Evaluating Effects of Medium-Resolution Optical Data Availability on Phenology-Based Rice Mapping in China

Round 1

Reviewer 1 Report

The manuscript "Evaluating effects of data availability on phenology-based rice 2 mapping in China" is interesting and coherent with topics of the Journal.  

Some necessary corrections emerge from the revision, highlighted on the manuscript.

The conclusion must be improved.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

General comments:

This study discusses the effects of data availability on phenology-based rice mapping in China, which provides guidelines about how to obtain enough valid observations for rice mapping. Since the authors mainly evaluates the data availability of Landsat 7/8 and Sentinel-2 images, it would be more appropriate to change the “data availability” in the title as “medium-resolution optical data availability”.

The main issues I concern are: (1) the target application is phenology-based rice mapping in this study, which means that all discussions of data availability are preferably based on rice growing areas; (2) the authors are expected to give a clear definition of effective identification and the minimum requirement of the number of valid observations. 

Detail comments:

Page 2, Line 62: suggest changing “LSWI+α>EVI/NDVI” as “LSWI+α>EVIorNDVI”.

Page 2, Line 64: please add citations.

Page 3, Line 130: I don’t think the involvement of Sentinel-1 is necessary in this study.  

Page 3, Line 138: “scape up” to “scale up”.

Page 3, Line 148: “Over 10 years of observations during 2011 – 2013” to “Over 10 years of observations during 2001 – 2013”?

Page 5, Line 191-209: The definition of time window is important and relevant to the results in this study. The authors separate the time windows for the north and south, effectively accounting for regional differences. However, in my opinion, the selection of time window based on six rice-growing sub-regions (as shown in the Figure 4) would be better because these regions have different environmental backgrounds.

Page 6, Line 217-243: Based on the equations, I am afraid the calculated results cannot represent the total valid observations of Landsat and Sentinel-2 during the flooding windows of single rice, early rice, late rice because large amounts of non-rice observations are included.

Combining with the figure 3, the results are overestimated in my perspective given that some regions with more than 20 valid observations locate at arid/semi-arid environments where paddy rice cannot be planted. Thus, the authors are expected to filter out non-rice pixels firstly based on some reliable products to make results more convincing.

Page 7, Line 296: The authors need to define what is effective identification and what is the minimum number of observations that can achieve effective identification of single/early/late rice field respectively.  

Page 7, Line 299: It is hard to understand why Sentinel-1 is necessary for late rice mapping based on Figure 3. There is no significant difference in the percentage of identifiable pixels among single, early and late rice.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Dear Authors,

The manuscript is well prepared and presents an effective methodology and exploratory analysis for assessing the effects of data availability on phenology-based rice mapping in China. Results reveal how important the availability of data is for the assessed context, and indicates ways for developing a national-scale effort for mapping rice in China. For me, it is an excellent initiative and is on the way to being published. Therefore, I would like to present some adjustments that can help to improve the study:

Based on the interesting Results, the Discussion section can be improved on three topics to expand the phenology-based range of analysis:

The potential and capability of the improvements in data availability to separate rice from classes traditionally confused with rice. Did the increase in temporal resolution solve confusion with similar classes? Also, what input features can be benefitted from the increasing valid number of observations to identify rice in China?

The limitations found and potential strategies to overcome the presented adversities.

Edaphoclimatic and crop management practices may be important conditioning factors for phenology-based crop mapping. For territorial planning purposes, there are variations of land management practices on a seasonal scale in China? As cited in Lines 192-193, rice systems differ. Given this, are crop and land management (especially crop calendars) or edaphoclimatic conditions significantly varying among regions? If yes, strategies for the temporal stratification of planting/sowing dates (Liu et al., 2020; Chaves et al., 2021*; He et al., 2021*; Li and Lei, 2021*) can be discussed as strategies to overcome differences in crop calendars and optimize the search for available remote sensing data.

* Liu et al. (2020). A new framework to map fine resolution cropping intensity across the globe: Algorithm, validation, and implication. Remote Sensing of Environment, 251, 112095. https://www.sciencedirect.com/science/article/pii/S0034425720304685.

*Chaves et al. (2021). CBERS data cubes for land use and land cover mapping in the Brazilian Cerrado agricultural belt. International Journal of Remote Sensing, 42(21), 8398-8432. https://www.tandfonline.com/doi/abs/10.1080/01431161.2021.1978584.

* He et al. (2021). Examining rice distribution and cropping intensity in a mixed single-and double-cropping region in South China using all available Sentinel 1/2 images. International Journal of Applied Earth Observation and Geoinformation, 101, 102351. https://www.sciencedirect.com/science/article/pii/S0303243421000581.

* Li and Lei (2021). Tracking the spatio-temporal change of planting area of winter wheat-summer maize cropping system in the North China Plain during 2001–2018. Computers and Electronics in Agriculture, 187, 106222. https://www.sciencedirect.com/science/article/pii/S0168169921002398.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Thank authors for addressing my questions. This paper gives other important message on the data issue for rice mapping.