Home People Lab News Research Publications Contact Links Code and Data

 

Effects of deforestation and climate change on carbon & water cycling in Amazonia

Improving the predictive ability of terrestrial biosphere models

Mechanistic models of animal movement

Enhancing vegetation structure for terrestrial biosphere modeling using Lidar and Radar techniques

Monitoring tropical tree growth and mortality rates in relation to climage change

Past research projects

 

 

 

Effects of deforestation and climate change on carbon & water cycling in Amazonia

Andes-Amazon Project
Tom Powell, Naomi Levine
The purpose of the Andes-Amazon Project is to predict how land-cover along with changes in climate will affect the composition, structure, and functioning of the Amazonian ecosystem over the next century. Using a combination of terrestrial models and field measurements, researchers will discover the role of land-cover change and climate in driving savannization of the Amazon and determine forest sensitivity to associated changes in climate, carbon flux, and hydrology.

Is deforestation changing the hydrologic climate and vegetation dynamics of the Amazon?
Marcos Longo
The Amazon forest has experienced significant deforestation since the 1970s. Approximately 17% of the original forest has been converted to pasture for cattle grazing or croplands. In the past, deforestation has occurred near the availability of paved road access. Currently, the Brazilian government plans to increase the number of paved roads in the Amazon, which may increase the deforestation of yet undisturbed areas.
The impact of deforestation on rainfall in the Amazon is still unclear, but a massive deforestation is likely to reduce evapotranspiration, affecting the overall water cycle strength. This can lead to a decrease in precipitation and an increase in drought frequency. Deforestation, however, is not yet widespread, and the current deforestation scale may induce atmospheric circulations that can potentially reorganize patterns of precipitation rather than merely reducing it. In both cases, the remaining forest experiences a significant change in water and light availability, which could alter the state of the ecosystem.
To understand these biosphere-atmosphere interactions resulting from deforestation, we use a state-of–the-art fully-coupled biosphere-atmosphere model (ED2-BRAMS). By considering the complex interactions between the biome and the atmospheric circulations, and by utilizing land use change scenarios that realistically represent regions likely to be deforested, we aim to better understand dynamic changes in precipitation and vegetation in the Amazon.

What impact will projected changes in climate and atmospheric [CO2] have on the Amazon forest?
Tom Powell
Climate models predict that a reduction in precipitation over much of the Amazon basin will be one of the consequences of climate change.  Both field studies and models have shown massive dieback of the Amazon forest when exposed to such a prolonged and extreme reduction in precipitation.  We use the ED model to explore the mechanism by which tropical trees succumb to drought and how their resilience might be modulated by rising atmospheric CO2.  This research is supported by the NSF-Amazon PIRE project, which seeks to build international partnerships between emerging scientists in the U.S. and Brazil.

 

Improving the predictive ability of terrestrial biosphere models
Constrained Biosphere Model


A recent analysis of terrestrial carbon dynamics in the Northeastern US, has shown how the Ecosystem Demography model version 2 (ED2) can be jointly constrained against eddy-flux measurements and forest-inventory measurements to yield accurate short-term and longer-term ecosystem dynamics. The ED2 model was initialized with the observed canopy composition structure in the footprint of the Harvard Forest flux tower, and then fitted simultaneously to the 1995 and 1996 hourly, monthly and yearly CO2 and ET flux data, and to the observed rates of deciduous and coniferous tree basal area growth and mortality in these years. Prior to the optimization, the model significantly underestimated the seasonal cycle of Net Ecosystem Productivity and significantly over-estimated rates of tree growth and mortality. After fitting, the model accurately captured the observed CO2 fluxes, ET fluxes, and canopy growth and mortality dynamics over timescales spanning hours to decades.
We then evaluated the performance of the optimized ED2 model’s performance at a different site, Howland forest. As before, the model was initialized with the observed canopy composition in the tower footprint, but the model parameters were not re-optimized. Despite the markedly different forest composition between the Howland and Harvard Forest sites (conifer-dominated as opposed to mixed-hardwood), and no subsequent adjustment of model parameters, there was a substantial improvement in the models predictions of the 5-year CO2 flux record, and measured tree growth and mortality dynamics at Howland. The optimized values all fell within specified acceptable ranges for each parameter. Changes in parameters responsible for the improved goodness-of-fit include: an increased maximum photosynthetic rate of hardwoods, a marked increase in the rate of fine root turnover, and a decrease in the carbon allocation to fine roots in conifer species.

Mechanistic models of animal movement

Mechanistic Home Range Analysis
James Forester
The objective of this research is to use mechanistic home range models to develop a predictive, reductionist theory of animal home range patterns from an understanding of the ecological and evolutionary forces that govern the movement behavior of individuals. In contrast to traditional home range models where arbitary statistical distributions are used to characterize spatial patterns of animal relocations, mechanisitic home range models formally derive expected patterns of space-use from underlying correlated random walk descriptions of individual movement and interaction behavior.By developing a series of mechanistic home range models for carnivore populations, this research has shown how mechanistic home range model models can be used to:

coy huntmodellamar valley

Enhancing vegetation structure for terrestrial biosphere modeling using Lidar and Radar techniques

Lidar and Radar techniques for ED model
Alex Antonarakis
There are uncertainties in regional and global terrestrial carbon budgets  due to the current state heterogeneity of forest ecosystems, the dynamics of carbon storage, and the changes in forest ecosystems resulting from disturbance and recovery processes. Therefore, consistent measurements of canopy structure and forest attributes such as canopy height, vegetation age, trunk diameter and height, canopy gap, aboveground biomass, and species identification amongst others, may help in providing information regarding the current state forest structure that is vital for assessments of current and future forest sustainability, biodiversity, and terrestrial carbon budgets.
Lidar and radar remote sensing techniques are capable of these measurements, with full waveform lidar measurements at near-infrared pulse emissions being sensitive to the vertical vegetation profile, and radar measurements a P, L, and C-bands sensitive to forest volume and density.
Lidar and radar remote sensing techniques on their own right are powerful tools in determining forest structure. In this project they will be investigated separately and also fused at the signal and parameter level to extract 3D forest structure and biomass at a variety of distinct ecosystems. The uncertainties in determining these parameters at different scales and spatial heterogeneity will be quantified by simulating regional carbon fluxes and performing sensitivity analyses using the Ecosystem Demography (ED) model. An end goal of this research is to assimilate lidar and radar remote sensing measurements of vegetation structure into the ED biosphere model in order to improve predictions of log-term vegetation responses to climate change. With this information the two active remote sensing techniques will be considered for global coverage on a spaceborne mission concept.

Monitorning tropical tree growth and mortality rates in relation to climage changegraph

Shirley Xiaobi Dong
Over 10,000 dendrometer bands have been installed on size- and space-stratified
samples of trees at various forest plot sites. In this way we monitor and record the circumference growth and mortality of trees from two to four times per year. The knowledge of how climate influences tree demographic rates will help us better understand the structure and dynamics of different tropical forests, and explain how and why they will change differently in relation to different changes in climate.

 

 

Past Research Projects

Impacts of disturbance, climate variability & CO2 on the North American carbon cycle

Spatial dynamics of fires & forest pathogen outbreaks in Yellowstone National Park

Influence of Hemlock Wooly Adelgid on forest dynamics & the terrestrial carbon cycle