the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Global mapping of oil palm planting year from 1990 to 2021
Abstract. Oil palm is a controversial crop, primarily because it is associated with negative environmental impacts such as tropical deforestation. Mapping the crop and its characteristics, such as age, is crucial for informing public and policy discussions regarding these impacts. Oil palm has received substantial mapping efforts, but up-to-date accurate oil palm maps for both extent and age are essential for monitoring impacts and informing concomitant debate. Here, we present a 10-meter resolution global map of industrial and smallholder oil palm, developed using Sentinel-1 data for the years 2016–2021 and a deep learning model based on convolutional neural networks. In addition, we used Landsat-5, -7, and -8 to estimate the planting year from 1990 to 2021 at a 30-meter spatial resolution. The planting year indicates the year of establishment for an oil palm plantation as of 2021, either newly planted or replanted oil palm in an existing plantation. We validated the oil palm extent layer using 17,812 randomly distributed reference points. The accuracy of the planting year layer was assessed using field data collected from 5,831 industrial parcels and 1,012 smallholder plantations distributed throughout the oil palm growing area. We found oil palm plantations covering a total mapped area of 23.98 Mha, and our area estimates are 16.66 ± 0.25 Mha of industrial and 7.59 ± 0.29 Mha of smallholder oil palm worldwide. The producers’ and users’ accuracy is 91.9 ± 3.4 % and 91.8 ± 1.0 % for industrial plantations, and 72.7 ± 1.3 % and 75.7 ± 2.5 % for smallholders, which improves upon a previous global oil palm dataset, particularly in terms of omission of oil palm. The overall mean error between estimated planting year and field data was -0.24 years and the root-mean-square error was 2.65 years, but the agreement was lower for smallholders. Mapping the extent and planting year of smallholder plantations remains challenging, particularly for wild and sparsely planted oil palm, and future mapping efforts should focus on these specific types of plantations. The average oil palm plantation age was 14.1 years, and the area of oil palm over 20 years was 6.28 Mha. Given that oil palm plantations are typically replanted after 25 years, our findings indicate that this area will require replanting within the coming decade, starting from 2021. Our dataset provides valuable input for optimal land use planning to meet the growing global demand for vegetable oils. The global oil palm extent layer for the year 2021 and the planting year layer from 1990 to 2021 can be found at https://doi.org/10.5281/zenodo.11034131 (Descals, 2024).
- Preprint
(2395 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
CC1: 'Comment on essd-2024-157', Denis murphy, 19 May 2024
A very useful study but the 'headline' figure of OP global OP plantation area, which I assume is dated as of 2021 is given as 23.98 Mha.
However, just a few lines later an FAO estimate from 2022 is quoted as a much higher area of 30 Mha - this discrepancy is not discussed anywhere in the paper. A figure of 30 Mha is also reported by FAO in 2023 and is widely used by other sources such as https://ourworldindata.org/palm-oil
Moreover, the 2024 value quoted by USDA (which might just be using the FAO data) is also 30 Mha.
In contrast, the much-quoted 2022 value of 23 Mha by Statista is much closer to the current authors' value
It would be nice to have some discussion about these discrepancies, which are from reputable bodies like FAO, USDA etc, because these headline values are widely quoted and used by policymakers, NGOs, the media etc.
Citation: https://doi.org/10.5194/essd-2024-157-CC1 -
AC1: 'Reply on CC1', Adrià Descals, 27 Aug 2024
Dear Denis Murphy,
Thank you for your comment. The revised manuscript includes a comparison with FAO and USDA statistics and discusses the potential reason for the discrepancy in the total oil palm area.
Please see a detailed response in the attached file.
Best regards, A. Descals on behalf of all the Authors
-
AC1: 'Reply on CC1', Adrià Descals, 27 Aug 2024
-
RC1: 'Comment on essd-2024-157', Anonymous Referee #1, 13 Jun 2024
Oil palm is an important oilseed that has received widespread attention because of its involvement in issues such as tropical deforestation. In this paper, based on their 2021 global oil palm mapping, the algorithm is optimized to achieve a higher accuracy of the global 10m oil palm mapping. Meanwhile, they further track the planting year of oil palms from 1990 to 2021 and demonstrated to achieve the fulfilling performance. In summary, this manuscript is well-written and presents a high-quality global oil palm dataset, which meets the ESSD standard for high data quality. Below are some comments to this manuscript:
The methodology should be strengthen appropriately, some parts are difficult to follow. For example:
- In section 2.2.1, how to correct the effect of local incident angle in Sentinel-1 imagery? The annual composites come from the average of the ascending and descending scenes, why not just ignore ascending and descending orbits in the median compositing?
- In section 2.2.2, I suggest add some descriptions about the “closed-canopy industrial oil palm” and “closed-canopy smallholder oil palm”, what are the significant differences between these two oil palms?
- In section 2.2.5, authors proposed to use the time-series NDWI to identify the planting years, and gave an example. However, I have two concerns as: 1) the reasonableness of your choice of NDWI, it must be explained. Does it also apply in case of conversion of cropland to oil palm? 2) You've labeled vegetation, clouds, and bare land with random forest classification over each Landsat observations, but how do you account for the effects of classification error? If an area has gone through "deforestation - bare land - grassland/cropland - oil palm", how can the year of planting be determined?
In section 2.2.4, the rigor of the validation sample is very clearly stated, but the number of validation samples for oil palm is too sparse (only 973 points). Thus, I suggest increasing the number of oil palm sample points.
I suggest that the authors give a map of the global spatial distribution of oil palm in 2021 in Section 3.1.
The new developed oil palm has been validated to achieve better performance, thus, I suggest that the authors can give some regional, visual comparison results.
If possible, I would also suggest a global map of the age distribution of oil palms is presented in the manuscript, it would be very interesting!
I agree with the comment in the Community Comment (CC1) as: the global oil palm area discrepancy between your dataset of 23.98 Mha and the FAO estimate about 30 Mha should be explained in the manuscript
Citation: https://doi.org/10.5194/essd-2024-157-RC1 -
AC2: 'Reply on RC1', Adrià Descals, 27 Aug 2024
Dear Reviewer #1,
We appreciate your time and effort for revising the manuscript. Your suggestions truly improved the manuscript. Please see our detailed response and updates to the manuscript in the attached file.
Best regards, A. Descals on behalf of all the Authors
-
RC2: 'Comment on essd-2024-157', Anonymous Referee #2, 01 Aug 2024
Oil palm has received substantial mapping efforts, but up-to-date and accurate maps detailing both the extent and age of oil palm plantations are essential for monitoring impacts and informing concurrent debates. This study achieved global 10-meter resolution mapping of oil palm and confirmed plantation years from 1990 to 2021. Overall, the manuscript is well-written, but there are still areas that could be improved:
- In line 22, "We found oil palm plantations covering a total mapped area of 23.98 Mha," while in line 38, "30 Mha in 2022" is mentioned. It is recommended to explain the differences between these two data points in the subsequent discussions.
- It is recommended to add descriptions for "closed-canopy industrial oil palm" and "closed-canopy smallholder oil palm". Additionally, using comparative charts to illustrate the differences between these land cover categories both qualitatively and quantitatively is advised, enabling readers to better understand the distinctions between these categories."
- In line 92, it is suggested to briefly explain the data organization format, whether temporal information was used, whether monthly synthesis was employed, and if data stacking akin to multi-channel image data was implemented when inputting into DeepLabv3+.
- Due to regional distribution differences, it is recommended to increase the number of validation samples to better demonstrate the effectiveness of the method.
- In line 119, "manually digitized false positives, and subsequently reclassified these commission errors as class ‘other’", was it first masking and then checking all categories, followed by manually correcting misclassified information?
- In the figure on line 179, the author uses NDWI to determine the planting year and provides an example. It is recommended to explain the rationale for using NDWI.
- In line 228, "only represent the regions where oil palm was found worldwide in this study", did the study detect globally and then filter out these current regions, or was it based on previous research?
- In Section 3.1, it is suggested to first present the global oil palm distribution map obtained in this study.
- In Section 3.3, consider increasing the number of sample points to assess validity further.
It is recommended to add comparative results with previous studies in Section 3.3 to visually demonstrate the effectiveness of the current research.
Citation: https://doi.org/10.5194/essd-2024-157-RC2 -
AC3: 'Reply on RC2', Adrià Descals, 27 Aug 2024
Dear Reviewer #2:
We thank the reviewer for the thoughtful observations. We have now addressed your comments. By doing so, the article has substantially improved. Please see our detailed response in the attached file.
Best regards, A. Descals on behalf of all the Authors
Data sets
Global oil palm extent and planting year from 1990 to 2021 Adrià Descals https://doi.org/10.5281/zenodo.11034131
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
1,952 | 481 | 62 | 2,495 | 23 | 32 |
- HTML: 1,952
- PDF: 481
- XML: 62
- Total: 2,495
- BibTeX: 23
- EndNote: 32
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1