Science @ Earthshot

Our world class science team applies ecology and machine learning to global ecosystem restoration.

Ecological Simulation

Our science team works to advance the state of the art in forest growth modeling, by understanding the effects of different restoration strategies and risk factors that could affect their outcomes.
We fuse data, machine learning, and process-based modeling techniques to provide detailed, accurate ecological insights on land restoration outcomes.
modeling techniques
Earthshot is combining empirical models with land surface process models.

We leverage scientific literature to inform state-of-the-art modeling techniques using both on the ground and remotely sensed data. These predictions sit within a larger framework that considers a variety of biological and non-biological risks to forest-based climate solutions, such as fire and the effect of climate change on tree growth.
How much carbon can trees and soils store?

What happens to water resources when you grow a forest?

How does biodiversity respond to ecological regeneration?

What risks does climate change pose to the natural world?

How are natural regeneration efforts susceptible to drought and fire?

Current Projects

Predictive model of biomass in unaided natural regeneration case (UNR), globally applicable, trained on data from Cook-Patton paper.
Stephen Klosterman
Principal Scientist
Deforestation risk analyses and avoided CO2 emissions projections for REDD+ projects.
Pedro Ribeiro Piffer
Remote Sensing Scientist
Natalia Gonzalez Pestana
GIS Specialist
Xingu AGB
Xingu downscaled AGB
Annual time series of biomass downscaled to 30m resolution.
Joe Hughes
Planetary Ecologist