the Materials and Methods. Rising temperatures will also lead to an earlier start and a later end of the growing season, thereby increasing its duration. Studies (Jia et al., 2020; Chen et al., 2021) show that P. tsushimanus is monovoltine, with both larvae and adults overwintering. Rising temperatures will not increase the number of generations, but they may induce summer diapause and affect the ratio of overwintering larvae to adults, particularly the proportion of older larvae.
5. tsushimanus larvae bore galleries in phloem and xylem and weaken trees. The beetles overwinter in branch forks in facultative diapause. At higher spring temperatures, the beetles will begin feeding on branch bark and buds, further weakening the trees. Both larvae and beetles promote pathogen penetration into the tree. The risk of pathogen transmission by this weevil may increase under a changing climate, especially given potential changes in the timing of bud break, the dates of adult feeding after overwintering, and the cycles of potential pathogens. This should also be considered in the Discussion.
6. The indices and parameters used to predict the potential global distribution of P. tsushimanus are described. How was the annual growth index (GIA) calculated, and what are the units of measurement (sq. dm of trunk surface? tree?). Apparently, the influence of various climatic factors on the weevil population depends on its stage. Drought, probably, is most dangerous during the egg stage and young larvae in the phloem.

 

6.References.

 

Line 325: Jia et al., 2020, and it is correct (JIA Zhiyi, CHEN Cong, MA Yuxuan, LI Shouyin, FAN Binqi, WANG Yan, HAO Dejun. Effects of temperature on growth and development of Pagiophloeus tsushimanus Morimoto[J]. Journal of Nanjing Forestry University (Natural Sciences Edition. 2020, 44(4): 131-136 https://doi.org/10.3969/j.issn.1000-2006.201904025). However, other authors are in the reference 27 (Wang, Y.; Chen, C.; Zhu, H.; Fan, B.Q.; Hao, D.J. Effect of temperature on growth and development of Pagiophloeus tsushimanus Morimoto (In Chinese with English abstract). J. Nanjing For. Univ. (Nat. Sci. Ed.) 2020, 44, 131–136. https:/ 903 /doi.org/10.3969/j.issn.1000-2006.201901007)

Source 2 in the reference list is “Chen, C.; Li, S.Y.; Wang, Y.; Zhu, H.; Fan, B.Q.; Hao, D.J. Biological traits and life history of Pagiophloeus tsushimanus (Coleoptera: Curculionidae), a weevil pest on camphor trees in China. Insects 2021, 12, 200. https://doi.org/10.3390/in 837 sects12030200”. In fact, it is “Chen, C., Zhang, C., Li, S., Zhu, H., Fan, B., Wang, Y., ... & Hao, D. (2021). Biological traits and life history of Pagiophloeus tsushimanus (Coleoptera: Curculionidae), a weevil pest on camphor trees in China. Journal of Forestry Research, 32(5), 1979-1988. https://doi.org/10.1007/s11676-020-01227-2”

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript focuses on predicting the potential distribution of Pagiophloeus tsushimanus, a specialist pest of Cinnamomum camphora, under current and future climatic conditions. The study is relevant to forest entomology, the biogeography of invasive species, and phytosanitary risk forecasting.

After reading the manuscript, I identified several critical issues:

1. Lines 211-217. The SSP3-7.0/CMIP6 scenario was used for the host plant, whereas the SRES A1B/AR4 scenario was used for the insect. These pathways cannot be considered methodologically equivalent solely because both may be broadly classified as intermediate or intermediate-to-high emission scenarios. They belong to different generations of climate models, are based on different assumptions, and produce different spatial patterns of temperature and precipitation. Furthermore, the historical period used for the RF model is reported as 1970-2000 (line 419), whereas the CLIMEX model uses 1961-1990 (lines 293-295). Consequently, overlaying these rasters and subsequently claiming a “synchronous” shift in the distributions of the host plant and phytophagous insect does not provide a rigorous comparison. The two components of the analysis should either be harmonized using the same generation of climate data, emission scenario, baseline period, and preferably a consistent GCM ensemble, or a separate uncertainty analysis should be conducted and the interpretation of the combined projection substantially qualified. The claimed correspondence between SSP3-7.0 and A1B requires quantitative justification rather than a purely verbal analogy.

2. Lines 185-195. Randomly dividing spatially clustered occurrence records into training and testing datasets commonly places nearby records in both subsets, thereby inflating AUC, TSS, and kappa values. Spatial thinning alone is not a substitute for spatial block cross-validation. The authors should apply spatial or block cross-validation and report the block size, number of repetitions, variability in performance metrics among repetitions, and the procedure used to tune the RF parameters. The global extent used to generate 7,200 pseudo-absence points, the 1:4 ratio, and the 10-km buffer are also insufficiently justified. Table 3 is particularly problematic because “accuracy” metrics are reported separately for the future SSP3-7.0 period, although no independent observations of the distribution of C. camphora in 2041-2060 exist. Discrimination metrics evaluate a model against observed contemporary data and therefore cannot validate the accuracy of a future projection. The authors should explain precisely how the AUC, kappa, and TSS values for the future period were obtained or remove them. Assessment of model transferability also requires maps of extrapolation or novel climatic conditions, such as MESS maps or an equivalent analysis, together with an evaluation of inter-model uncertainty.

3. Dividing 17 occurrence records in a 75:25 ratio produces a test dataset of approximately four records, which is insufficient for a reliable assessment of model transferability. Model parameters were fitted to the known distribution, and model reliability was subsequently justified by stating that all known records occurred within climatically suitable areas. However, it is unclear whether “all records” refers only to the test dataset or also includes calibration records. Such agreement does not constitute independent model validation and does not assess false-positive predictions. Repeated data partitioning or leave-one-out/spatially structured validation should be performed. Calibration and validation records should be clearly distinguished, quantitative measures of sensitivity and specificity should be reported, and the sensitivity of the prediction to the principal CLIMEX parameters should be examined.
In addition, Table 2 is inconsistent with the accompanying text. The table reports SM1 = 0.6, SM2 = 0.8, SM3 = 1.5, and HDS = −0.005 week⁻¹, whereas lines 339-343 give SM1 = 0.5, SM2 = 1.5, and SM3 = 2.5, and lines 361–366 give HDS = −0.01 week⁻¹. The analysis therefore cannot be reproduced. The selection of DV0 = 13.04 °C as the maximum threshold among the developmental stages also requires verification. Summing stage-specific thermal constants calculated using different lower developmental thresholds is not necessarily equivalent to calculating a single accumulated degree-day requirement above the larval threshold. The authors should provide the original calculations or a biologically consistent justification for aggregating the developmental stages. This point requires clarification.

4. The manuscript states that, of the 17,395 C. camphora occurrence records, 1,800 records were retained for modelling after spatial thinning and 340 records were used for the figure. However, the accompanying C. camphora dataset contains all 17,395 rows, including 4,210 exact coordinate duplicates, and contains no variable that would allow the 1,800 modelling records or the 340 mapping records to be identified. The P. tsushimanus supplementary dataset contains 17 records, but their distribution among sources 3 GBIF records, ten field records, and four literature records does not correspond to the statement that five records were obtained from the literature (lines 171-183). Information on georeferencing accuracy, observation date, original record identifier, and the rules used to remove spatially dependent records is also absent.
The authors should provide the final datasets actually used in each model, including unique identifiers and sources, together with the analysis code, random seed, RF parameters, pseudo-absence generation procedure, and final raster outputs. The statement that the data are available “upon request” (lines 824-825) is inconsistent with the level of transparency required to reproduce such a complex spatial modelling procedure.

5. The heat-stress threshold of 38 °C and the moisture-stress parameters are CLIMEX calibration settings. However, the Discussion presents them as experimentally validated physiological limits above which survival and reproduction “decline significantly” (lines 688-690). References [25] and [27] do not support all elements of this statement. Similarly, reference [54] concerns the general vulnerability of forests to drought and does not substantiate the claim that cavitation in C. camphora will inevitably occur and subsequently cause a local collapse of the P. tsushimanus population (lines 645-650). Statements concerning “oasis corridors” created by irrigation (lines 658-663), the inability of the species to establish in tropical forests (lines 700-705), and the effectiveness of pheromone traps specifically for P. tsushimanus (lines 783-784) require direct supporting references or more cautious wording. The model estimates potential climatic suitability but does not demonstrate actual invasion, dispersal rates, or the effectiveness of a specific monitoring method.

After the methodological issues have been addressed, the English text requires careful technical editing. The manuscript contains typographical errors and inconsistent terminology or abbreviations, including “Juen,” “Spices,” “Perdicted,” and “HM/gHM.” There are also errors in the numbering or formatting of headings and tables, as well as statements referring to “significant” changes without supporting statistical tests.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors,

I appreciate the thorough revision of the manuscript. However, a few points still require final adjustment.

1. The future AUC, kappa, and TSS values ​​must be removed from Table 3. Currently, they remain in the manuscript despite the authors' response.
2. Please provide and reconcile the complete package of data and materials: final tables, script, filtering rules, parameters, randomization details, output rasters, and a description of exactly which models and maps they reproduce.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf