Impacts of Climate Change and Human Activities on Runoff Variation of the Intensive Phosphate Mined Huangbaihe River Basin, China

Round 1

Reviewer 1 Report

After 2 round of review, I think the paper can be published.

Author Response

We are very grateful to the Reviewer for the positive response!

Reviewer 2 Report

Review Bo et al. Impacts of climate change and human activities on runoff variation of the intensive phosphate mined Huangbaihe River Basin, China

 

This manuscript presented a study that attempted to investigate the influence of climate variability and human activities on watershed processes mainly on runoff. I find the issue at hand interesting and worthwhile investigating because it will help inform decision makers and water managers about the influence phosphate mining on the water resources and availability. However, I believe the manuscript requires a major work on improving the organization and the language before accepting for publication. I found the manuscript difficult to understand due to the poor language and not good enough results reporting and discussion. I have outlined below a few major and minor points that need to be addressed.

 

Major comment:

The authors reported no significant change in the trend of precipitation and runoff. Yet evapotranspiration (ET) showed significant trend change. On the other, the watershed is dominated by forest and there is no as such considerable shrinkage in forest cover between 1980 and 2008. I would like to underscore here precipitation is the only locally measured data, even though there is no good network of gauges for the entire study period. There is also measure discharge that is useful especially before the dam construction. Therefore, I argue that the authors finding about the statistically significant trend change in ET could be due to a data issue. How do you justify this? GLEAM is a model output that uses different datasets including remote sensing products. I understand the lack of data but why didn’t you try also MODIS ET with a much better spatial resolution and some studies in Asian reported MODIS ET is good for forest covers. And other available ET products. That I think would help to give some confidence to the trend analysis. I wonder how deep the mining activities have been taking place. Because there is no any appreciable change at the surface as reported in land use classes from 1980, 1995 and 2008. The authors discussed this in 435-440. This tells me that the surface processes that controls runoff generation are stable. And the variation could come from the input rainfall. Much of the changes related the mining is taking place underground and it would be more interesting if the authors give focus on the ground water component of the river flow – baseflow. I’m not sure if the SWAT model could handle that but I would expect consider change in that part of the flow. The baseflow could also contribute further downstream of TFM gauge, meaning the increased in recharge due to the rock fractures and porosity would not be visible at Tianfumian site. The authors discussed this in line 452 -467. It would be great if you could show this using SWAT model parameters that controls recharge and ground water flow. The manuscript requires further consolidation from literature and the language needs to be improved. Rest of the manuscript also requires improvement.

 

Minor

# Line 60-01. Rephrase this sentence

# line 60 remove ‘affecting’  

# line 91 “In this study, the studied area” improve the language  

#line 125 add reference

# 187  I assume you selected the parameters with their sensitivity rank.  You may improve the language further.

#263 add ‘in’ after “change” and it reads as “The change in trend and magnitude…”

#263 Remove the word “annual’ before precipitation and evapotranspiration

#263-271 please improve the reporting of this paragraph. It might be difficult for a reader with less statistical background.

#321 capitalization “It” to it

#331 Table 3, what is the difference between the two ET values. Check the foot note ‘b’ .

#393 is good explanation why SOL_BD is the most sensitive SWAT parameter for the study area? That is not often the case in many of SWAT related sensitivity analysis in the literature.

#452-454 Check this sentence, it is vague and unclear.

#459 replace “interception” by “retention”

#468 Table 9, why the water area decreased in 2008? There reservoirs are still there.

 

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Thank you for addressing most of my comments!

Author Response

We are very grateful to the reviewer for the positive response!

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

The paper has been significantly improved at this time. However, the following issues still need to be addressed.

Major comments:

How was the mining activities considered in SWAT model?

Instead of driving evapotranspiration based on precipitation, the authors could have used alternative sources such as remote sensing products (e.g. MODIS).

Since the authors used SUFI-2 for SWAT calibration, they need to report percent of observations bracketted at 95% confidence and corresponding unceratinity band width.

As the used SWAT model handles dams/reservoirs, why the authors did not implement dams/reservoirs in their modleing?

The authors used climate change and variability but there is significant difference between them. Did they assess climate change or climate variability?

Was the trend change really due to mining activity? Some years e.g before 1985 don't follow the pattern of rainfall. This requires strong evidence and statistical test to see if there is significant change due to phosphate mining.

The manuscript needs significant language improvement

Minor comments:

Ln 17, replace "had" with "have been"

Ln 24, replace "due to" with "since" and "here" with "this study"

Ln 34, delete "from this paper"

Ln 46-52, mining or groundwater mining? Also, the sentences are not clear. Please consider re-writing

Ln 61, provide reference

Ln 69, provide references for each models

Ln 70, replace "has" with "have"

Ln 76, replace "found that" with "that reported"

Ln 92, replace "and its" with "with drainage"

Ln 109, delete "study"

Ln 116, basin or catchment? be consistent just use one term throughout the text

Ln 126, delete "from"

Ln 141-142, suggest to delete

Ln 150, delete "the researchers"

Ln 150, and throughout the text remove the year after "et al."

Ln 188, spell out NSE and RMSE in the first call

Ln 209, define variables immediately after the equation

Ln 229, high uncertainties. Better to use alternative data sources

Figure 2, combine the panels into one for easy comparison

Ln 249, replace "decades" with "decadal"

Ln 251-22, why? You also said affected by another factor in 2007?

Table 1, add runoff-rainfall ratito for additional information

Table 2, change "N" to "ns or full name"

Ln 280-281 and Figure 3, I don't think that holds true since some years before 1985 don't follow the rainfall patterns? E.g. what explains the noticebale decline in runoff in 1995, 2013, 2015 or before 1985?

Ln 286, MODIS or other sources?

Ln 303-305, still lower than Tb, which was probably due to precipitation decrease

Ln 306-311, looks like comparable. Are they significant values?

Table 3, almost the same fraction of forest cover but noticeable difference in PET. Please precisely explain

Figure 4, where is ET?

Ln 322, how did you incoporate anthropogenic activities in SWAT model?

Table 4, please add description of parameters and sensitivity rank, including what r_ and v_ refers to

Add p-factor and r-factor in Table 5, evaluate the performance of the model

Figure 6, what do we learn from the figure. Runoff-rainfall is linearly related that probably indicates less impact of mining activities. Maybe you need to plot runoff against rainfall

Ln 371, runoff or streamflow?

Ln 395-397, consider re-writing. You started with "while" ended with one phrase

Ln 418, replace "be" with "been"

Ln 445, "....cracks and the interception by goafs" please clarify how did consider these processes during modeling?

Ln 456, use the p-factor and r-factor info here

Ln 462-476, why did not you use reanalysis data or at least the elevation band in SWAT?

Ln 488, delete "the" after "show"

Ln 489, change "was" to "were"

Ln 502-503, why did not you analyze in this study as SWAT provides those components?

Reviewer 2 Report

The paper improved a lot. Here just a few minor points.

1. Lines 250 - 251: the authors state that the correlation between precipitation and runoff is positive or negative. What is meant is that the difference between the precipitation and its overall mean as well as the difference between the runoff and its overall mean both have the same sign (positive correlation) or a opposite sign (negative correlation). I think the term correlation is not appropriate. A negative correlation would mean that there is a causal relation such that the runoff decreases with increasing precipitation. I don't think that is true.

2. In line 313 the Standard deviation is mentioned, but it is absent in table 3 (and elsewhere). Either present the STD in the table or remove note. 

3. Before you can conclude that the phosphate mining is the cause the change (line 434), you must exclude other possible causes. I understand that there are no other possible causes, but it should be mentioned in the text.

4. I not familiar with the method of Zhang et al. I wonder whether the values 1410 and 1100 are fixed values in the method or case specific. In the latter case the reader should know how to determine these values.

Reviewer 3 Report

After reading the article, I have the following suggestions for the authors.

Please check the manuscript for typos and grammar mistakes, as. for example:

- line 34: These results from this ... should be The results from....   

- line 44: [2- 44 5].Direct - It should be a space before Direct

- line 59: There are a large number - should be: There is a ...

 - line 67: The hydrological model are used - should be The hydrological model is used ...

- lines 69-72 must be reformulated

- lines 75-76 There 76 are few studies found - incorrect. Some studies found that...

-line 242- much less - maybe significantly smaller

and so on.

   2. References are double numbered, Please remove one of the numbers.

There are serious methodological misunderstandings, as, for example, those related to wrong terminology and misunderstanding of statistical tests, as for example:

   3. lines 145-146: Mann -Kendall trend test is designed to test the hypothesis of the existence of a monotonic trend of the series against the inexistence of such a trend, not for testing the stationarity. Please see the original articles:

Mann, H.B. 1945. Non-parametric tests against trend, Econometrica 13:163-171.

Kendall, M.G. 1975. Rank Correlation Methods, 4th edition, Charles Griffin, London.

  4. the series relationships - point of view of statistics the terminology is autocorrelation .

   5. lines 146-157. It seems that the authors are not familiar with the statistical terminology and proper applications of the tests. Please refer to the original works of Mann, Kendall and Sen, not to some interpretations.

  6. line 159: the more confident change point - what does it mean? Maybe most probable...

  7. line 254: why 5-point moving mean has been used and not 4 or 6. The correct terminology is moving average, not moving mean

  8. The trend of annual runoff and annual precipitation is also detected using Mann-Kendall method and trend magnitude is used Sen’s slope. - Incorrect from grammar and scientific viewpoints.

  9. line 266: Inflexion point has a different significance.

The examples can continue. 

The methodology used in this article is almost the same as that from 

Zhang, A.; Fu, G.; Wang, B.; Bao, Z.; Zheng, H. Assessments of Impacts of Climate Change and Human Activities on Runoff with SWAT for the Huifa River Basin, Northeast China. Water 534 Resources Management 2012, 26, 2199-2217.

Based on these observations and the wrong interpretations of some of the results presented in the tables, I suggest rejecting the article.