The Response of a Subalpine Forest Ecosystem to Earlier Spring Warm-up

Principle Investigator: Russell K. Monson, University of Colorado, Boulder

Abstract: Recent analyses have shown widespread declines in the winter snow pack of mountain ecosystemsin the Western U.S. that are coupled to early-spring temperature anomalies. We hypothesize that early spring warm-up in western forests causes increased water stress and reduces the capacity for the forest to assimilate carbon, while at the same time accelerating the loss of carbon due to soil respiration. We will test this hypothesis using observations and modeling. Our research contains elements of three different NICCR foci, including eddy covariance measurements, modeling and manipulative experiments; however, it is principally intended to fulfill Focus 3, with an emphasis on “the use of measurements and analyses to evaluate mechanisms that might be included in climate and carbon models”.

Location: The studies will be conducted at the Niwot Ridge AmeriFlux site in Colorado. The site is located on U.S. Forest Service land in the Roosevelt National Forest. The manipulation experiment will be conducted on University of Colorado property, 500 m downslope from the AmeriFlux tower.

Questions: The proposed studies have four aims. The first aim is to continue eddy covariance measurements ofCO2, H2O, and energy fluxes at the Niwot Ridge site and extend the data record to 11 years. In additionto maintaining the data record, we will use it to address an important priority: to improve our knowledge of observational errors. The second aim is to conduct a factorial experiment involving manipulation of rhizodeposition and snow depth. In the manipulation study, four formal hypotheses will be tested:1. Low snow and early snow melt cause proportional reductions in xylem sap flow (a measure of transpiration rate, E) and VCmax (a biochemical limitation to gross ecosystem exchange, GEE).2. Xylem sap flow scales proportionally with soil temperature when compared across treatments.3. Springtime needle respiration rate (Rd), is more dependent on stored carbohydrates than current photosynthesis, and does not scale proportionally with GEE, when compared across treatments.4. The concentrations of rhizodeposited sugars and total dissolved organic carbon, and microbial biomass scale proportionally with soil respiration rate (Rs) when compared across treatments. The third aim is to use results from the manipulation experiment to modify and test the Simplified Photosynthesis and Evapotranspiration (SIPNET) model, which will be used to discern the components of NEE that control ecosystem responses to early spring warming. The fourth aim is to use the optimized SIPNET model to partition the anticipated response of NEE to variation in the early spring snow melt over the next 100 years into its component CO2 fluxes (GEE, RE).

Methods: The measurements of NEE and climate will be made using established micrometeorological instrumentation and sensors at the Niwot Ridge AmeriFlux site. In the manipulation experiment we will alter snow deposition with barriers and snow blowers, and eliminate tree rhizodeposition by tree girdling. Photosynthesis, sap flux, water potential, RS and soil sugar content will be measured in the experimental plots. We will use the SIPNET model to assimilate the eddy covariance record and generate optimized estimates of ecosystem process parameters. The optimized model will then be used to predict NEE for selected decades to the year 2100 using scenarios generated by the NCAR Community Climate Model(ver 3) coupled with a regional hydrology model (the Variable Infiltration Capacity model) and local SNOTEL snow pack observations. By the end of the research project, we will be able to: (1) explain which carbon exchange processes will be most affected by early spring warm-up in this ecosystem, (2) determine how to improve an ecosystem process model that is currently favored for conducting model-data assimilation studies, and(3) explain how future changes in the springtime climate of the Rocky Mountains will influence ecosystem carbon exchange. Deliverables will be an 11-year eddy flux data set for use in independent modeling efforts, and an improved ecosystem process model for use in the analysis of NEE from a broad range of Ameriflux sites.

last updated: 28 August 2006 PLH