Albedo Impacts of Changing Agricultural Practices in the United States through Space-Borne Analysis
Round 1
Reviewer 1 Report
This paper presents interesting findings on the dynamics of albedo during the growing season of major field crops grown in the lower 48 US. Database provided and suggestions regarding how to properly use NDVI as proxy for albedo certainly useful for the scientific community. The authors should address questions and concerns listed below and revise the paper accordingly in order to make it fit for publication in this journal.
Figure 2. The term high quality for the post-filtered data is ambiguous. The author should consider change it to a clearer tem such as ‘filtered’ or other similar terms.
Line 170, please use the proper term that is “day of year” instead of “Julian day". “Julian day” is continuous count of days since the arbitrarily selected point of the year 4713 b.c. (Merriam-Webster Dictionary) the data used in this study are clearly day of year not Julian day.
Figure 3. The map projection used is not ideal as it display lower 48 US in distorted form. The authors should consider using USA_Contiguous_Albers_Equal_Area_Conic, or USA_Contiguous_Lambert_Conformal_Conic or USA_Contiguous_Equidistant_Conic map projection to avoid distortion of area and distance. And also scale bar and north arrow are missing in the maps. The outline of the lower 48 US with appear to be black color (which is also too thick) can mislead the readers as the points selected for developed land use is also black.
Figure 4. Please see comment about figure 3 regarding map projections. Same applied here.
It is not clear why spring wheat, instead of winter wheat was selected among the 4 main crops shown in Figure 4 given winter accounts for 70-80 percent of total wheat produced in the U.S. (USDA ERS).
I am not sure how useful Table 1 for most readers. What are we supposed to see there?
It also appears that Figure 5 already present the same data.
Line 193-194. Please use ha for area unit instead of acres
Line 225-226. It is not clear here the discussion on the BSA changes refers to which wavelength? If it refers to 634 nm, then the authors made a mistake here, for cotton the 645 nm channel actually changes from 0.15 at day 150 (planting) to 0.10 at day 290 (harvest) and not the other way around.
Figure 7 & 8. The term ‘Year-by-year variation’ in the captions are rather misleading. The text suggest that annual pattern of BSA was relatively consistent across different years and therefore the more suitable term in the caption should be ‘Annual variation in albedo across multiple years’
Table 2. I don’t know if this table is necessary. The authors mentioned rice area accuracies in the context of uncertainties in the crop layer dataset used but it is not clear in the present form of the paper how the authors actually used the accuracy data for further analysis.
Line 336. Figure 22 not found in the paper.
Line 356-357. Please provide reference for the statement about double cropping maize practice found in southern Asia. Why not mention double cropping rice, which is a more commonly found throughout tropical Asia.
Conclusions section is too long and written just like a plain summary rather than presenting concluding thought about the study and plus why mention about database provided as supplement in the conclusion, it was already mentioned in the main text.
Author Response
Please see the attachment
Responses to reviewer 1
This paper presents interesting findings on the dynamics of albedo during the growing season of major field crops grown in the lower 48 US. Database provided and suggestions regarding how to properly use NDVI as proxy for albedo certainly useful for the scientific community. The authors should address questions and concerns listed below and revise the paper accordingly in order to make it fit for publication in this journal.
We thank the reviewer for his/her constructive comments.
Figure 2. The term high quality for the post-filtered data is ambiguous. The author should consider change it to a clearer tem such as ‘filtered’ or other similar terms.
We have changed the description in the figure to “Quality Controlled” to better describe the selected pixels. (Fig 2)
Line 170, please use the proper term that is “day of year” instead of “Julian day". “Julian day” is continuous count of days since the arbitrarily selected point of the year 4713 b.c. (Merriam-Webster Dictionary) the data used in this study are clearly day of year not Julian day.
We have replaced “Julian day” with “day of year” throughout the paper to avoid any confusion (Multiple Lines)
Figure 3. The map projection used is not ideal as it display lower 48 US in distorted form. The authors should consider using USA_Contiguous_Albers_Equal_Area_Conic, or USA_Contiguous_Lambert_Conformal_Conic or USA_Contiguous_Equidistant_Conic map projection to avoid distortion of area and distance. And also scale bar and north arrow are missing in the maps. The outline of the lower 48 US with appear to be black color (which is also too thick) can mislead the readers as the points selected for developed land use is also black.
We have revised the figure to use the “USA Contiguous Albers Equal Area Conic” (ESRI 102003) projection. Additionally, we added a north arrow and scale bar, narrowed the US outline thickness, improved resolution of the outline, as well as modified the “developed” category to use a distinct color. (Fig 3)
Figure 4. Please see comment about figure 3 regarding map projections. Same applied here.
We have revised the figure to use the “USA Contiguous Albers Equal Area Conic (ESRI 102003) projection. We have also added a north arrow and scale bar to this figure and revised the line thicknesses. (Fig 4)
It is not clear why spring wheat, instead of winter wheat was selected among the 4 main crops shown in Figure 4 given winter accounts for 70-80 percent of total wheat produced in the U.S. (USDA ERS).
We have included spring wheat in this figure specifically as we highlight this crop in later figures due to the overlap with maize and soybean in HZ4.
I am not sure how useful Table 1 for most readers. What are we supposed to see there?
Table 1 is used as a quick reference to show the amount of change in albedo that can be expected during the growing season for readers who do not want to process the database
It also appears that Figure 5 already present the same data.
Figure 5 is provided for an example of the daily variation in albedo for the four highlighted crops.
Line 193-194. Please use ha for area unit instead of acres
We have replaced acres with ha in this section (L 198)
Line 225-226. It is not clear here the discussion on the BSA changes refers to which wavelength? If it refers to 634 nm, then the authors made a mistake here, for cotton the 645 nm channel actually changes from 0.15 at day 150 (planting) to 0.10 at day 290 (harvest) and not the other way around.
We have updated the section and changed the order of the numbers to clarify the albedo response. (L 229-231)
Figure 7 & 8. The term ‘Year-by-year variation’ in the captions are rather misleading. The text suggest that annual pattern of BSA was relatively consistent across different years and therefore the more suitable term in the caption should be ‘Annual variation in albedo across multiple years’
We have updated these two figure captions to the suggested wording (Figure 7&8 captions)
Table 2. I don’t know if this table is necessary. The authors mentioned rice area accuracies in the context of uncertainties in the crop layer dataset used but it is not clear in the present form of the paper how the authors actually used the accuracy data for further analysis.
We would still like to document user accuracy of the CDL data for users who are interested
Line 336. Figure 22 not found in the paper.
A small section of text was erroneously excluded and replaced with “figure 22,” this missing text has been restored. (L 345-347)
Line 356-357. Please provide reference for the statement about double cropping maize practice found in southern Asia. Why not mention double cropping rice, which is a more commonly found throughout tropical Asia.
We have updated the sentence to a more generic form as follows: “For example, planting date and irrigation practices of crops in regions that often double crop (such as those seen in the southern Asia region[28]) can vary greatly from management practices found in the United States.” and added a reference highlighting the prevalence of multi-cropping practices common to southern Asia. (L 367-369)
Conclusions section is too long and written just like a plain summary rather than presenting concluding thought about the study and plus why mention about database provided as supplement in the conclusion, it was already mentioned in the main text.
Some readers only read the abstract and the conclusion sections. We include as much information in the conclusion section as possible so it contains needed information for those readers.
Reviewer 2 Report
Review of Starr et al. (submitted to Remote Sens.)
This is definitely interesting and publishable work, but only after some revisions suggested below.
L11: Cropland Data Layer (CDL) needs to be introduced since the name itself is too vague. What kind of cropland data is this? Is it simply a map of crop types? Who produces it?
L13: Julien -> Julian
L36: “solar” should be removed from the definition of albedo
L39-40: provide a reference for daily changes
L70: again, CDL is discussed without mentioning who generates this data (until L84). The term needs to be introduced here (or before this point).
L74: Add a comma after “From this collocation” and all leading prepositional phrases as correctly done on L70, for example.
L93: “A kernel driven approach” only means something if you read the cited paper. It should not be this way. This paper should be intelligible without consulting another paper. I am fine with dropping “kernel driven approach” since it essentially adds nothing. Is the kernel part of Eq. 2 of the manuscript? If so, “(see equation below)” could be added after “kernel driven approach”.
L97: add ‘(diffuse)’ after ‘indirect’ so that the SR subscript makes sense to the reader.
L104: 5000 nm is part of the shortwave? That seems to be too far into the mid-IR, but I trust that this is not a typo.
L116: I think this sentence is misleading. I understand that the CDL does not provide reflectances, just crop type or land use, but the sentence implies otherwise. The second half of this sentence “to derive…” belongs later in the paragraph (i.e. next sentence or two).
L135: profiles -> profile
L135: I am quite sure this is not a grammatically correct sentence. The last part of the run-on sentence starting with “thus” needs to be reworded.
L137: What is the pointing uncertainty of MODIS? If it is more than 30 metres, I wonder if this should be mentioned as a source of error. I think the pointing uncertainty should be mentioned either way in Sect. 2.4, especially if there is a systematic (non-random) component to pointing error.
L163: Add “were” before calculated.
L172: is the comma between “crop” and “land” supposed to be there?
L185: Delete “of which”
L188: “non-field crop” -> “non-field-crop”
L189: for -> to
L190: “at” -> “for” or “in”
L198: Remove “.” After “Figure 5”
L199: seven -> three
L200: Trim wavelength list.
L212: “due to …” is a conclusion that is not supported in this paper by direct evidence, but is almost certainly correct.
L215: following the peak growth period, the albedo of maize crops is noticeably higher (0.15-0.2) than at the start of the growing season (0.1). I wonder if there are other factors (rainfall, soil moisture) besides the factors mentioned by the authors.
L224: The sentence “Still …” needs to be reworded. Perhaps delete “, was found”.
L226: Another possibility is that cotton is more white (i.e. brighter) when viewed from above. This could explain the higher albedo than the 3 other main crops and the lack of seasonal variation. The foliage of the cotton plant is hidden to some extent by the white cotton.
L238: I believe there are other substances in plants that absorb in the blue portion of the spectrum, in addition to chlorophyll.
L239: There is a space between ‘nm’ and the numeric value. This is how it should be. The rest of the paper butts the units to the numeric value.
L265: I suggest that “Whereas” is deleted from this sentence. I think this word is misused here.
L324: Define NASS acronym.
L329: Reword “Areas that appear noisy”. Is the “noise” in the albedo time series?
L336: “Figure 22. (…)”. There is no Figure 22 and a period should not follow ‘22’. The sentence also does not make sense.
L337: This following sentence comes out of nowhere. I believe a chunk of text is missing. The authors need to ease the reader into this discussion of cirrus, which currently seems disconnected from the paper. Also, there are a couple of sentences starting with “Still,” and I am not sure what is meant by “still”. “Nevertheless” seems more appropriate.
L369: index -> index,
L370: The authors should tell the readers for which wavelengths previous researchers have tried to exploit the NDVI-albedo relationship. I doubt anyone would expect this relationship to hold from 470 to 1600 nm.
L387: Here, the reader should be told how this new database is different than an albedo database for homogeneous, non-barren conditions.
L420: “but may be due to…” What follows this excerpt seems like an explanation only for the harvesting season. It does not seem to explain the behaviour in the peak growing season. Thus, this sentence should be rewritten to clear this up.
L429: ; -> ,
L439: “to economic modeling of crop acreage response functions”. I have no idea what this term means and I doubt whether many in the remote sensing community will also. To which response are the authors referring?
Author Response
Please see the attachment
Responses to reviewer 2
This is definitely interesting and publishable work, but only after some revisions suggested below.
We thank the reviewer for his/her suggestions and comments.
L11: Cropland Data Layer (CDL) needs to be introduced since the name itself is too vague. What kind of cropland data is this? Is it simply a map of crop types? Who produces it?
We have updated the description to remove any confusion over which product we are referring to. (L 11)
L13: Julien -> Julian
Based on other reviewer’s comments we have replaced “Julian” with day of year to clear up confusion (Multiple Lines)
L36: “solar” should be removed from the definition of albedo
We have removed the word “solar” from this sentence (L 37)
L39-40: provide a reference for daily changes
We have removed references to daily changes in this sentence as well as the next. (L 40,42)
L70: again, CDL is discussed without mentioning who generates this data (until L84). The term needs to be introduced here (or before this point).
We have updated the description here as well to remove any confusion over which product we are referring to. (L 71)
L74: Add a comma after “From this collocation” and all leading prepositional phrases as correctly done on L70, for example.
We have updated this line and several others like it that use leading prepositional phrases by adding commas where appropriate. (Multiple lines)
L93: “A kernel driven approach” only means something if you read the cited paper. It should not be this way. This paper should be intelligible without consulting another paper. I am fine with dropping “kernel driven approach” since it essentially adds nothing. Is the kernel part of Eq. 2 of the manuscript? If so, “(see equation below)” could be added after “kernel driven approach”.
We have removed this phrase as suggested. (L 95)
L97: add ‘(diffuse)’ after ‘indirect’ so that the SR subscript makes sense to the reader.
We have added “(diffuse)” as recommended (L 99)
L104: 5000 nm is part of the shortwave? That seems to be too far into the mid-IR, but I trust that this is not a typo.
This is the shortwave spectral range in use by some science teams. For example, the NASA’s CERES instruments that on board various satellites measure SW radiance at the sepctral range of 0.3-5µm (https://ceres.larc.nasa.gov/instruments/)
L116: I think this sentence is misleading. I understand that the CDL does not provide reflectances, just crop type or land use, but the sentence implies otherwise. The second half of this sentence “to derive…” belongs later in the paragraph (i.e. next sentence or two).
We have updated the sentence to read “To calculate the impact of crop type on surface albedo, precise information about the land cover homogeneity within each MODIS pixel is required.” to avoid any confusion. (L 113-114) as well as “derive the reflectance of the shortwave bands to for CDL pixel classification” (L 119) to help improve the readability.
L135: profiles -> profile
Changed as recommended (L 136)
L135: I am quite sure this is not a grammatically correct sentence. The last part of the run-on sentence starting with “thus” needs to be reworded.
We have condensed the sentence in order to avoid a run-on sentence. (L 135-136)
L137: What is the pointing uncertainty of MODIS? If it is more than 30 metres, I wonder if this should be mentioned as a source of error. I think the pointing uncertainty should be mentioned either way in Sect. 2.4, especially if there is a systematic (non-random) component to pointing error.
The pointing uncertainty for MODIS is around 50m. We have mentioned that in the text in section 2.4
“Note that as a potential source of error in this aspect, the MODIS satellite has a pointing uncertainty of around 50 m [16], which may lead to unforeseen issues when attempting to collocate over specific fields.” (L 152-155)
as suggested with the following reference.
16 Wolfe, RE, Nishihama, M, Fleig, AJ, Kuyper, JA, Roy, DP, Storey, JC, Patt, FS (2002). Achieving sub-pixel geolocation accuracy in support of MODIS land science. REMOTE SENSING OF ENVIRONMENT, 83(2-Jan), 31-49
L163: Add “were” before calculated.
Added as requested (L 168)
L172: is the comma between “crop” and “land” supposed to be there?
We have cleared up this sentence (L 177)
L185: Delete “of which”
We have removed this (L 190)
L188: “non-field crop” -> “non-field-crop”
Typo has been corrected (L 192)
L189: for -> to
Changed as requested (L 194)
L190: “at” -> “for” or “in”
Changed as requested (L 195)
L198: Remove “.” After “Figure 5”
Fixed typo (L 203)
L199: seven -> three
We have adjusted this sentence to correctly identify which channels are being discussed (L 204)
L200: Trim wavelength list.
List has been trimmed to only include wavelengths being discussed (L 205)
L212: “due to …” is a conclusion that is not supported in this paper by direct evidence, but is almost certainly correct.
We changed “due to” to “possibly due to” (L 217)
L215: following the peak growth period, the albedo of maize crops is noticeably higher (0.15-0.2) than at the start of the growing season (0.1). I wonder if there are other factors (rainfall, soil moisture) besides the factors mentioned by the authors.
We are unsure of the sum of factors and their relative impacts to the surface albedo post-harvest. We believe this deserves a follow up paper.
L224: The sentence “Still …” needs to be reworded. Perhaps delete “, was found”.
We have reworded this sentence (L 229-231)
L226: Another possibility is that cotton is more white (i.e. brighter) when viewed from above. This could explain the higher albedo than the 3 other main crops and the lack of seasonal variation. The foliage of the cotton plant is hidden to some extent by the white cotton.
Great suggestion. We have added “Additionally, the physiology of the cotton plant, which includes a unique vegetative structure and white cotton, may lend to this effect.” to the discussion. (L 235-236)
L238: I believe there are other substances in plants that absorb in the blue portion of the spectrum, in addition to chlorophyll.
We have added a mention of other absorptive substances in the text as requested:
“with relatively less absorption of green by chlorophyll a and b and other absorptive pigments”
and included a reference. (L 244-245)
L239: There is a space between ‘nm’ and the numeric value. This is how it should be. The rest of the paper butts the units to the numeric value.
We have updated the unit label as necessary (Numerous lines)
L265: I suggest that “Whereas” is deleted from this sentence. I think this word is misused here.
Sentence has been modified to replace “whereas” with “however” (L 272)
L324: Define NASS acronym.
The definition has been added to the initial introduction of NASS in the paper (L 71)
L329: Reword “Areas that appear noisy”. Is the “noise” in the albedo time series?
We have reworded to clarify that the noise is spatial in nature, probably due to classification difficulties. (L 337-338)
L336: “Figure 22. (…)”. There is no Figure 22 and a period should not follow ‘22’. The sentence also does not make sense.
“Figure 22” was added erroneously, it has been removed (L 345)
L337: This following sentence comes out of nowhere. I believe a chunk of text is missing. The authors need to ease the reader into this discussion of cirrus, which currently seems disconnected from the paper. Also, there are a couple of sentences starting with “Still,” and I am not sure what is meant by “still”. “Nevertheless” seems more appropriate.
A small section of text was erroneously excluded, this has been restored. (L 345-347)
L369: index -> index,
Corrected typo (L 373)
L370: The authors should tell the readers for which wavelengths previous researchers have tried to exploit the NDVI-albedo relationship. I doubt anyone would expect this relationship to hold from 470 to 1600 nm.
For the mentioned study, the NDVI is estimated using observed radiances at 0.65 and 0.86 µm. We added “particularly seen at the 650 nm and 860 nm wavelengths” the discussions. (L 382-383)
L387: Here, the reader should be told how this new database is different than an albedo database for homogeneous, non-barren conditions.
Per your suggestion we have added the following:
“These results can be further improved through implementation of the new database, which, in contrast to static or generic albedo databases, implements daily albedo changes for specific crops, incorporating regional management and growing patterns over homogenous croplands.” (L 400-403)
L420: “but may be due to…” What follows this excerpt seems like an explanation only for the harvesting season. It does not seem to explain the behaviour in the peak growing season. Thus, this sentence should be rewritten to clear this up.
We have removed this section of the sentence to remove uncertain details. (L 434)
L429: ; -> ,
We have split this into two sentences to avoid a run-on sentence. (L 442)
L439: “to economic modeling of crop acreage response functions”. I have no idea what this term means and I doubt whether many in the remote sensing community will also. To which response are the authors referring?
We changed the term to “to modeling of crop acreage changes driven by economic forces” and added a reference to an example of the described modeling effort. (L 452-454)