11 March 2018 in Global Change Biology
6 August 2021 in Journal of Geophysical Research: Biogeosciences
A long‐standing goal of ecology has been to understand the cycling of carbon in forests. This has taken on new urgency with the need to address a rapidly changing climate. Forests serve as long‐term stores for atmospheric CO2, but their continued ability to take up new carbon is dependent on future changes in climate and other factors such as age. We have been measuring many aspects of carbon cycling at an unmanaged evergreen forest in central Maine, USA, for over 25 years. Here we use these data to address questions about the magnitude and control of carbon fluxes and quantify flows and uncertainties between the different pools. A key issue was to assess whether recent climate change and an aging tree population were reducing annual C storage. Total ecosystem C stocks determined from inventory and quantitative soil pits were about 23,300 g C m−2with 46% in live trees, and 48% in the soil. Annual biomass increment in trees at Howland Forest averaged 161 ± 23 g C m−2 yr−1, not significantly different from annual net ecosystem production (NEP = −NEE) of 211 ± 40 g C m−2 y−1measured by eddy covariance. Unexpectedly, there was a small but significant trend of increasing C uptake through time in the eddy flux data. This was despite the period of record including some of the most climate‐extreme years in the last 125. We find a surprising lack of influence of climate variability on annual carbon storage in this mature forest.
18 April 2022 in Journal of Geophysical Research: Biogeosciences
16 December 2022 in Oecologia
10 February 2026 in Tree Physiology
In forests of the southwestern US, the seasonality — or phenology — of tree growth is affected by a combination of limiting temperatures and water availability. But, in this topographically diverse area, temperature and precipitation vary by elevation and therefore may have differing effects on tree phenology across the landscape. Our over-arching research question was: how does variation in temperature and water availability drive differences in the timing of tree growth and water use across species and sites in northern Arizona? We analyze three years of high-frequency measurements of stem growth and sap flow velocity collected across five sites from 1400 to 2600 meters elevation. We supplemented these data with lower-frequency measurements of xylogenesis, budburst, and potential photosynthetic quantum efficiency. Our study species — Pinus ponderosa, Pinus edulis, Juniperus osteosperma, Juniperus scopulorum and Quercus gambelii — represent the dominant tree species across northern Arizona. We found that all metrics of tree phenology tended to be limited by water availability at lower elevations. Lower elevations had bimodal patterns of growth and sap flow velocity which reflected precipitation inputs. Higher elevations had more consistent unimodal patterns which aligned with daylength, and growth was rarely limited by water availability. Microcore data supported growth rates from dendrometers, but microcores were able to capture xylogenesis when dendrometers were unable to — surprisingly, even when stem diameter was shrinking due to water limitation. A machine learning model showed soil volumetric water content was the best predictor of radial growth and sap flow velocity at lower elevations but was only marginally better than daylength and temperature at higher elevations. Thus, tree activity in trailing-edge, low elevation forests was more impacted by moisture, and less affected by temperature, compared to forests at higher elevations. Results, which show that arid woodlands and forests in the American Southwest are adapted to grow opportunistically when neither water availability nor air temperature is limiting, are synthesized in a new conceptual model.