the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 20 Sep 2024
CropSuite – A comprehensive open-source crop suitability model considering climate variability for climate impact assessment
Abstract. Increasing demand for agricultural land resources and changing climate conditions require for strategic land-use planning and the development of adaptation strategies. Therefore, information about the suitability of agricultural land is a necessary prerequisite. Current suitability approaches often focus on single crops, can only be applied regionally and usually neglect the impact of climate variability on crop suitability. Here, we introduce CropSuite, a new comprehensive and easy-to-use open-source crop suitability model that makes it possible to overcome these shortcomings. CropSuite uses a fuzzy logic approach and is based on the assumption of Liebig’s law of the minimum. It includes a spatial downscaling approach for climate data, which allows for performing crop suitability analysis at very high spatial resolution. Several factors that impact on crop suitability can flexibly be integrated into CropSuite by determining membership functions. CropSuite allows for the consideration of irrigated and rainfed agricultural systems, vernalization requirements for winter crops, lethal temperature thresholds, photoperiodic sensitivity and several other limitations. The model calculates and outputs climate-, soil-, and crop suitability, the optimal sowing date, the potential for multiple cropping, the (most) limiting factor(s), as well as the recurrence rate of potential crop failures.
In this study, we apply CropSuite for 48 crops at a spatial resolution of 30 arc seconds (1 km at the equator) for Africa. Thereby, we consider regionally important staple and cash crops, such as coffee, cassava, banana, oil palm, cocoa, cowpea, groundnuts, mango, millet, papaya, rubber, sesame, sorghum, sugar cane, tobacco, and yams. We find that the consideration of climate variability for calculating crop suitability makes a significant difference on suitable areas, but also affects optimal sowing dates, and multiple cropping potentials. The most vulnerable regions for climate variability are identified in Somalia, Kenya, Ethiopia, South Africa, and the Maghreb countries. The results provide valuable crop-specific information that can be further used for climate impact assessments, adaptation and land-use planning.
- Preprint
(2579 KB) - Metadata XML
-
Supplement
(334 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-2526', Anonymous Referee #1, 22 Oct 2024
In this manuscript, Zabel et al. describe a new piece of software, CropSuite, that generates maps of crop suitability and related information based on climate, soils, and terrain. This builds on previous work by themselves and other authors to include, importantly, (a) a consideration of climate variability in addition to averages and (b) less-widespread but regionally important crop types. Noting that such crops are under-studied but are especially important in Africa, the authors focus their analyses there. The results look reasonable when compared to real-world crop distributions and sowing dates.
One of the goals of CropSuite was to make something that is easy-to-use and flexible enough to be used by a variety of stakeholders, not just scientists. As a scientist, I can’t really assess how accessible it is to less-technical users, but the inclusion of a graphical user interface (GUI) is a really important development. I do think, however, that this tool will be useful to scientists and model developers as well. Global gridded crop models and especially integrated assessment models need to be able to endogenously represent things like sowing date, the potential for multiple cropping, and shifts in what crops are planted where; tools like CropSuite can help.
That said, I do have some questions and concerns about the manuscript as currently written. Thus, I recommend it be considered for publication after minor revisions. See attachment for details.
-
AC1: 'Reply on RC1', Florian Zabel, 05 Dec 2024
Dear Reviewer #1, thank you very much for your time to review our paper! We appreciate your thoughtful comments and suggestions a lot! These were really helpful and we were able to improve our study accordingly.
Since the initial submission of this paper, we were able to improve the CropSuite model and the GUI. We uploaded an updated version (v1.0) of CropSuite to Zenodo and GitHub. In addition, we uploaded the complete GeoTIFF dataset and the compiled maps for all 48 crops to Zenodo.
We also hope that CropSuite will be further used not only by stakeholders, but also be further developed by scientists and model developers, which is our main motivation to provide the source code of CropSuite as open source. We prominently added this goal to the end of the abstract, since this was hidden so far. We also agree that the development of the GUI is an important aspect that may have been somewhat neglected in our paper. To highlight the importance of the GUI, we added it also to the abstract.
Thank you very much that you recognize the potential of CropSuite to improve crop models and integrated assessment models, which is very motivating for us.
See Supplement for details on our point by point reply.
-
AC1: 'Reply on RC1', Florian Zabel, 05 Dec 2024
-
RC2: 'Comment on egusphere-2024-2526', Anonymous Referee #2, 31 Oct 2024
This study improves the crop suitability assessment method by considering the climate variability. It compares results with SPAM2020, GAEZv4, and the crop calendar from GGCMI. The manuscript is clearly written. I have a few comments.
Major:
1. Make the dataset and results more accessible. Now, all the results in Zenodo are *.png, which contain no geographical information. I highly recommend authors to provide geographic file format, e.g., Geotif and netcdf.
Minor:
1. line 77, Is the soil texture for >200cm also needed? if yes, where does it come from?
2. Why the soil layer in lines 79-80 is different with Table 2
3. Line 81, reference formate. And, why are the weights needed? How are these weights applied? It's a bit confusing here. Did the authors mean that weights were used to multiply with original value to generate the new value?
4. line 233. OK, MapSPAM2020 may introduce some uncertainties, then why not using MapSPAM2010?
5. Line 277, is it because that nutrient and soil fertility are not considered in this study?
6. In theory, I would expect a smaller area in this study because this study considers additional climate variability. However, Figure 8 shows a larger area by this study. Can the authors explain more about this?
Citation: https://doi.org/10.5194/egusphere-2024-2526-RC2 -
AC2: 'Reply on RC2', Florian Zabel, 05 Dec 2024
Dear Reviewer #2, thanks a lot for your comments and for reviewing our paper!
In the following, we refer to your comments and answer them directly below your comment. Please note that line numbers in our reply refer to the revised version of the manuscript with track changes. Thank you!
See attached supplement for our detailed point by point reply.
-
AC2: 'Reply on RC2', Florian Zabel, 05 Dec 2024
Data sets
Additional maps for all 48 crops Florian Zabel, Matthias Knüttel, and Benjamin Poschlod https://doi.org/10.5281/zenodo.13285542
Africa Agriculture Adaptation Atlas Matthias Knüttel and Florian Zabel https://adaptationatlas.cgiar.org
Model code and software
CropSuite (v0.9) Source Code Matthias Knüttel and Florian Zabel https://doi.org/10.5281/zenodo.13285636
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 280 | 106 | 15 | 401 | 23 | 8 | 6 |
- HTML: 280
- PDF: 106
- XML: 15
- Total: 401
- Supplement: 23
- BibTeX: 8
- EndNote: 6
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1