the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
GMIE-100: a global maximum irrigation extent and irrigation type dataset derived through irrigation performance during drought stress and machine learning method
Abstract. Irrigation stands as the primary sector of human water consumption and plays a pivotal role in enhancing crop yields and mitigating drought effects. The precise distribution of irrigation is crucial for effective water resource management and the assessment of food security. However, the existing global irrigated cropland map is characterized by a coarse resolution, typically around 10 kilometers, and is often not regularly updated. In our study, we present a robust methodology that leverages irrigation performance during drought stress as an indicator of crop productivity and water consumption to identify global irrigated cropland. Within each irrigation mapping zone (IMZ), we identified the dry months occurring during the growing season from 2017 to 2019 or the driest months from 2010 to 2019. To delineate irrigated cropland, we utilized collected samples to calculate normalized difference vegetation index (NDVI) thresholds for the dry months of 2017 to 2019 and the NDVI deviation from the ten-year average for the driest month. By combining the results with the higher accuracy between these two methods, we generated the Global Maximum Irrigation Extent dataset at 100-meter resolution (GMIE-100), achieving an overall accuracy of 83.6 %. GMIE-100 reveals that the maximum extent of irrigated cropland encompasses 403.17 million hectares, accounting for 23.4 % of the global cropland. Concentrated in fertile plains and regions adjacent to major rivers, the largest irrigated cropland area is found in India, followed by China, the United States, and Pakistan, ranking 2nd to 4th, respectively. Importantly, the spatial resolution of GMIE-100, at 100 meters, surpasses that of the dominant irrigation map, offering more detailed information essential for supporting estimates of agricultural water use and regional food security assessments. Furthermore, with the help of the DL method, the global central pivot irrigation system (CPIS) was identified using Pivot-Net. We found that there are 11.5 million hectares of CPIS, accounting for about 2.9 % of total irrigated cropland. In Namibia, the US, Saudi Arabia, South Africa, Canada, and Zambia, the CPIS proportion was larger than 10 %. To our best knowledge, this is the first effort to identify irrigation methods globally. The GMIE-100 dataset containing both or irrigated extent and CPIS distribution is accessible on Harvard Dataverse at: https://doi.org/10.7910/DVN/HKBAQQ (Tian et al., 2023a).
- Preprint
(4719 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on essd-2023-536', Anonymous Referee #1, 21 Feb 2024
-
AC3: 'Reply on RC1', Fuyou Tian, 17 Apr 2024
We are highly appreciated for your constructive comments and suggestions on our manuscript. Those comments and suggestions are valuable and helpful for revising and improving our article, as well as inspiring our research. We have carefully reviewed the comments and have revised the manuscript accordingly.
1. The crop land definition from FAO was “Cropland is land used for the cultivation of crops, both temporary (annuals) and permanent (perennials), and may include areas periodically left fallow or used as temporary pasture.” Actually, we just focus temporary cropland because this was food producing crop type. The permanent crops were usually for fruit trees, nut trees, coffee, tea, and some types of vines, which is recognized as shrub or tree in most landcover system such as ESRI, FROM-GLC, GLAD-Map, GLC-FCS30 and WORDCOER. On the contrary, harvest crops, maize, soybean, wheat, and rice was most important feeding crops. So, we choose this definition to distinguish irrigated and rainfed cropland.
2. We are sure that the accuracy of this basically acceptable. Because this data integrated 10 existing land cover maps or cropland datasets to delimit the global cropland extent while masking out irrelevant non-cropland pixels for the period of 2016–2018 (Figure 1). More detailed information on these land cover and cropland layer products as well as their classes used in the integration refer to (Zhang et al., 2021). Although variations in classification systems among different products exist, a subset of classes of those land cover and cropland layer products were selected to best fit into the cropland definition. Spatially, FROM-GLC was selected for Europe, Africa, New Zealand, the majority of Asia, and part of Latin America. GFSAD30 was selected for tropical Asian islands, including Indonesia, Malaysia, and the Philippines. In addition to these two global-coverage cropland extent products, several national or regional datasets, including ChinaCover, CDL, AAFC ACI, NLCD, MapBiomass, CLUM, SERVIR, and INTA, were used because they have been extensively validated by local experts and hence exhibited high accuracies of cropland mapping. The data was at 30 meter resolution. You could view it online via http://desp.casearth.cn/data-preview/?id=GCL30_2020&lang=en or download it via https://data.casearth.cn/en/sdo/detail/62ff50e208415d271ab1b84a
3. Our responses are given in a point-by-point manner below and BLUE fonts. Please find our detailed responses in supplement to all these comments/suggestions and thank you again for everything you have contributed.
-
AC3: 'Reply on RC1', Fuyou Tian, 17 Apr 2024
-
RC2: 'Comment on essd-2023-536', Anonymous Referee #2, 22 Feb 2024
This study demonstrated a global irrigation dataset with 100 meters using irrigation performance during drought stress, which is a brand new way to detect irrigated and non-irrigated cropland. Furthermore, this MS finishes mapping the global central pivot irrigation system using the Deep Learning method. Also, it is interesting to detect special irrigation methods using deep learning methods. Overall, the MS was well-written and designed for readers. But there were still some concerns before this MS was accepted:
Major concerns:
- About the resolution: In section 2.1 some coarse data was described as input data, but the final resolution of the irrigation map is 100 meters, so this will mislead some readers on how to produce a 100-meter irrigation map using 500-meter data.
- About the IMZs: You mention that “65 MRUS in Copwatch served as the basis for further division of global cropland into 110 irrigation mapping zones (IMZs)”, what is the principle for further dividing these zones? Are these zones available or not?
- About accuracy assessment: you collect many field points using the GVG app. How to distinguish irrigation field points during the field survey?
- The irrigation map and GCPIS were identified using two ways (irrigation performance and DL), but some figures make me confused to display these two results.
Minor revision:
- The preprocess of NDVI data in Line 160 should be further explained.
- You could list some detailed maps of global CPIS in Figure 6 to make the global CPIS map clearer.
- In Figure 16 it will be significant if the satellite images were added to give the reader a basis for their judgment.
- IMZ was not so readable in Figure 1.
- The English should be further polished and improved.
Citation: https://doi.org/10.5194/essd-2023-536-RC2 -
AC1: 'Reply on RC2', Fuyou Tian, 17 Apr 2024
We are deeply grateful for the your positive assessment and constructive suggestions on our manuscript. Your insightful comments have played a pivotal role in enhancing the clarity and significance of our research. We have carefully considered each point in your comments and please kindly refer to our point-by-point responses in supplement.
-
RC3: 'Comment on essd-2023-536', Anonymous Referee #3, 26 Feb 2024
This manuscript introduced the GMIE-100 dataset, which identifies global irrigated cropland using drought stress performance and machine learning. This is a valuable dataset that could benefit various fields, including agriculture, environmental science, and water resource management. However, I have some major concerns about this MS that need the authors to clarify before it is further processed.
1 The title of the manuscript indicates that the dataset represents the largest irrigated area. How does the author interpret this "largest area"? This requires the author to provide explicit clarification within the text. Additionally, how does the author consider the possibility of overestimation of this largest area relative to the actual distribution, given that our focus is on the actual distribution range?
2 The samples are derived from different collection methods. It is crucial for the author to clarify whether samples collected through different methods exhibit consistent representation and describe irrigated land in the same manner. If their collection standards vary, the author needs to explicitly discuss the impact on the results.
3 In terms of accuracy assessment, merely providing overall accuracy is insufficient. Please refer to best practices for reporting accuracy as outlined in papers such as Olofsson et al. 2014 [1]. Moreover, I have not observed quantification of uncertainty, which necessitates further work from the author.
Olofsson P, Foody GM, Herold M, Stehman SV, Woodcock CE, Wulder MA. Good practices for estimating area and assessing accuracy of land change. Remote Sensing of Environment 2014; 148:42–57. https://doi.org/10.1016/j.rse.2014.02.015.
4 The results and discussion sections lack necessary citations. Many explanations proposed by the author lack corresponding literature support, which makes it difficult for me to be convinced of the correctness of your interpretations. Please see the annotations I've made in the manuscript.
5 I have made several annotations in the manuscript indicating areas that need revision. It is advised that the author make corresponding modifications and carefully review the entire document to rectify similar errors.
-
AC2: 'Reply on RC3', Fuyou Tian, 17 Apr 2024
We are highly appreciated for your constructive comments and suggestions on our manuscript. Those comments and suggestions are valuable and helpful for revising and improving our article, as well as inspiring our research. We have carefully reviewed the comments and have revised the manuscript accordingly. Our responses are given in a point-by-point manner below and BLUE fonts. Please find our detailed responses in supplement to all these comments/suggestions and thank you again for everything you have contributed.
-
AC2: 'Reply on RC3', Fuyou Tian, 17 Apr 2024
Data sets
GMIE: a global maximum irrigation extent and irrigation type dataset derived through irrigation performance during drought stress and machine learning method Fuyou Tian, Bingfang Wu, Hongwei Zeng, Miao Zhang, Weiwei Zhu, Nana Yan, and Yuming Lu https://doi.org/10.7910/DVN/HKBAQQ
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
715 | 163 | 54 | 932 | 60 | 51 |
- HTML: 715
- PDF: 163
- XML: 54
- Total: 932
- BibTeX: 60
- EndNote: 51
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1