Anne Huth and Jennifer Hamblen (UA HWR) taking water quality measurments at the Boquillas site

D. Williams (UA-RNR), D. Goodrich (USDA-ARS/UA), R. Scott (USDA-ARS), G. Lin (Biosphere 2)

The goal of this work is to quantify many of the water, nutrient, and energy exchanges of a mature, riparian forest ecosystem and identify the eco-physiological mechanisms that are responsible for these exchanges. An exceptional strength to this project is that many different aspects of the ecosystem cycling are being looked at simultaneously via an interdisciplinary approach.

Science goals

Approach

The main activities for the reporting period were to complete the first full year (2001) and begin the second year (2002) of monitoring mesquite biohydrology. Continuous monitoring includes distributed meteorological stations, above-canopy water, carbon and energy exchanges, stem- and branch-level sapflow, soil water and energy stores, groundwater depth, and stream stage. Additionally, a series of intensive monitoring campaigns were conducted 19 - 23 September 2001, 12 - 19 June 2002. These dates were chosen to capture the effects of variable climatic forcing on ecosystem functioning. During these campaigns, additional measurements were made of leaf gas exchange, leaf area index, leaf water potential, plant water isotopes, atmospheric profiles of carbon and water isotopes, understory water/carbon/energy exchanges, and soil respiration.

Results

First, we have gathered considerable evidence to suggest that there are two principal water sources for the riparian terrace ecosystem. The deep-rooted mesquite trees are accessing groundwater and appear mainly to rely on this relatively stable source. The understory vegetation, however, is highly dependent on recent precipitation and is active mainly during the summer monsoon. Evidence for this conclusion is based on the nearly constant overstory water use (i.e., mainly tree transpiration) from the pre-monsoon to monsoon period and the nearly constant relationship between tree water use and water table fluctuations (Figure 3-1) during 2001. Thus, changes to total ecosystem water use were due principally to changes in the understory evapotranspiration. These results, determined by micrometeorological techniques, indicate that the maximum tree water use seems to be nearly constant at an estimated rate of 2.5 - 3.0 mm/day throughout the growing season (i.e., after the leaves were fully mature around 1 June). These results emphasize the importance of separating out understory versus overstory water use in mesquite forests. The growing season for the mesquite was bounded by the last spring freeze and the first fall freeze. Since the nighttime temperatures within the riparian corridor of the San Pedro are typically 5 - 10 ^oC less than the surrounding valley floor, the growing season is only about 5 months.

Secondly, the ability of the trees to access a stable water source and the dependence of the understory on precipitation leads to interesting results in regards to ecosystem nutrient (C and N) fluxes. Net ecosystem uptake of carbon dioxide actually decreased during the rainy season even though the understory vegetation (and thus a greater total vegetation water use and carbon uptake) was active during the monsoon. This decreased uptake was due to the greater increase in soil respiration. Mesquite is a C₃, N-fixing legume that produces very high-quality litter compared to the C₄ grasses. When mesquite roots reach stable groundwater supplies, relatively large quantities of this litter can be added to the soil because photosynthesis and growth will no longer be limited by plant water availability. Large quantities of C and N are accumulating in the mature mesquite community especially in the surface litter. The presence of large amounts of carbohydrates and amino acids in the soils is of interest as these compounds are generally short-lived in soil due to the preferential metabolism by soil microorganisms. The large quantity of available C that can be readily degraded in the mesquite soil was confirmed with soil cores collected from the mesquite community that were incubated in the laboratory under moist field conditions for 80 d. The large amounts of litter that persist in the mesquite community are due to soil-water limitations that restrict the activities of soil microorganisms. Thus, shifts in precipitation patterns to greater/lesser summer events will speed/decrease the mineralization of the litter layer present in the mesquite communities.

Our assembled research team constitutes a highly integrated, multi-disciplinary team. The benefits of this multi-disciplinary approach and longer-term funding cycle of SAHRA have enabled us to look at the how component fluxes of the riparian vegetation water and nutrient cycling are related and interact (e.g., how the eddy fluxes of carbon dioxide relate to the cycling of carbon and nitrogen in the soil). One of the broader implications of this work will be a novel understanding of riparian vegetation functioning.

Future Plans

• Continuous eddy flux measurements will be collected through 2003 in order to determine the effects of wintertime precipitation on ecosystem fluctuations.



• Additional intensive measurement campaigns are planned for August 12-19, 2002 along with an attempt to monitor understory/overstory flux partitioning using stable isotopes immediately after and during the drydown of a large rain pulse.



• We are expanding this work to additional sites along a grassland to forest mesquite invasion gradient to look at the consequences of mesquite invasion on nutrient and water fluxes. An additional collaborator, Dr. Travis Huxman (Ecology and Evolutionary Biology faculty, UA) has been brought on to help in this effort.



• We will also be developing a GIS-based, riparian ET tool to assist land managers in helping to identify and predict total riparian ecosystem water use.