Standardized Drought Indices for Pre-Summer Drought Assessment in Tropical Areas
Emanuel Storey
Round 1
Reviewer 1 Report
Review of Atmosphere manuscript 952384
This manuscript addresses an important topic, namely the application of drought indices in tropical regions. The authors propose a metric based on aggregation of precipitation and temperature data from regionally-specific seasons that are most important in drought assessment, and this is the major novelty of the study. Overall, this research could be valuable to meteorologists and other scientists working in tropical regions. However, this reviewer perceives several weaknesses in the study and in the way it is organized and presented.
Methodologically, the authors have not provided a sufficiently convincing basis for evaluating the drought indices [it is my recommendation to compare the indices with soil moisture, streamflow, or NDVI in specific crop or forest types]. The indices are evaluated based on correlation with NDVI data, aggregated at sub-regional scales to include many different land cover types, producing weak correlations. One conclusion supported by the results is that the proposed indices are more correlated to NDVI than the Standardized Precipitation Index (SPI); however, the SPI is tabulated at annual time steps, whereas the other indices were computed for seasonal periods, so this finding is simply to be expected. The most important findings must be highlighted more clearly, in the Abstract as well as Discussion and Conclusion sections.
The presentation of the study also requires improvement. Most of all, the methods are poorly organized and unclear in many places. It remains unclear which methods and steps were actually employed by the authors, as the English tenses used suggest that the methods are practiced by others [please see specific comments below]. Also, there are several misleading and contradictory aspects of the study, as specified below. This reviewer recommends that the authors conduct a thorough revision of the Methods section in particular, according to the comments given below. Also, the Discussion should focus more on the drought indices and the way they were evaluated, in comparison to other studies. This work should be published, but requires substantial improvement. I acknowledge effort and inspiration contained in the research.
Abstract -
General: your abstract is the most important section of the document, as it will determine whether people download the full paper. Currently, it contains many statements that are vague
15: please consider using different word than “elaborated”, such as “developed”.
17: suggest “with respect to seasonal rainfall variations” rather than “regarding the rainfall regime”
18: this use of “site conditions” will be vague to readers. Is it not the case that you defined the drought periods uniquely according to districts within the Yucatan? If so, please make this explicit.
19-20: this sentence is redundant with the previous. Consider eliminating to make space for other information.
21: did you not also evaluate mid-summer drought? Please clarify in text.
23: On what basis were the indices evaluated? Please specify (e.g., by correlation with AVHRR NDVI time series?).
26: suggest ending the sentence by stating “…, and long-term increase in drought severity observed over the period 19XX-20XX is consistent with climatic change.”
Introduction –
31: rather than “…linked to overexploitation”, I would recommend stating “exacerbated by overextraction of freshwater aquifers….”
36: it is not application of PDSI, but rather computation or derivation of PDSI which requires the data noted here.
42: please provide reference to URL in text, otherwise as formal reference in references list (e.g., CliCom 2020).
38: it seems to require spatially detailed information rather than “broad knowledge”
40: here you state that SPI requires only precipitation data, but later in the Methods, you describe how you also used temperature in its computation. This seems very inconsistent, and must be clarified here whether the inclusion of additional variables is standard practice? If the input variables can be selected at will, then you should cite a study in which variables are included or omitted as options.
44-45: it is unclear what is meant by the distinction between dry or wet, versus rainy or dry. Consider revising sentence for clarity.
46: I think this is a good juncture to state that the SPI must be defined according to periods that are relevant to the seasonal drought characteristics of a particular area (and provide reference), which seems to be the emphasis of your study.
49-53: this information is good, but does not flow well with the rest of the introduction, and could be replaced by some description of the relevant periods for drought in your region, as recommended below.
56-59: The statements made on these lines are unclear in meaning, and potentially untrue. Tropical regions and other regions also experience months of scarce rainfall, followed by abundant rainfall (for example, Mediterranean type climate in California). I recommend stating instead that, in tropical regions, rainfall is far more important than temperature variation for estimating drought severity.
60-61: this is also not true – you cannot be sure that temperate regions experience more regular rainfall, especially with climatic change and the great variation among temperate regions. Please find reference to substantiate claim, or remove it. Also, please note that PDSI was developed also for arid regions where ET is very important.
66: you are not creating a drought index, because the SPI was created by others before. Instead, you are investigating the utility in computing SPI according to regionally specific pre-summer and mid-summer drought periods.
68: please also give formal citation of original Koppen classification document or system.
84: Consider stating “In this study, we utilized monthly data from the selected stations, including…which were processed in R version 4.0)”
88: here you state that you used the NDVI imagery “in order to compare the indices”. I believe this is misstated. Instead, you are comparing the correspondence of the indices (which at this point, have not been defined) to a remotely sensed index of vegetation abundance and condition. I am not confident this is the best way to evaluate a drought index, but will respect the decisions made by the authors.
89-90: only define these acronyms if they will be used again in the text.
90: You state that the period of available images begins in 1996. Why, then, does the study period covered by climatic data begin in 1952??
91: you state above that the images are from AVHRR. How then, do they originate from “various platforms and sensors”? This could bring up doubts regarding the validity of the analysis.
Section 2.2.1: I recommend condensing this text so it is more concise, and moving it to the Introduction (to replace lines 49-53). This will give readers a better sense of what you are trying to accomplish, and of the regional particularity.
Section 2.2.2: Generally, the methods sections are used to describe the methodology you followed. In this sub-section, there is virtually no description of your methods, but instead you have included several paragraphs that would be more appropriate in an introduction section. Given that your article is very lengthy, consider making this section more concise (max two paragraphs) and including it in the Introduction to better describe the SPI. Avoid redundancy with other information already in the Introduction.
162: Here you discuss “emptying of the water reserves” (which could be important) but this is not something addressed in your index which is based only on precipitation and temperature. Again, you are giving a description of many different things, but not telling the readers what you actually did in your own study. You should include more statements like “In this study, we applied a particular methodology using these variables, etc.”.
191: How was temperature included in the assessment of droughts?? This is unclear, and must be addressed. If you did not estimate different variables in the water balance (beyond precipitation), in what way was temperature included? Please clarify in the text.
195-200: Here you imply that you have computed the water balance, even though you state the opposite just above.
207: In this sentence you state “…values ranges are applied…”. Did you apply these value ranges, or did someone else? When you state that a certain thing IS done, this suggests that you are describing a generally accepted way of doing it. If you state that it WAS done, it becomes ambiguous whether it was you or someone else who did the action. You must therefore state explicitly that it was you (‘We did X’ or ‘In this study, X method was applied’). When this is made clear within a paragraph, it is then acceptable to include some statements in the form of ‘X was done’.
214: This reviewer questions strongly the appropriateness and utility of using Moran’s I to evaluate the drought indices. Spatial autocorrelation should be used to inform sampling, not to test whether an index is valid. Please consider eliminating this part of the analysis, as it is not well justified and is not meaningful in result.
227: Please clarify that temporal variation in NDVI can result from changes in either vegetation condition (e.g. drought stress, phenology) or abundance (e.g. biomass, cover).
228-231: Even though I was able to infer what is meant in these sentences after reading several times, many readers will struggle to understand. Please revise to improve clarity.
Results –
Figure 2: the placement of years on the X axes is not clear. The marks corresponding to the labelled years (e.g. 1950, 1957, etc.) should be longer or bolder than those of the other years. Consider 5-year intervals if possible. Also, figure text is too small relative to main text. Caption: please state full names of the indices in the caption (e.g. standardized precipitation index, etc.). Figures must stand on their own, not relying on information from the main text. Also, you state “all stations in the Yucatan Peninsula”, even though in the methods you describe using a subset with longer record periods. This is inconsistent. Please address this.
Table 2, Section 3.3: Here you find that the coefficients are positive, which suggests that droughts became less frequent or severe over the past decades. This finding contrasts with what you state in the abstract, that these show evidence of climate warming. Generally, the result is not surprising, as climatic change is predicted to result in less precipitation in arid regions, and as increased rainfall in tropical areas. Maybe this is a result worthy of further discussion and emphasis.
Discussion –
An essential part of this section should be comparison and contrast of your findings with a variety of other studies. You discuss other literature to a limited extent, and this is mostly about the particular phenomena observed in Mexico, rather than in the development and interpretation of tropical drought indices more broadly. Please augment this section accordingly.
Author Response
The authors appreciate your extensive review that improved notably the quality of the paper.
Thank you also for comments that pointed out that this manuscript addresses an important topic, namely the application of drought indices in tropical regions and that the propose metric, based on aggregation of precipitation and temperature data from regionally-specific seasons, are most important in drought assessment, been this the major novelty of the study. Also because you mention that this research could be valuable to meteorologists and other scientists working in tropical regions and that this work should be published.
We are going to attend your suggestions and the ones from other reviewers to improve the manuscript.
Abstract -
General: your abstract is the most important section of the document, as it will determine whether people download the full paper. Currently, it contains many statements that are vague
15: please consider using different word than “elaborated”, such as “developed”.
R/ Changed as suggested.
17: suggest “with respect to seasonal rainfall variations” rather than “regarding the rainfall regime”
R/ Changed as suggested.
18: this use of “site conditions” will be vague to readers. Is it not the case that you defined the drought periods uniquely according to districts within the Yucatan? If so, please make this explicit.
R/ Changed as suggested to make it explicit.
19-20: this sentence is redundant with the previous. Consider eliminating to make space for other information.
R/ In order to avoid redundancy as suggested, the content of both sentences have been simplified and reordered
21: did you not also evaluate mid-summer drought? Please clarify in text.
R/ No, it was evaluated for the pre-summer season. Several changes in the text clarify that.
23: On what basis were the indices evaluated? Please specify (e.g., by correlation with AVHRR NDVI time series?).
R/ The evaluation method has been inserted “by correlation with AVHRR NDVI time series”.
26: suggest ending the sentence by stating “…, and long-term increase in drought severity observed over the period 19XX-20XX is consistent with climatic change.”
R/ In the study, we do not identify a general long-term increase in drought severity as average trends were positive for all the station. We denoted a long-term effect of temperature on drought and with the differences between SPI12 and other indices (SPIdrought, ISPP and ISPB), a change un rainfall regime. The last could not be identified for all the meteorological stations.
Introduction –
31: rather than “…linked to overexploitation”, I would recommend stating “exacerbated by overextraction of freshwater aquifers….”
R/ Changed as suggested.
36: it is not application of PDSI, but rather computation or derivation of PDSI which requires the data noted here.
R/ Changed as suggested.
42: please provide reference to URL in text, otherwise as formal reference in references list (e.g., CliCom 2020).
R/ The URL has been added in the text.
38: it seems to require spatially detailed information rather than “broad knowledge”
R/ Changed as suggested.
40: here you state that SPI requires only precipitation data, but later in the Methods, you describe how you also used temperature in its computation. This seems very inconsistent, and must be clarified here whether the inclusion of additional variables is standard practice? If the input variables can be selected at will, then you should cite a study in which variables are included or omitted as options.
R/ In this study, the authors developed 2 new indices. These are both standardized indices but are not SPI. They are not calculated the same way, for the same period and include weighting.
44-45: it is unclear what is meant by the distinction between dry or wet, versus rainy or dry. Consider revising sentence for clarity.
R/ We clarify this in the text.
46: I think this is a good juncture to state that the SPI must be defined according to periods that are relevant to the seasonal drought characteristics of a particular area (and provide reference), which seems to be the emphasis of your study.
R/ We included your suggestion and a reference in the text.
49-53: this information is good, but does not flow well with the rest of the introduction, and could be replaced by some description of the relevant periods for drought in your region, as recommended below.
R/ Since you considered that this information is good, we prefer to keep the text for clarity. The description you mentioned was included below.
56-59: The statements made on these lines are unclear in meaning, and potentially untrue. Tropical regions and other regions also experience months of scarce rainfall, followed by abundant rainfall (for example, Mediterranean type climate in California). I recommend stating instead that, in tropical regions, rainfall is far more important than temperature variation for estimating drought severity.
R/ Changed as suggested.
60-61: this is also not true – you cannot be sure that temperate regions experience more regular rainfall, especially with climatic change and the great variation among temperate regions. Please find reference to substantiate claim, or remove it. Also, please note that PDSI was developed also for arid regions where ET is very important.
R/ Removed as suggested
66: you are not creating a drought index, because the SPI was created by others before. Instead, you are investigating the utility in computing SPI according to regionally specific pre-summer and mid-summer drought periods.
R/ Please notice that we are not proposing the creation of SPI, we proposed the creation of a new drought indices ISPP and ISPB.
68: please also give formal citation of original Koppen classification document or system.
R/ Köppen climate classification and reference added
84: Consider stating “In this study, we utilized monthly data from the selected stations, including…which were processed in R version 4.0)”
R/ Changed as suggested.
88: here you state that you used the NDVI imagery “in order to compare the indices”. I believe this is misstated. Instead, you are comparing the correspondence of the indices (which at this point, have not been defined) to a remotely sensed index of vegetation abundance and condition. I am not confident this is the best way to evaluate a drought index, but will respect the decisions made by the authors.
R/ NDVI has been used extensively to monitor drought. In an area like the Yucatan Peninsula, with dense vegetation cover and no surface water flow, it would be difficult to obtain time series of reliable information more comparable to the indices than the NDVI.
89-90: only define these acronyms if they will be used again in the text.
R/ We eliminated the acronym LAGE.
90: You state that the period of available images begins in 1996. Why, then, does the study period covered by climatic data begin in 1952??
R/ We agree that the major novelty of the study is the metric based on aggregation of precipitation and temperature data from regionally-specific seasons that are most important in drought assessment. For this only purpose, we could only retain climatic data for the NDVI period. However, long term analysis of drought indices gave valuable information on drought evolution respect to global warming and climate change. We could detect critical events and calculate trends.
91: you state above that the images are from AVHRR. How then, do they originate from “various platforms and sensors”? This could bring up doubts regarding the validity of the analysis.
R/ We clarified this point, add reference information relative to method and quality and gave reference of large time-series computed from NOAA-AVHRR images.
Section 2.2.1: I recommend condensing this text so it is more concise, and moving it to the Introduction (to replace lines 49-53). This will give readers a better sense of what you are trying to accomplish, and of the regional particularity.
R/ Other reviewers asked to add information in this section. They also suggested to move text line 174 to this section and they asked for adding explanations of complexity of drought at different time of the year and how they impact agriculture.
As the section would be too extended to be included in the introduction, general description of drought types were introduced in the introduction and specificity of pre-summer drought now takes part of the new 2.1 section with zone study climate descriptions, apart of the data section 2.2.
Section 2.2.2: Generally, the methods sections are used to describe the methodology you followed. In this sub-section, there is virtually no description of your methods, but instead you have included several paragraphs that would be more appropriate in an introduction section. Given that your article is very lengthy, consider making this section more concise (max two paragraphs) and including it in the Introduction to better describe the SPI. Avoid redundancy with other information already in the Introduction.
R/ Modified as suggested
162: Here you discuss “emptying of the water reserves” (which could be important) but this is not something addressed in your index which is based only on precipitation and temperature. Again, you are giving a description of many different things, but not telling the readers what you actually did in your own study. You should include more statements like “In this study, we applied a particular methodology using these variables, etc.”.
R/ The integration by the authors of an exponential weighting in the indices computing to take into account the emptying of the water reserves function is now clearer in the text
191: How was temperature included in the assessment of droughts?? This is unclear, and must be addressed. If you did not estimate different variables in the water balance (beyond precipitation), in what way was temperature included? Please clarify in the text.
R/ We clarified the computation of the water balance previous to ISPB obtention in the text
195-200: Here you imply that you have computed the water balance, even though you state the opposite just above.
R/ We clarified the difference between ISPP and ISPB
207: In this sentence you state “…values ranges are applied…”. Did you apply these value ranges, or did someone else? When you state that a certain thing IS done, this suggests that you are describing a generally accepted way of doing it. If you state that it WAS done, it becomes ambiguous whether it was you or someone else who did the action. You must therefore state explicitly that it was you (‘We did X’ or ‘In this study, X method was applied’). When this is made clear within a paragraph, it is then acceptable to include some statements in the form of ‘X was done’.
R/ Changed in all the text as suggested
214: This reviewer questions strongly the appropriateness and utility of using Moran’s I to evaluate the drought indices. Spatial autocorrelation should be used to inform sampling, not to test whether an index is valid. Please consider eliminating this part of the analysis, as it is not well justified and is not meaningful in result.
R/ Eliminated as suggested.
227: Please clarify that temporal variation in NDVI can result from changes in either vegetation condition (e.g. drought stress, phenology) or abundance (e.g. biomass, cover).
R/ Included as suggested.
228-231: Even though I was able to infer what is meant in these sentences after reading several times, many readers will struggle to understand. Please revise to improve clarity.
R/ It was changed as you and another reviewer suggested.
Results –
Figure 2: the placement of years on the X axes is not clear. The marks corresponding to the labelled years (e.g. 1950, 1957, etc.) should be longer or bolder than those of the other years. Consider 5-year intervals if possible. Also, figure text is too small relative to main text. Caption: please state full names of the indices in the caption (e.g. standardized precipitation index, etc.). Figures must stand on their own, not relying on information from the main text. Also, you state “all stations in the Yucatan Peninsula”, even though in the methods you describe using a subset with longer record periods. This is inconsistent. Please address this.
R/ Graph and caption modified as suggested.
Table 2, Section 3.3: Here you find that the coefficients are positive, which suggests that droughts became less frequent or severe over the past decades. This finding contrasts with what you state in the abstract, that these show evidence of climate warming. Generally, the result is not surprising, as climatic change is predicted to result in less precipitation in arid regions, and as increased rainfall in tropical areas. Maybe this is a result worthy of further discussion and emphasis.
R/ We included that sentence in the 3.3 section.
Discussion –
An essential part of this section should be comparison and contrast of your findings with a variety of other studies. You discuss other literature to a limited extent, and this is mostly about the particular phenomena observed in Mexico, rather than in the development and interpretation of tropical drought indices more broadly. Please augment this section accordingly.
R/ Information relative to tropical drought monitoring with indices has been added to the discussion, with references on Vietnam and Central American Dry Corridor.
Reviewer 2 Report
In the manuscript “Standardized Drought Indices for pre-summer
drought assessment in tropical areas”, Romero et al. proposed a new standardized drought indices for the evaluation of the pre-summer drought in tropical zone in Mexico. They tested two new indices: one elaborated from the precipitation and the other also considering the temperatures. These indices were validated and used to identify the most severe drought conditions in the Yucatan Peninsula and seems to show a good match with NDVI. I only have some editorial comments.
Line 60-63: I don’t understand why the rainy season can have a precipitation deficit with respect to evaporation in temperate zones.
Line 80: Add a very brief description of Ax’, Aw etc.
Line 96: Which months are pre-summer?
Line 107: Please add explanations of complexity of drought at different time of the year and how they impact agriculture.
Line 135: Please define the efficiency of precipitation.
Line 174: This information could come earlier (related to Line 96)
Line 212: Please spell out the acronyms.
Author Response
Thank you for your feedback, we describe below how we incorporate your suggestions:
Line 60-63: I don’t understand why the rainy season can have a precipitation deficit with respect to evaporation in temperate zones.
R/ In temperate zones with Df climate type, for which SPI has been developed (McKee et al 1993) the most rainy season occurs during summer but the amount of rain does not always compensate for the evaporation generated by the increase in temperatures, we clarify that in the text.
Line 80: Add a very brief description of Ax’, Aw etc.
R/ Description of climate types has been inserted in the text.
Line 96: Which months are pre-summer?
R/ Information about pre-summer end and duration has been added at the end of the 2.2.1. first paragraph. The information related to the comment of the reviewer “Line 174: This information could come earlier (related to Line 96)” has also been inserted.
Line 107: Please add explanations of complexity of drought at different time of the year and how they impact agriculture.
R/ We added the following text to explain it better: The study the onset of the rainy season is important because early (late) onsets tend to be associated with wetter (drier) May-June season, the first peak of the rainy season. So, having a late start of the rains, followed by a drier season in May-June in conjunction with a deep MSD in July and August, would affect significantly key socioeconomic sectors like agriculture during the “primera” period.
Line 135: Please define the efficiency of precipitation.
R/ As “efficiency of precipitation” can refer to other process, the expression has been changed to “capacity of precipitation events to recharge soil moisture”. The new expression is clearer and avoid confusion.
Line 174: This information could come earlier (related to Line 96)
R/ The information has been moved as suggested.
Line 212: Please spell out the acronyms.
R/ Acronyms were spelled in English translation
Reviewer 3 Report
The article is interesting somehow, but has nothing to do with hydrological science. Namely, the authors deal with the possibility of using different remote sensing products. They compared the images of remote sensing from various sources. But, there is no connection with field measurements and on-site observations. Authors developed some simple empirical models related only to remote sensing data. In doing so, the authors accepted several assumptions without specific hydrological analysis.
Uncertainty of the results is unknown and they only just look as something. The question is for what they can be used.
Author Response
We thank you for your comments, clarifications and improvement have been applied for validation methods and results.
The article is interesting somehow, but has nothing to do with hydrological science.
R/ We strongly disagree with this sentence. Notice also that this special number is related with hidroclimatology applications in Meso America. Since the develop of PDSI in 1948, drought indices have been used and published in important hydrological journals like Journal of Hydrology or Water Resources Research, se for example:
Asadi Zarch, M.A., Sivakumar, B., Sharma, A. (2015) Droughts in a warming climate: A global assessment of Standardized precipitation index (SPI) and Reconnaissance drought index (RDI), Journal of Hydrology, 526, doi:10.1016/j.jhydrol.2014.09.071
Jacobi, J., Perrone, D., Duncan, L. L., and Hornberger, G. (2013), A tool for calculating the Palmer drought indices, Water Resour. Res., 49, 6086– 6089, doi:10.1002/wrcr.20342.
Ma, M., Ren, L., Yuan, F., Jiang, S., Liu, Y., Kong, H., and Gong, L. (2014), A new standardized Palmer drought index for hydro‐meteorological use, Hydrol. Process., 28, 5645– 5661, doi: 10.1002/hyp.10063
Pedro-Monzonís, M., Solera, A., Ferrer, J., Estrela, T., Javier Paredes-Arquiola, J., (2015), A review of water scarcity and drought indexes in water resources planning and management, Journal of Hydrology, 527, doi:10.1016/j.jhydrol.2015.05.003.
Also, drought evaluation with indices is necessary for drought forecasting (Sene K. (2016) Droughts. In: Hydrometeorology. Springer, Cham. https://doi.org/10.1007/978-3-319-23546-2_10)
Namely, the authors deal with the possibility of using different remote sensing products. They compared the images of remote sensing from various sources.
R/ We clarified this point, add reference information relative to method and quality and gave reference of large time-series computed from NOAA-AVHRR images.
But, there is no connection with field measurements and on-site observations. Authors developed some simple empirical models related only to remote sensing data. In doing so, the authors accepted several assumptions without specific hydrological analysis.
R/ Please notice that data from remote sensing allows the validation of the developed indices applied to in situ measurements, comparing ISPP, ISPB and SPI, since drought indices were calculated from in situ data. Monthly average rainfall and temperature data for 52 meteorological stations of the National Meteorological Service of Mexico (SMN) network have been used.
The annual period of NDVI data have been reduced to April. This month corresponds to the end of the drought and represents better its evolution than a larger period, in consequence, the correlation coefficient increased. Although the correlation averages are low, a significant difference was detected between the traditional 12-month SPI and the other indices, evidencing the importance of restricting the computation period to the usual dry months. Also, high correlations were found for the most forested areas of the northern half of the peninsula, where hydromorphy is low and NDVI more dependent on weather conditions than on fluctuations in the underlying water table.
Uncertainty of the results is unknown and they only just look as something. The question is for what they can be used.
R/ The authors are sorry but do not understand the comment because the uncertainty is not assessable for the drought indices. Regarding NDVI extraction using AVHRR images, the margin of error reported in the literature is 10% (Goward et al 1991). In the updated version of the document, the P-value is calculated for the trend coefficients.
Reviewer 4 Report
The authors should provide more detailed information in the part of results.
In page 5, from line 168 to line 177, there are two “First,” which are very confusing for the description of the method. Please correct it.
In page 5, please explain what are P and T in line 170.
In page 6, please rewrite the second paragraph in the session of 2.4 “Validation of the indices”.
In page 8, the values of correlation coefficients from table 3 are all lower than 0.4, which are very low. Some scatter plots may be useful to show the relationship between two variables.
Author Response
We thank you for your comments.
Comments and Suggestions for Authors
The authors should provide more detailed information in the part of results.
R/ We hope that improvements done in the manuscript, following the suggestions of the reviewers, gave more information about the results. Information have been added relative to trends and validation. Extensive information has not been added as other reviewers reported that the document is lengthy.
In page 5, from line 168 to line 177, there are two “First,” which are very confusing for the description of the method. Please correct it.
R/ The redaction of the paragraph has been modified according to your suggestion. The methodological steps are clearer.
In page 5, please explain what are P and T in line 170.
R/ P and T were explained “, where T is the average monthly precipitation (mm) and T the average monthly temperature (°C)”
In page 6, please rewrite the second paragraph in the session of 2.4 “Validation of the indices”.
R/ Section 2.4 has been rewritten and reorganized according to your suggestion.
In page 8, the values of correlation coefficients from table 3 are all lower than 0.4, which are very low. Some scatter plots may be useful to show the relationship between two variables.
R/ The annual period of NDVI data have been reduced to April. This month correspond to the end of the drought and represent its evolution, in consequence, the correlation coefficient increased. Although the correlation averages are low, a significant difference was detected between the traditional 12-month SPI and the other indices.
Also, we highlighted the high correlations found for the most forested areas of the northern half of the peninsula, where hydromorphy is low and NDVI more dependent on weather conditions than on fluctuations in the underlying water table.
As we have 52x4 correlations calculated between time series of indices and NDVI, a scatterplot could not give information on the relationship between the large variables, we would have to choose the stations to represent, which could be considered as bias.
Round 2
Reviewer 1 Report
The authors have made substantial revision to the manuscript, and it is much improved. Please consider the following minor comments:
Abstract: its correct just to state “precipitation” or “temperature” rather than “the precipitation” or “the temperature”, unless you are referring to specific values.
19: correct “lead the authors” rather than leaded
144: as mentioned in the first review, you state that the images were from “various platforms and sensors”. This will confuse some readers, because you state also that you are using AVHRR images, which are generated by one particular type of sensor, but with two platforms. Consider eliminating this sentence.
150: Please consider different word than “elaboration”, here and elsewhere. See English definition: “the addition of more information to or an explanation of something that you have said. Examples:
- This point needs greater elaboration.
- His photographs require no written elaboration - everything is in the images themselves.”
Author Response
The authors have made substantial revision to the manuscript, and it is much improved. Please consider the following minor comments:
R/ Thank you for your feedback, we describe below how we incorporate your suggestions:
Abstract: its correct just to state “precipitation” or “temperature” rather than “the precipitation” or “the temperature”, unless you are referring to specific values.
R/ Changed as suggested
19: correct “lead the authors” rather than leaded
R/ Changed to “leads”
144: as mentioned in the first review, you state that the images were from “various platforms and sensors”. This will confuse some readers, because you state also that you are using AVHRR images, which are generated by one particular type of sensor, but with two platforms. Consider eliminating this sentence.
R/ Eliminated as suggested
150: Please consider different word than “elaboration”, here and elsewhere. See English definition: “the addition of more information to or an explanation of something that you have said. Examples:
R/ Changed to “computation”, also “elaborated” line 292 changed to “calculated”
Reviewer 3 Report
Describe the phenomenon of droughts accurately enough with only one index, which should be practically useful; it is not possible. The article also proves this The authors view the position given in the conclusions ". Particularly, the application of the indices allowed the identification of the most problematic drought years in the Yucatan Peninsula, as well as the temporal evolution of the phenomenon." have not proved.
Author Response
Describe the phenomenon of droughts accurately enough with only one index, which should be practically useful; it is not possible. The article also proves this
R/ We agree with the comment, the statement “The description of droughts accurately enough with only one index is not possible [16] although it could be practically useful.” has been added to the text
The authors view the position given in the conclusions ". Particularly, the application of the indices allowed the identification of the most problematic drought years in the Yucatan Peninsula, as well as the temporal evolution of the phenomenon." have not proved.
R/ The sentence has been modified as suggested. ”Particularly, the application of the indices could allow the identification of the most problematic pre-summer drought years in the Yucatan Peninsula [58,59], as well as the multi-year trend of the phenomenon.”
