Forest land under different scenarios of future global change

Forest loss is one of the most threats to biodiversity, forested lands drive a key role in the climate earth system that also affects species diversity and ecosystem services (Vale et al. 2021). Therefore, several initiatives had been driven to map land-use time series, mainly projections into the future (Chen et al. 2020 and reference therein). The availability of those products is valuable for earth science, ecology, conservation, and other research fields once land-use land-cover data are important predictors of species occurrence and biodiversity threat (Ruiz-Benito et al. 2020., Valet et al. 2021) 

A recent land-use product called GCAM-Demeter presents the highest global spatial resolution (0.05 º) until now (Chen et al. 2020).  It provided current and future projections (2015-2100) under different scenarios of climate change (Shared Socioeconomic Pathways (SSPs) and Representative Concentration Pathways (RCP) ) and according to the most recent framework of Coupled Model Intercomparison Project phase 6 (CMIP6). The data in each year include grid-explicit fraction (in percent) of each of the 32 plant functional types (PFTs) that are widely used in current Earth system models. The complete dataset is available in five General Circulation Models (GCMs): gfdl, hadgem, ipsl, miroc, and noresm. Also includes the mean and standard deviation of those GCMs (Chen et al. 2020).

Although the valuable contribution of the GCAM-Demeter to provide those data,  it is compressed in NetCDF format, a complex file format that needs management to become usable in several analyses, especially in ecology and biodiversity analyses (Vale et al., 2021 and reference therein). Here I managed the outputs of the mean of five GCMs (harmonized projection)  based on  the sum analysis of plant functional types considering:

1- Global Extent at 0.05-degree resolution 

2- Years 2020, 2030 and 2050

3- SSPs and RCP as follow: SSP1_RCP2, SSP2_RCP45, SSP4_RCP6, SSP5_RCP85

The goals are to assess quantitatively the forest lands under different scenarios of global change and make these data available in the Tag Image File Format (TIFF)  which is a more friendly and useable format to incorporate in several spatial analyses, mainly in ecology and biodiversity studies for conservation purposes.

Methods

I downloaded the GCAM-Demeter NetCDF files  (the outputs of the mean of five GCMs and the first version, i.e the harmonized projection)  freely available at DataHub (Chean et al. 2020).  I selected, extracted, and performed the sum analysis of  plant functional types (codes PTF1 to PTF11- described in README attached) considering:

1- Global Extent at 0.05-degree resolution 

2- Years 2020, 2030 and 2050

3- SSPs and RCP as follow: SSP1_RCP2, SSP2_RCP45, SSP4_RCP6, SSP5_RCP85

The data manipulation and analysis were done using ncdf4 and raster packages in the R environment (R Core Team 2020, Pierce 2019; Hijmans et al. 2020).  The outputs range from 0 to 100 and can be identified by their file name, for example: 2020_SSP5_RCP85_Forest_GCAM-Demeter_GCMsMean_Harmonized.tif . Also, outputs are provided in the Tag Image File Format (TIFF) which is a more friendly and useable format (Vale et al. 2021 and references therein). The methods and the quantitative results for forested areas are detailed better in the GitHub repository .

Acknowledgments.

This initiative was possible due to the high-quality data maintained and made publicly available by GCAM-Demeter authors.  Also,  the study  was developed within the scope of the Earth System Modeling Program funded by CAPES (Coordination for the Improvement  of  Higher  Education  Personnel  -  Grant  No. 88887.373031/2019-00) 

References

Chen, M., Vernon, C. R., Graham, N. T., Hejazi, M., Huang, M., Cheng, Y., & Calvin, K. (2020). Global land use for 2015–2100 at 0.05 resolution under diverse socioeconomic and climate scenarios. Scientific Data, 7(1), 1-11. 

Hijmans, R. J. (2020). raster: Geographic Data Analysis and Modeling (R package version 3.3-13)[Computer software]. Retrieved form https://CRAN. R-project. org/package= raster.

Pierce, D. (2019). ncdf4: Interface to Unidata netCDF (Version 4 or earlier) Format Data Files. R package version 1.16.

Ruiz-Benito P, Vacchiano G, Lines ER, Reyer CP, Ratcliffe S, Morin X, Hartig F, Mäkelä A, Yousefpour R, Chaves JE, Palacios-Orueta A. Available and missing data to model impact of climate change on European forests. Ecological Modelling. 2020 Jan 15;416:108870.

Team, R. C. (2020). R: A language and environment for statistical computing.

Vale, M. M., Lima-Ribeiro, M. S., & Rocha, T. C. (2021). GLOBAL LAND-SE AND LAND-COVER DATA: HISTORICAL, CURRENT AND FUTURE SCENARIOS. Biodiversity Informatics, 16, 2021, pp. 28-38.