TABLE 1.

Uncertainties and gaps in our understanding of past, present and future plant species distributions in the Andes and priorities for research

Topic	Subtopic	Uncertainties and gaps	Priorities
Observations	Species data	Number and list of native and non‐native species	Compile a plant species list for the whole Andes and per biome, including native and non‐native species
			Increase taxonomic treatments for Andean plant taxa
		Difference between under‐sampled and narrowly distributed species, spread of non‐native species	Increase species collections (native and non‐native) with high‐quality geographical and location data, beyond easily accessible areas
			Increase availability of existent specimen/occurrence data in public platforms Keep collecting to enable monitoring the spread of non‐native species and changes in native distributions
	Climate data	Observed trends and patterns of climate variability in specific regions and locations	Increase the collection of climate data at high frequency across the complete elevational gradient, significantly above the upper forest line
			Increase availability of existent climatic data, promoting a collaborative data‐sharing culture
			Increase the understanding of natural climate processes, including soil–vegetation–atmosphere interactions
		Spatial variation of temperature patterns at micro‐scales	Consider microclimatic variations using air and soil temperature sensors at finer spatial scales
Models	Climate models	The inability of models to represent clouds and convection	Develop new approaches to reduce errors directly related to shortcomings in process parametrisations
			Increase modelling resolution and complexity
		Poor land‐surface representation, including land surface–atmosphere interactions and feedbacks	Increase computational resources and technologies for archiving and sharing datasets
			Develop novel approaches to regional downscaling
	Plant distribution models	Representation of biological and ecological processes	Collect dispersal data and develop approaches to incorporate dispersal in SDMs and DVMs
			Collect demographic data (mortality, germination and establishment success) to improve parametrisation in DVMs and incorporate these data into SDMs
		Representation of external processes	Develop integrated models of land use change and plant distribution
			Include spatially explicit simulations of disturbance regimes (e.g. fire, building of infrastructure and roads)
		Model validation	Instal and monitor climate change experiments in field conditions Incorporate palaeo data in predictive models which could account for non‐analogous climate
		Representation of intraspecific variation	Collect data on functional traits on understudied areas, for natives and non‐native species, recording intraspecific variation (trait variation at population level), accounting for differences at local scales (phenotypic plasticity at elevational and latitudinal gradients, local adaptation)
		Spatial representation of DVM	Make DVMs spatially explicit, expanding their spatial scale without losing detail at local scales