Patterns in data of extreme droughts/floods and harvest grades derived from historical documents in eastern China during 801–1910

In China, historical documents record a large quantity of information related to climate change and grain harvest. This information can help to explore the impacts of extreme drought or flood on crop production, which can provide implications for the adaptation of agriculture to higher-probability extreme climate in the context of global warming. In this paper, reported extreme drought/flood chronologies and reconstructed grain harvest series derived from historical documents were adopted in order to investigate the association between the reported frequency of extreme drought/flood in eastern China and reconstructed poor harvests during 801–1910. The results show that extreme droughts were reported more often in 801–870, 1031–1230, 1481–1530, and 1581–1650 over the whole of eastern China. On a regional scale, extreme droughts were reported more often in 1031–1100, 1441–1490, 1601–1650, and 1831–1880 in the North China Plain, 801–870, 1031–1120, 1161–1220, and 1471–1530 in Jianghuai, and 991–1040, 1091–1150, 1171–1230, 1411–1470, and 1481–1530 in Jiangnan. The grain harvest was reconstructed to be generally poor in 801– 940, 1251–1650, and 1841–1910, but the reconstructed harvests were bumper in 951–1250 and 1651–1840, approximately. During the entire period from 801 to 1910, the frequency of reporting of extreme droughts in any subregion of eastern China was significantly associated over the long term with lower reconstructed harvests. The association between reported frequency of extreme floods and reconstructed low harvests appeared to be much weaker, while reconstructed harvest was much worse when extreme drought and extreme flood in different subregions were reported in the same year. The association between reconstructed poor harvests and reported frequency of regional extreme droughts was weak during the warm epoch of 920–1300 but strong during the cold epoch of 1310–1880, which could imply that a warm climate could weaken the impact of extreme drought on poor harvests; yet other historical factors may also contribute to these different patterns extracted from the two datasets.

2) There is a clear and well-visible improvement in the 2.1 Data subchapter with examples from original source references and also with listing major advantages of the compiled dataset and indices they apply. However, the authors could also address any weaknesses of their databases, as their data are mainly available in compilations, and usually compilations are based on all kinds of contemporary and non-contemporary data mixed, with the (sometimes high) probability of misdating, doubling (tripling) events etc. Consequently, majority of the Chinese database, the paper is based on, is only available in a form that retracing the sources and the reconstruction steps ate not possible. Naturally, the authors cannot solve this problem all within this article; nevertheless, in one sentence (e.g. on page 4, btw. lines 20-25, after listing all advantages of the dataset) they should refer to this general problem or weakness.
Accepted. The weakness about this database has been added. (P4, L30-32) However, this dataset also has weaknesses. For example, the reconstruction derived from the historical documents relies highly on the accuracy of the compilation, and is only available in a form that retracing of the original sources and reconstruction steps are not same as reconstructions from natural evidences.

To Anonymous Referee #1:
The authors have accepted the key changes requested in the previous reviews. However, they have not been entirely consistent in applying those changes. It is important that throughout the text the authors carefully distinguish (1) the analysis of data in these databases of historical weather and harvests from (2) their interpretation of the underlying climatic and human history. Therefore, I would request the following minor revisions: The abstract has added terms such as "reported" but needs to rework the language more carefully to reflect the new approach of the article. The following changes would be appropriate in order to make the meaning more precise and avoid inaccurate claims: -"reported extreme droughts [floods] occurred" should be: "extreme droughts [floods] were reported" (or a similar phrase). The key distinction here is a report of a drought [flood] is not the same as the historical occurrence of drought [flood].
-"reconstructed grain harvest was poor [medium, high] in" should be: "the grain harvest was reconstructed as poor [medium, high] for" (or a similar phrase).
-"occurrence of reported extreme drought in any sub-region of eastern China was significantly associated with reduced harvests in the long-term average" should be: "frequency of reporting of extreme droughts was significantly associated over the long term with lower reconstructed harvests" (or a similar phrase).
-"association between harvest and extreme floods" should be: "associated between the reported frequency of extreme floods and reconstructed low harvests" (or a similar phrase).
-"other social factors" should be "other historical factors" to include other historical environmental changes, both natural and anthropogenic Accepted. All related expressions have been revised throughout the text to distinguish "the analysis of data in these databases of historical weather and harvests" from "their interpretation of the underlying climatic and human history".
On page 2, lines 20-34 have not been reworked to reflect the new approach of the article --that is, to first discuss patterns in the data derived from the historical documents, and only then to discuss the interpretations of those patterns as historical climate impacts. This paragraph makes the unwarranted assumption that the reported frequencies of events in the historical documents represent real frequencies and that associations between the frequencies of reported disasters and extremes and variations in reconstructed grain harvests represent causation (i.e., climate impacts). The studies In the sentence at the bottom of page 4 and top of page 5, it is important that the authors clarify that their method is based on an assumption that the probabilities for omitting drought and flood events in reporting and transmission of historical records were random and unbiased, for the reasons they have stated. They haven't actually proven that omissions were random and unbiased. They have merely made a reasonable argument that it would be appropriate to proceed on this assumption. Now that the authors have worked to distinguish their results (that is, the patterns and associations in their datasets) from their discussion (that is, the climate and historical interpretation of those associations), their use of a combined "Results and Discussion" section has become more confusing. For the sake of clarity, I would encourage the authors to rename section 3 as simply "Results" and turn subsection 3.4 into a new section, "Discussion." Their discussion of results in the bottom of page 9 to the top of page 10 as well as their discussion of results at the top of page 12 might then be moved into the new "Discussion" section, as part of the authors' climatic and historical interpretation of the patterns in the data. The fact that these pattern makes sense from a meteorological perspective supports the case that droughts had a significant historical impact on grain harvests. The authors may wish to state this point clearly in their discussion.  (Zhang, 1982).
Although it should be noted that these meteorological interpretations only apply for short-term extreme flood events. When it comes to long-term wet periods, it could as well be responsible for bad harvest, due to the low-temperature and short sunlight.  Zhang et al. (2007), limited resources could also cause social turbulence, such as famine, peasant uprising, the outbreak of war, and population decline, all of which may further increase agricultural vulnerability. Therefore, even though the frequency of reporting of extreme droughts was slightly higher in the warm period of 920-1300, the frequency of reconstructed poor harvests did not increase significantly.
The revised sections, while mostly clear, should receive further review for correct English grammar and word use before publication.
Accepted and revised.