Global Wheat Cultivation Distribution under Future Climatic and Socio-economic Conditions (RCP-SSP combinations)

This is the outcome data from our study titled "Prediction of global wheat cultivation distribution under climate change and socioeconomic development" which was published in The Science of The Total Environment.  The present study represents a significant extension of our previous research on "The Potential Distribution and Dynamics of Global Wheat under Multiple Climate Change Scenarios".

Socioeconomic and climate change are both critical factors influencing the global distribution of crop cultivation. However, there has been limited exploration of the role of socioeconomic factors in predicting future crop cultivation distribution under climate change.

We have proposed the MaxEnt-SPAM approach under the assumption that environmental conditions are the primary determinants of land suitability for cultivating wheat, while socioeconomic factors play a crucial role in influencing farmers' crop choices. In essence, the distribution of wheat cultivation is contingent upon maximizing potential revenue and ensuring suitability for wheat planting.

The proposed MaxEnt-SPAM approach was utilized to estimate the distribution of wheat cultivation in three combined Representative Concentration Pathway (RCP) - Shared Socioeconomic Pathway (SSP) scenarios, namely RCP2.6-SSP1, RCP4.5-SSP2, and RCP8.5-SSP3. The methodology involved estimating wheat planting suitability under future RCP scenarios using the MaxEnt model, predicting farmers' crop choices under future SSP scenarios through Time series-Backpropagation (TS-BP) models, and ultimately estimating global wheat cultivation distribution based on the SPAM model. Validation of this approach against major known datasets on the distribution of wheat cultivation demonstrated satisfactory accuracy, with a predictive accuracy exceeding 85% and a significant positive correlation (p < 0.01) between the predicted global wheat cultivation and multiple known datasets.

Based on the aforementioned concept and methodology, a global wheat cultivation distribution grid (0.5 degree × 0.5 degree) was projected under the RCP2.6-SSP1, RCP4.5-SSP2, and RCP8.5-SSP3 scenarios.

The findings suggest that RCP8.5-SSP3 may offer the most favorable conditions for wheat cultivation. Additionally, socioeconomic development significantly constrains the potential distribution of wheat cultivation, with estimated areas accounting for an average of 77% of the potential distribution determined by climatic factors under the selected RCP-SSP scenarios. Socioeconomic development appears to have a positive impact on wheat cultivation in Africa.

Our results illustrate the influence of socioeconomic factors on crop distribution within a market economy framework, underscoring the importance of integrating socioeconomic factors and climate change for accurate predictions of crop cultivation distribution.

We contend that the global wheat cultivation distribution datasets under future climatic and socio-economic conditions (RCP-SSP combinations) are a valuable addition to existing products. This prediction data is among the few products to consider both climate change and socio-economic development, providing a more comprehensive understanding of crop cultivation distribution dynamics.

The Global Wheat Cultivation Distribution under Future Climatic and Socio-economic Conditions (RCP-SSP combinations) is expected to enhance our comprehension of the dynamics and distribution of global wheat cultivation under different climate change and socio-economic development paths in the future, potentially supporting research in earth system simulation and agricultural sciences.

The dataset for the Global Wheat Cultivation Distribution under Future Climatic and Socio-economic Conditions (RCP-SSP combinations) and the Maxent-SPAM approach code is stored in a zip package named SPAM_MaxEnt.zip, which contains two folders (code and data).

code:

This sub-folder provides the main program and example data for the MaxEnt-SPAM approach. Codes are written in Matlab language by Puying Zhang. There are also 'read me.txt' files under the code folder to provide the necessary information.

The exampleData contains

1. h_pri.tif: prior data

2. h_res.tif: global C3 crop cultivation proportion

Run the main programme: cross_entroy.m

data:

This sub-folder contains global wheat cultivation distribution stored in GeoTIFF file format.

1 Global distribution of the long-term wheat-cultivation area fraction:

This sub-folder contains the data for the global distribution of the long-term wheat-cultivation area fraction in RCP2.6-SSP1, RCP4.5-SSP2, and RCP8.5-SSP3 scenarios. The value of each data ranges from 0 to 1, indicating the long-term wheat-cultivation area fraction in each grid, and the higher the value, the more wheat cultivated.

r2s1f_sub.tif: the data for global distribution of the long-term wheat-cultivation area fraction in RCP2.6-SSP1 scenario

r4s2f_sub.tif: the data for global distribution of the long-term wheat-cultivation area fraction in RCP4.5-SSP2 scenario

r8s3f_sub.tif: the data for global distribution of the long-term wheat-cultivation area fraction in RCP8.5-SSP3 scenario

2 Spatial overlap between the long-term period of land suitability for wheat planting and wheat cultivation distribution:

This sub-folder contains the data for Spatial overlap between the long-term period of land suitability for wheat planting and wheat cultivation distribution in multi-scenarios. The value of each data contains three values:{1, 2, 3}, 1 wheat cultivation existed but was predicted to be unsuitable to plant wheat; 2 presented a reduction in the wheat cultivation area compared to the land's suitability; 3 presented the region that wheat cultivation existed and was predicted to be suitable to plant wheat.

com_suit_fra126.tif: the spatial overlap between the long-term period land suitability for wheat planting and wheat cultivation distribution in (a) RCP2.6-SSP1 scenario and RCP2.6

com_suit_fra245.tif: the spatial overlap between the long-term period land suitability for wheat cultivation and wheat cultivation distribution in (b) RCP4.5-SSP2 scenario and RCP4.5

com_suit_fra385.tif: the spatial overlap between the long-term period land suitability for wheat cultivation and wheat cultivation distribution in (c) RCP8.5-SSP3 scenario and RCP8.5

3 Differences in the proportion of long-term wheat cultivation:

This sub-folder contains the data for the difference in the proportion of long-term wheat cultivation under the RCP-SSP scenarios and the distribution of long-term wheat planting suitability under the same RCP scenarios. The value of each data ranges from -1 to 1, This data is obtained by using the wheat-cultivation area fraction minus planting suitability grid to grid. the negative value indicates that the proportion of wheat cultivation is lower than the wheat planting suitability, while this positive value indicates that the proportion of wheat cultivation is higher than the wheat planting suitability.

r2s1_f.tif: Difference in the proportion of long-term wheat cultivation under the RCP2.6-SSP1 scenario and the distribution of long-term wheat planting suitability under the RCP2.6 scenario

r4s2_f.tif: Differences between the proportion of long-term wheat cultivation in RCP4.5-SSP2 and the suitability of long-term wheat planting under the RCP4.5 scenario

r8s3_f.tif: Differences between the proportion of long-term wheat cultivation in RCP8.5-SSP3 and the suitability of long-term wheat planting under the RCP8.5 scenario 

References:

Yaojie Yue, Puying Zhang, Yanrui Shang. The Potential Distribution and Dynamic of Global Wheat under Multiple Climate Change Scenarios. Science of the Total Environment, 2019, 688: 1308-1318.

Xi Guo, Puying Zhang, Yaojie Yue. Prediction of global wheat cultivation distribution under climate change and socio-economic development. Science of the Total Environment, 2024, 919: 170481.

For more details on the MaxEnt (Maximum entropy) model, please refer to (Phillips et al., 2006; Elith et al., 2011). SPAM (spatial production allocation model) refers to (You et al., 2009; You et al., 2014).

Elith, J., Phillips, S.J., Hastie, T., Dudík, M., Chee, Y.E., Yates, C.J., 2011. A statistical explanation of maxent for ecologists. Divers Distrib 17 (1), 43-57. https://coi.org/10.1111/j.1472-4642.2010.00725.x.

Phillips, S.J., Anderson, R.P., Schapire, R.E., 2006. Maximum entropy modeling of species geographic distributions. Ecol Model 190 (3-4), 231-259. https://coi.org/10.1016/j.ecolmodel.2005.03.026.

You, L.Z., Wood, S., Wood-Sichra, U., 2009. Generating plausible crop distribution maps for sub-Saharan Africa using a spatially disaggregated data fusion and optimization approach. Agr Syst 99 (2-3), 126-140. https://coi.org/10.1016/j.agsy.2008.11.003.

You, L.Z., Wood, S., Wood-Sichra, U., Wu, W.B., 2014. Generating global crop distribution maps: from census to grid. Agr Syst 127, 53-60. https://coi.org/10.1016/j.agsy.2014.01.002