XO FILES SCIENCE eXtraordinary Opportunity PHILANTHROPY ALLIANCE

Predicting the Future of Earth’s Forests Forests absorb carbon dioxide and are thus an important bufer against climate change, for now. Understanding forest dynamics would enable both better management of forests and the ability to assess how they are changing.

Stuart J. Davies Frank H. Levinson Chair in Global Forest Science; Director, Forest Global Earth Observatory-Center for Tropical Forest Science, Smithsonian Tropical Research Institute

Forests cover much of the land surface of Earth, outside of deserts and polar regions. But forests are changing rapidly in ways we don’t fully understand. The gold dots are current observing sites for the Forest GEO research efort. Source: ForestGEO.

Trees are the most conspicuous form of can live many centuries. More than half Currently, forests absorb nearly a terrestrial plant life. They have helped to of all known terrestrial plant and animal quarter of the carbon dioxide emitted sustain humans as well as many mam- species also live in forests, so they are a each year by human activities—and thus mals and birds, and they grow nearly storehouse of biodiversity; they also pro- help to mitigate the impact of climate everywhere. Yet we tend to take them vide medicines, food, and fuel for a huge change. But whether that will continue is for granted. Collectively, trees are the fraction of humanity. Forests also play uncertain, because forests are changing forests—perhaps 60,000 diferent species major roles in climate and hydrological rapidly through the combined efects of ranging from fast-growing, short-lived cycles, and they store half of the world’s logging and over-exploitation, pollution, varieties to the towering redwoods that terrestrial carbon. disease, invasive species, and a warming XO FILE | Predicting the Future of Earth’s Forests

climate. Predictive models—of forest factors. The core data are standardized and have removed both ground-dwelling and cover and of carbon uptake—have been shared, enabling comparative analyses. In tree-dwelling animals that disperse seeds, virtually impossible, because we are only addition, the network builds local scientifc thus reducing the ability of the forest beginning to understand the complexity capacity by training hundreds of research- to regenerate trees. In another part of of forest ecosystems, to fully map their ers. Our goal is to expand to 100 observing Malaysia, an explosion in the population diversity, and to unravel how forests sites, enough to cover all major forests and of wild pigs (which Muslim populations actually work. The key to improving our forest types, and to transform the network do not eat and so don’t hunt) destroyed understanding and our predictive ability into a permanent global observation sys- young trees in neighboring forests to make is to develop an integrated program of sys- tem for the world’s forests. their nests. Since tropical forests absorb the tematic data collection and modeling. These eforts have already generated largest amount of carbon dioxide, these Past attempts to model and predict new insights into the diferent dynamics pressure have important implications for the future of the world’s forests have been and stresses among forest types: the ability of forests to provide a bufer severely limited by a lack of fundamen- Perhaps surprisingly, trees grow quite against climate change. tal data on how forests work. The most well on the permafrost in arctic regions. Nutrients play an important role, but widespread and simplistic forest ecosys- With rapidly rising temperatures from their availability varies enormously. Soils tem models, so-called “big leaf” models, climate change, these boreal forests in the central Amazon region are so poor characterize the entire diversity of species are seeing longer growing seasons and that nutrients are recycled in the living in a forest with one set of physiological expansion of their potential range, as biomass. When those forests are cut attributes. More advanced models may treelines move north. But as temperatures down to grow crops, the crops typically include two distinct functional categories rise, the permafrost thaws, causing ground fail within a few years. In the western (e.g., evergreen and deciduous species), but surface subsidence and waterlogging of Amazon region, in contrast, soils are rich in even these fall far short of representing the the soil, efectively drowning trees. The nutrients and supported very productive actual diversity and complexity of forests. result is net forest loss. farming in pre-Columbian times. Detailed The predicted outcomes for the future of Temperate forests worldwide (especially understanding of such patterns could forests and climate derived from these in the U.S.) are expanding rapidly as they provide much more efective policies to current, limited models remain hugely recover from heavy logging in earlier guide development. uncertain. Real progress requires models centuries and get a boost from higher Forests vary widely from one part of that incorporate vastly greater realism on atmospheric carbon dioxide levels, the world to another. The tropical forests the processes driving forests. The only way longer growing seasons, and increasing of Borneo and the Amazon are each to do this is through an integrated program soil nutrients. On the downside, these home to thousands of species, but have of data collection and model development. conditions also favor forest pests and not a single species in common. Both the Over the past 30 years, systematic diseases. In Europe, for example, a fungal famous redwoods and the lesser-known data collection has been the task of the disease has killed many European ash but extremely long-lived bristlecone pines Forest Global Earth Observatory (Forest- trees and is now spreading to England. In (some as old as 3000 years) are found only GEO), a worldwide network of scientists eastern North America, hemlock trees are in the western U.S. led by the Smithsonian Institution that dying from a sap-sucking, very temperature Forests tend to be very clumpy. now includes 100 partner organizations sensitive insect that originated in Japan Whether from specifc habitat in 25 tropical and temperate countries. and has been moving north over the past requirements or the dynamics of seed ForestGEO conducts detailed and regular few decades. The mountain pine beetle is dispersal, species are far from uniformly measurements in 64 protected forest plots. killing high altitude white-bark pine trees distributed within even a small forest. The measurements track the birth, growth, that previously were protected by cold Moreover, forests are diverse in functions and death of over six million individual temperatures. as well, because diferent types of trees trees spanning 12,000 species, as well as Tropical forests in many areas globally do diferent things; some thrive in valleys, changes in their ecological context: soil are losing ground from logging, land some on hills, some only in shady sites composition, co-habiting microbial and clearing, and hunting practices. In Sarawak, under taller trees. Variations in soil animal species, nutrient fows and other Malaysia, excessive hunting pressures nutrients and water availability help drive XO FILE | Predicting the Future of Earth’s Forests

these patterns. This makes sustainable every fve years, but also training more growing space-based monitoring capabil- forest management—and modelling scientists and supporting interdisciplin- ity. This requires partnerships with relevant forest dynamics—much more complex. ary teams to analyze, understand, model, space agencies and the interdisciplinary Predictive models capable of handling and interpret these data. By developing capacity to inter-relate biological and such geographical variability and ecological integrated collaborations among these ecological data with the spectral signatures complexity will need more detailed data researchers, modelers can help guide which from many diferent wavelengths observed on nutrient cycling, the functional biology data are necessary for model improvement, from above. Accomplishing this would of diferent tree species, the actual carbon and feld researchers can ensure that the allow for “real-time” evidence of forest fuxes into and out of soils and trees, and on critical processes driving forest dynamics change and more global coverage, provide the patterns of interactions between local are not left out of the models. In efect, it important benchmarks for testing forest animal species and trees (whether eating requires the creation of a new, quantitative models, and might ofer governments more their leaves or dispersing their seeds). This science of forest dynamics. It also requires a efective tools for protecting forests and in turn requires not just re-measuring the long- term commitment. planning their sustainable usage. growth of millions of individual trees and Equally important is the need to inte- collecting other data from their habitats grate ground-based forest data with the

The eXtraordinary Opportunity | How to Create a Global Forest Observation System

Forest dynamics play a complicated yet To analyze and poorly understood role in infuencing the interpret those data rate of climate change. That’s because forests requires building and are changing rapidly, and because the maintaining an expanded fraction of the additional man-made carbon global research team of emissions forcing climate change that interdisciplinary forest forests will absorb remains unknown. And scientists with a wide while climate issues are an important reason variety of expertise. That to understand forest dynamics, forests includes collaborations are also important as a store of biological with space agencies and diversity and for their impact on watersheds climate modelers. We and water supplies. expect this research to ofer Only by repeated, standardized new quantitative insights observations of individual trees and their into forest evolution, surroundings in multiple locations can we species interactions, large- gather the data that could enable a scientifc scale landscape change, understanding of forest dynamics and build physiological responses Researchers at a Forest GEO research site in Huai Kha Khaeng, predictive models of their interaction with of diferent species to Thailand, measuring tree growth, nutrient fows, and other the global climate. To ensure that we are changing conditions, energy ecological data on one of more than 5 million trees now under monitoring all forest types and covering and mass balances within regular surveillance by the program. Source: Christian Ziegler, STRI. all major forests geographically, we need forest ecosystems, and to expand the current observation sites carbon fuxes and storage. We estimate the on how forests can bufer or contribute to from 64 to 100. Creating and maintaining need for an additional $100 million over climate change. This is an extraordinary this expanded observation system, with ten years to train and sustain the required opportunity for private philanthropy, which repeated measurements of about 10 million scientifc talent. can best address the need for a long-term individual trees, requires an endowment of The result will be powerful, predictive commitment, a multinational research focus, $100 million. That will enable the necessary models of forest dynamics for both improved and an intellectual breadth that spans many global dataset on forests. management and quantitative insights diferent disciplines.

SCIENCE PHILANTHROPY ALLIANCE 480 S. California Ave. #304, Palo Alto, CA 94306 | 650-338-1200 | [email protected] | www.sciphil.org