||This project is collaboration between North Carolina State University and Duke University. The principal investigator at North Carolina State University and Duke University are Dr. Jean-Christophe Domec and Dr. Sari Palmroth, respectively.
An underlying assumption for reduced sensitivity to rainfall and drought and increased carbon sequestration for unmanaged forest is that deep root systems provide a stable supply of water to plants. Plants do not only lose water from the canopy through transpiration, but may also lose a portion of water taken up at night from deep, moist soil layers through flux from roots to shallow, dry soil layers. This process is termed ‘hydraulic redistribution’ (HR). Recent studies suggest that HR can significantly delay further drying of the upper portion of the soil profile by replacing more than 25% of the water utilized during the day with water taken up by deep roots. Furthermore, recent numerical studies have shown that the daily replacement of transpired water by HR can potentially affect land-surface climatology and so the linkages between soil moisture dynamics and rainfall imply that HR may serve as a mechanism for the interaction between deep layer soil-moisture and regional climate dynamics.
The climate projections include forecasts of increasing atmospheric CO2, higher mean and night temperatures and higher vapor pressure deficit (VPD). They all contribute to greater evaporative demand for plants (via increased growth, leaf area and vapor pressure gradient) and decreasing moisture availability. The future predicted higher VPD is likely to increase nighttime transpiration. This will have two significant effects at the ecosystem level: it will result in greater demand on soil water and it will lower the replenishment of surface soil water by HR from deeper layers because the atmosphere will compete with the upper soil for water absorbed at night. This may translate to reduced carbon assimilation and sequestration by canopy trees, as well as understory trees and shrubs. In contrast to managed stands with no understory, unmanaged forest stands, in which a few species tap deep water and most or all tap shallow water, are likely to have their hydrological balance highly dependent on HR. Furthermore, utilizing unique experimental settings such as free-air CO2 enrichment (FACE) to examine the effect of water stress and predicted elevated night transpiration on trees under elevated CO2 will increase our ability to forecast future environmental impacts on tree and forest function and productivity and it is expected that the decrease in HR due to dryer nights under future climatic scenario may in fact be accentuated under elevated CO2.
The first objective of this proposed research will be to investigate the temporal variability of HR in three active Ameriflux sites that our group is already managing, and to use published data of fluxes, soil moisture, root profiles and tree hydraulic characteristics of two inactive Ameriflux sites. Specifically, it will be determined how water redistributed at night by deeper roots increases carbon sequestration and reduces the sensitivity of carbon, water and energy exchange to drought. The second objective will be to study the soil-root interactions and hydraulic interdependence between plant communities. Using a soil-plant-atmosphere model, the third objective will be to forecast how the predicted increases in temperature, vapor pressure deficit and CO2 will affect gross ecosystem productivity and net ecosystem exchange as well as the flux partitioning between understory and overstory species through changes in HR. The fourth objective will be to incorporate HR into a large scale model for further evaluation of the impact of root functioning in land-surface climatology by examining the effect of soil moisture on convective rainfall triggers.
Existing sites that are part of the FLUXNET network we will be used to fit these objectives. The three sites will include two uneven-aged and unmanaged forests: a 80–100-year old oak/hickory forest growing on low fertility, acidic clay-loam soil (US-Duke 2), and a 60–150-year old tupelo/pine/baldcypress forest growing on deep organic soil (Alligator River Site), as well as one drained pine plantation growing on thick organic soil (> 1.8 m; US-NC2). The magnitude of HR will be measured using soil moisture probes installed at several depths. Extensive measurements of root profiles and root resistances to drought will help answer the question of whether deep roots provide sufficient water to maintain photosynthesis and transpiration during drought. Eddy covariance and sapflow data from these different ecosystems will be analyzed via top-down approaches in conjunction with a mechanistic ecosystem soil-plant-atmosphere model to test current understanding of the effects of soil types, drought, VPD, elevated temperature and elevated CO2 concentration on ecosystem respiration and canopy CO2/H2O exchange. Sap flux and gas exchange (leaf-level and chamber) measurements will also be used to evaluate responses of small woody plants and herbaceous components of the ecosystems to VPD and water deficits. Additionally, published data of soil profiles, rooting depths, root functioning and HR from two additional AmeriFlux sites having distinct differences in physical characteristics of the rooting zones will be used to further expand our field measurements to other natural systems (both sites are old-growth forest from the Pacific Northwest: Metolius old site- US-Me4 and Wind River Crane US-Wrc). Data collected at the Duke free-air CO2 enrichment (FACE) site will also be incorporated to assess the effect of elevated CO2 on HR.
This project will provide quantitative information on carbon, water and energy exchange in response to water deficits, elevated future temperature, vapor pressure deficit and atmospheric carbon dioxide across non-managed ecosystems. In addition, coupling the soil-plant-atmosphere model with a biosphere model will predict whether those unmanaged stands will have a greater influence on possible feedback mechanisms between soil moisture and convective rainfall triggers. This will have significant impact on determining whether future climate scenario would affect carbon sequestration of unmanaged stands. Given the potential importance of land-atmosphere feedback for improving short- and long-term weather predictions, a demonstration of the existence of feedback in nature would be of tremendous value. The effects of switching from un-managed forests to plantations would have on future climate will be predicted, as well as the kind of management strategies that could be implemented to optimize carbon sequestration in the future. This project will also foster a recently initiated partnership with the Alligator River National Wildlife Refuge and thus the outcome of this research will aid in developing ecosystem management strategies that can be implemented today to ensure continued quality wildlife habitat and provision of ecosystem services in the face of a global warming. We will also develop a scientific research program that will help protect the highly threatened natural ecosystems throughout the coastal areas of the U.S Southeast.