||We propose a multi-faceted research project that will ask and answer a set of hypothesis-driven research questions relating to the ‘Effects of projected precipitation changes on NPP, NEP, and changes in soil carbon storage in western semi-arid ecosystems’ using a suite of eddy covariance flux measurements, field and laboratory manipulation studies and biophysical modeling.
The proposed research will be conducted at a regional cluster of AmeriFlux field sites in central California. The field sites reside in a Mediterranean type climate with wet, cool winters and hot, dry summers. This climate space is an ideal natural laboratory for studying the effects of precipitation on ecosystem function (e.g. assimilation and respiration) and structure (e.g., leaf area index, tree mortality) because the sites receive greater inter-annual variability in annual precipitation (548 +/- 196 mm) than long-term trends in precipitation that have occurred or are predicted by regional climate change models.
The primary objective of this research proposal is to extend the duration of our time series on eddy covariance flux measurements of carbon dioxide, water vapor and energy beyond a decade at the oak savanna and annual grassland AmeriFlux sites; we have been conducting carbon flux measurements at these AmeriFlux sites since 2001. Extending the duration of our flux data record beyond a decade is required to observe the impact of long-term periodicities (e.g. 4 to 11 years) in precipitation, which are associated with El Nino/La Ninas, on ecosystem function and structure. In addition, collecting a carbon flux time series that extends beyond a decade enables us to address unique carbon cycle topics, like legacy or lag effects on ecosystem carbon assimilation and respiration and mortality that are associated with inter-annual variations in rainfall.
The second objective of this research proposal is to study the interactive effects between rainfall and ecosystem-scale carbon dioxide exchange. With our cluster of AmeriFlux field sites we will examine how the interaction between rainfall and carbon dioxide exchange is modulated by: 1) plant functional type (oak woodlands vs annual grasslands) and 2) access to ground water (shallow-rooted grasses and herbs vs deep-rooted trees). Specifically, we are interested in how plant functional type alters the access of roots to groundwater and if such access to groundwater buffers this ecosystem from the vagaries of a highly variable precipitation regime. To answer this question, we have drilled three wells at the oak savanna site and will monitor the daily and seasonal variation in the water table during the extended summer dry period.
The third objective is to use our seasonal carbon flux measurements to study how photodegradation of litter primes rain-induced soil/litter respiration pulses (sunny vs shady sites) and how variations in soil water availability (wet spring vs dry summer) alters the photosynthetic priming of soil respiration. To complement, and interpret, our ecosystem-scale eddy flux measurements, we propose a set of laboratory and field manipulation experiments to study how litter decomposition is affected by photodegradation and how rain-induced pulses in soil/litter respiration are modulated by sun/shade exposure and the number and size of antecedent rain events.
Together, we intend to use these ecosystem-scale carbon flux data and manipulative experiments to develop, test and improve a hierarchy of land-surface models that compute the biophysical fluxes (carbon, water and energy) in climate models. Errors in trace gas fluxes associated with the current class of land surface models can have unintended consequences in predicting direct and indirect carbon-climate feedbacks like how respiration responds to rain, the growth of the planetary boundary layer, the generation of convective clouds and rain. We will compute mass and energy fluxes using radiative transfer models that either consider the oak savanna and grassland as a one-dimensional turbid medium or consider its explicit three-dimensional structure. One objective of this modeling work is to work in tandem with our experimental dataset to produce new insights on the relative accuracy/inaccuracy of using simple, one-dimensional radiative transfer schemes to compute photosynthesis and the surface energy balance in heterogenous canopies, like savanna. A second objective of the modeling work is to use the three-dimensional model to produce a simpler one-dimensional model that parameterizes spatial clumping of oak trees. The third objective of the modeling work is to develop and incorporate new algorithms and models that simulate how ecosystem respiration responds to conditional rain events and how access to ground water moderates ecophysiological drought stress.
We will comply with the BER ‘long-term measure’ by contributing our eddy flux and meteorological data and site metadata to the AmeriFlux database in a regular and timely manner. It is our intension that these data will be used by us and the scientific community to improve, parameterize and validate a variety of models and subroutines that are central to coupled carbon-climate models. For example, none of the coupled climate-carbon cycle models used in the IPCC assessment consider the effects of photodegradation on litter decomposition, simulate rain-induced pulses in soil respiration or consider the photosynthetic priming of soil respiration. Nor do many, if any, coupled carbon cycle-climate models consider the role of vegetation accessing the water table on the ecosystem water budget or light transfer through three-dimensional, vegetation space. With the experiments and manipulations discussed above, we intend to develop new algorithms that may be incorporated into the next generation of coupled carbon cycle-climate models; this step is needed to improve our ability to simulate carbon, water and energy fluxes of semi-arid ecosystems that experience extended periods without rainfall.