B31B-01
Seasonal Water Availability and Mountain Forest Dynamics in the Southwestern U.S.A.
Water availability in the semi arid United States varies seasonally, annually, and on multi-year cycles, affecting mountain forest carbon exchange directly by influencing primary production and respiration and indirectly through drought, forest fires and insect infestation. Explicitly quantifying the carbon-water relationship in mountain island ecosystems depends largely on understanding both the temporal distribution of water availability and the response of the vegetation to seasonal inputs. Understanding how ecosystems, particularly those which are positioned along sensitive vegetation ecotones and steep transitional climate zones, function and respond to perturbations in climatic conditions is critical to our understanding of the expected range of changes that are possible with changes to our climate. What are the possible responses in ecosystem composition, diversity and biotic feedbacks, not only to changes in the amount of precipitation, but also shifts in its' seasonal pattern, frequency distribution and intensity? Southern Arizona provides a natural laboratory with an annual bimodal precipitation pattern, steep transitional climatic zones, sensitive ecotones, with clear demarcation resulting from large elevation and precipitation gradients and high water limiting conditions with extreme climate variability. This presentation highlights the seasonal carbon-water relationship of a coniferous mountain island forest in the Santa Catalina Mountains of Southern Arizona. Hydro-micrometeorological observations from an eddy covariance flux tower made over five years encompass periods of severe multi-year drought punctuated by strong El Nino winter seasons. Distinct seasonal water use efficiency and carbon assimilation highlights the unique dynamics of this ecosystem and provide part of the fundamental reasons for the observed northward migration of semi-arid coniferous species.
B31B-02
Causes and consequences of tree mortality in Piñon-Juniper woodlands. Introducing an ecosystem scale rainfall manipulation.
Tree mortality as a consequence of drought is widespread worldwide. In Piñon-Juniper woodlands of the
semiarid North American Southwest this phenomenon is patent but the consequences for the functioning of
these ecosystems remain largely unknown. Although several factors have been proposed to explain tree mortality
following drought (e.g. plant desiccation, bark beetle attack), no substantial experimental evidence has been
produced to give mechanistic explanations for the occurrence of these events, nor for the potential effects on the
ecosystem carbon and water cycles following this rapid landscape transformation. In this work, we introduce a
plant-to-ecosystem rainfall manipulation experiment in Piñon-Juniper woodlands at the Sevilleta LTER in
central New Mexico, USA. The goal of our study is to understand the causes (plant-level) and consequences
(ecosystem-level) of tree mortality and/or survival following experimental drought in these woodlands. Framed in
hydraulic concepts involving the soil-plant-atmosphere continuum, we are investigating how Pinus edulis and
Juniperus monosperma would respond to treatments of rainfall diversion and addition in replicated (n=3) 1600
m2 plots. Within this experimental framework, we are first describing the hydraulic architecture of both
species to predict plant allocation patterns (e.g. root vs. leaf area) and assess tree-level water transport capacity
and/or failure. We believe that a thorough understanding of the tree hydraulic characteristics controlling
transpiration will allow us to make robust predictions about the likelihood of plant death or survival during drought
episodes and concomitant effects on the ecosystem rain use efficiency. Pre-treatment results (summer-fall
2006) indicate that transpiration rates per unit of leaf area were highly sensitive to variation in hydraulic
conductance in the soil and plants and varied according to the contrasting vulnerabilities to xylem cavitation
between P. edulis and J. monosperma. In light of these preliminary results we expect to see effects on the
transpiration to evapotranspiration ratio and the soil water storage at depth as trees are subjected to chronic
drought
http:per.ornl.gov/McDowell.html
B31B-03
Plant and Root Growth Responses to Heterogeneous Supplies of Soil Water in Two Coastal Shrubs of California.
Much effort has been focused on identifying plant and root growth responses to heterogeneous supplies of soil nutrients. However, in many circumstances, soil water may limit plant growth and it too can have a patchy distribution. In our research we asked: 1) What is the ecological significance of soil moisture heterogeneity to plant growth in a California coastal dune habitat? 2) How does growth of whole plants and roots respond to soil moisture heterogeneity? and 3) Can roots of these species sense and grow towards moisture-rich areas (hydrotropism) in a natural medium? To address these questions: we conducted comparative field studies of water relations and growth of Artemisia californica and Eriogonum parvifolium; we performed a growth rate study of roots and plants in experimental pots with either patchy or homogeneous distributions of soil water; and we analyzed individual root growth in sand-filled observation chambers in response to moisture-rich patches and resultant soil water gradients. In the field, correlations between daily photosynthetic rates, active leaf display and predawn xylem pressure potentials (ΨPD) indicated that access to water limited growth in A. californica and E. parvifolium. These species, common in habit and habitat, differed in their ability to access water with E. parvifolium having overall higher ΨPD than A. californica (repeated measures ANOVA, P < 0.01). Our growth rate study revealed that patchy supplies of water did not reduce the relative growth rate or average size of E. parvifolium (two-tailed t-tests, P > 0.25). It appears that modified partitioning of growth both at the whole plant and root system level permitted E. parvifolium to maintain growth in patchy soil water conditions. We found that E. parvifolium increased allocation to roots and proliferated in moisture-rich patches in the patchy soil water treatment. Root length density and the proportion of root mass present in the patch was 20- to >100-fold greater in and near the moisture-rich patch than in a comparable but drier soil location (one-tailed matched pairs t-tests, P ≤ 0.05). While root hydrotropism could be a means by which plants are able to locate moisture-rich patches, from our chamber studies we found no compelling evidence for hydrotropic root behavior in seedlings of these two dune shrubs and suggest that roots instead may encounter patches of soil water serendipitously.
B31B-04
Coupled effects of soil heterogeneity and diffusion limitation on nitrogen availability
The heterogeneous distribution of nutrient-rich and nutrient-poor patches in soils strongly affects the rate of nitrogen cycling between organic and inorganic soil compartments. In particular, large differences in carbon-to- nitrogen ratios (C/N) among patches tend to increase cycling and the availability of inorganic nitrogen, which in turn affects plant nutritional status. We address the issue of substrate heterogeneous distribution by analyzing a two compartment model, where each fraction represents a class of micro-sites either mineralizing or immobilizing N from the inorganic forms, according to the C/N ratios of the substrate pools and the decomposer communities. Inorganic N is entirely or only partially shared by the two soil fractions depending on soil moisture and texture. We derive an equivalent lumped model for the macroscopic dynamics and show that the resulting structure embeds different schemes of commonly used biochemical models. We also assess the role of soil texture and soil moisture in limiting the diffusion of mineral N from N-rich to N-poor patches and show how microbial populations in dryer and finer-textured soils tends to experience stronger N-limitation than in more humid and coarser-textured soils. We close with a discussion of the effect of this interplay of micro-scale heterogeneity with soil moisture and textures on plant nutrient availability.
B31B-05
Soil Organic Carbon and Soil Nitrogen in Chinese Soils: Textural, Climatic, and Ecological Influences
The largest terrestrial carbon pool is in the soil, 1.5 to 2.5 times that of vegetation, yet this is also the area with the most uncertainty. Limited understanding of the dynamics and spatial distribution of soil organic carbon (SOC) and soil nitrogen (SN) leads to uncertainty in the global carbon and nitrogen cycles. Regionally, SOC and SN values can be summed across sample profiles to obtain a mean estimate of SOC or SN storage. However, this approach does not effectively account for the spatial variability of SOC and SN within a region, especially when the sampling density of a region is sparse, such as in some areas of western China. This remains one of the main challenges to using regional sums to calculate the total terrestrial carbon flux. This study explores an alternative technique, estimating SOC and SN from easily obtained proxy data. Data are analyzed from the extensive Chinese national soil survey, conducted between 1979 and 1992, containing records on SOC and SN content, soil texture (sand, silt and clay composition percentages), and land cover and land use classifications for over 5000 locations throughout China. These are combined with climate variables (including mean monthly temperature and mean monthly precipitation) to estimate evapotranspiration and plant available water. This is done to determine the relationships of climatic, ecological, land use, and land cover variables to SOC and SN storage. With these relationships, estimates can be produced for SOC and SN storage from larger-scale proxy data, as well as from the effect of land cover and land use changes.
B31B-06
Apatite Weathering and Phosphorus Availability in Deep Regolith, Luquillo Mountains, Puerto Rico
Rapid weathering and erosion rates in mountainous tropical watersheds lead to highly variable soil and saprolite thicknesses which in turn impact nutrient fluxes and biological populations. Here we investigate the weathering of primary minerals containing iron and phosphorous and the role of resident microorganisms in the cycling of these elements in the deep regolith of the Rio Icacos watershed in the Luquillo Experimental Forest, a tropical montane rainforest in northeastern Puerto Rico. In the Rio Icacos watershed, which has one of the fastest documented chemical weathering rates of granitic rocks in the world, the quartz diorite bedrock weathers spheroidally, producing a complex interface comprised of partially weathered rock layers called rindlets. This rindlet zone (about 0.2-3 m thick) is overlain by saprolite (2-8 m) topped by soil (0.5-1 m). Samples were taken from cores augered to 7.5 m on a ridgetop. The profile included about 5 m of regolith (soil \p saprolite), and more than 2.5 m of rindlets. A 0.5 m thick rindlet zone was also sampled in a nearby roadcut. Weathering reactions of primary minerals were examined in thin sections made from rindlets. Total chemistry was measured on all solid samples and Fe, Fe(II), and P were measured in 0.1 M NaOH and 0.5 M HCl extractions performed on augered samples. NaOH-extractable P was assumed to include inorganic and organic P that is bioavailable on both short and long timescales including P associated with secondary Fe(III)- (hydro)oxides. Concentrations of NaOH-extractable P are very low throughout the regolith but increase significantly in the rindlet zone (below 5 m depth). Residual P, believed to include primary apatite and occluded, resistant organic and inorganic forms of P, generally increases with depth. Below 5 m depth, this fraction of P is near zero. Solid-state reaction rates can be calculated for minerals in a weathering profile from the elemental distribution in the profile. Here we quantify a weathering rate for apatite using the P distribution across the rindlet zone in the Rio Icacos watershed. The resulting rate of apatite weathering is 1.1 x 10-14 mol m-2 s-1. In the road cut rindlet sequence, apatite dissolves over the entire 0.5 m rindlet zone, beginning just above the un-weathered bedrock. The high concentrations of P in the NaOH extracts from the rindlet zone samples indicate that the P released during apatite weathering remains relatively accessible within this zone and in the deepest overlying saprolite. HCl-extractable total Fe and Fe(II) follow a similar trend to NaOH-extractable P except in the soil zone (0- 0.5 m), where Fe concentrations are relatively high. In the soil zone, low P concentrations in all measured fractions probably indicate intense biological scavenging of P. High densities of microorganisms in the soil decrease with depth in the regolith, but increase again near the bedrock interface. Nutrient and biological cycles in the deep saprolite profiles are decoupled between the surface layers and the saprolite-bedrock interface zone. While surface communities depend on re-cycled nutrients and atmospheric inputs, deep communities survive primarily off of nutrients released by the weathering bedrock and thus are tightly coupled to processes related to saprolite formation including mineral weathering. Extremely low available P may limit microbial growth within the saprolite, leading to the decoupling of surface and deep communities.
B31B-07
Fire as a Factor of Variation of Soil Respiration in Amazonia of Peru
Severe changes are affecting the role of Amazonia in the Earth system. One of these possible effects could be the modification of the relevance of soil in the carbon cycle. In this sense, fire is an important factor for mobilizing C from the soil to the atmosphere, mainly as CO2. This could have an important effect in the global warming. Our proposal will evaluate the variation of the soil respiration related to the seasonality and the fire effects on soils in the Amazonia of Peru and Brasil. In experimental parcels of four locations of Peru with different vegetation cover (forest and pasture), we will measure soil respiration along with the organic carbon and the microbial biomass of soils during campaigns of wet and dry seasons, with complementary measurements of soil temperature, water and nutrient content. Also, we will reproduce a fire experiment simulating local activity of "slash and burn" to evaluate fire effects. Measurements will be taken after the soil cooled and 1, 3, 5, 7 and 10 days after the fire. Additionally, the carbon stock of the subparcels will be evaluated. Evaluation of the variations of CO2 fluxes and the capacity of adaptation to fire and water content will be done through the comparisons of the different locations, type of soils and concentration of available N as an indicator of nutrient content.