COS 168-4 - Global analysis of plant nutritional traits and effects of climate change on nutritional values update abstract

Background/Question/Methods

Plant nutritional properties (crude proteins, fibers, minerals, and dry matter) are key traits determining the quality and quantity of food intake of wild and domestic animals. These properties vary across plant functional types and are affected by environmental conditions. Present in future climatic changes will likely affect the nutritional quality of plants. However, most studies have focused on yield rather than quality and the consequences of climate change for feed quality are not well understood. We first investigate the factors controlling plant nutritional properties by compiling a large dataset containing 1121 plant species and 4085 observations of leaf properties. We called this database PNuts (Global database of Plant Nutritional properties). We then use a machine learning approach (Random Forest) to investigate how plant nutritional values will be affected under different climate scenarios until 2050.

Results/Conclusions

We show that to carbon dioxide concentrations, solar radiation, plant type, actual evapotranspiration, and precipitation are the driving factors of variability in nutritional properties. Future projections under +2 and +4 climate scenarios show a global decline of plant proteins (3%), minerals (9%-14%) and dry matter (18-22%), and an increase of fibers (7%). These results indicate that plants will become less digestible and nutritious within 2050 with important consequences for wild and domesticated animals’ health, physiology, and behavior.