Are geothermal streams important sites of nutrient uptake in an agricultural and urbanising landscape (Rotorua,New Zealand)?

1. Lakes in the Rotorua region of New Zealand are affected by eutrophication from urbanisation and agricultural land use. Some lake tributaries contain geothermally influenced waters, and it is currently unknown whether geothermal tributaries are active sites of nutrient cycling or represent point sources of nutrients to the lakes. 2. Using government data sets, we characterised the physicochemical conditions of geothermal and non‐geothermal streams. We then measured ecosystem metabolism and reach‐scale uptake of nitrate (), ammonium () and phosphate () in summer 2010 (n = 8 streams). Finally, we used government data to compare annual nutrient flux from geothermal and non‐geothermal surface water inputs to Lake Rotoiti. 3. As expected, geothermal streams had higher temperature, conductivity and nutrient concentrations and lower pH. However, primary production, community respiration and uptake rates in geothermal streams were not different from those in their non‐geothermal counterparts. Uptake rates of were higher in geothermal streams, and uptake was below detection in geothermal streams, probably due to the saturation by naturally high concentrations. 4. A comparison of Lake Rotoiti inputs suggested that geothermal streams are not significant sources of and , while geothermal inputs of represent an average of 46% of total flux from Lake Rotoiti tributaries. 5. Despite their high temperature and low pH, geothermal streams are active sites of photosynthesis, respiration and and cycling, indicating dynamic biofilm communities. 6. Management options for geothermal streams, if any, should focus on retention (e.g. uptake or coupled nitrification and denitrification) but could prove challenging given the persistent, naturally occurring high flux.     

Nitrogen cycling and dynamics in a macrophyte‐rich stream as determined by a release

1. A tracer release study was conducted in a macrophyte‐rich stream, the River Lilleaa in Denmark. The objectives of the study were to compare uptake rates per unit area of by primary producers and consumers in macrophyte and non‐macrophyte habitats, estimate whole‐stream uptake rates of and compare this to other stream types, and identify the pathways and estimate the rate at which enters the food web in macrophyte and non‐macrophyte habitats. 2. Macrophyte habitats had four times higher primary uptake rates and an equal uptake rate by primary consumers per unit habitat area as compared to non‐macrophyte habitats. These rates represent the lower limit of potential macrophyte effects because the rates will be highly dependent on macrophyte bed height and mean bed height in the River Lilleaa was low compared to typical bed heights in many lowland streams. Epiphytes accounted for 30% of primary uptake in macrophyte habitats, illustrating a strong indirect effect of macrophytes as habitat for epiphytes. N flux per unit habitat area from primary uptake compartments to primary consumers was four times lower in macrophyte habitats compared to non‐macrophyte habitats, reflecting much greater biomass accrual in macrophyte habitats. Thus, we did not find higher N flux from macrophyte habitats to primary consumers compared to non‐macrophyte habitats. 3. Whole‐stream uptake rate was 447 mgN m^?2 day^?1. On a habitat‐weighted basis, fine benthic organic matter (FBOM) accounted for 72% of the whole‐stream uptake rate, and macrophytes and epiphytes accounted for 19 and 8%, respectively. 4. We had expected a priori relatively high whole‐stream N uptake in our study stream compared to other stream types mainly due to generally high biomass and the macrophyte’s role as habitat for autotrophic and heterotrophic organisms, but our results did not confirm this. In comparison with other release study streams, we conclude that nutrient concentration is the overall controlling factor for N uptake rates across streams, mostly as a result of high biomass of primary uptake compartments in streams with high nutrient concentrations in general and not in macrophyte streams in particular. 5. Our results indicate that macrophytes play an important role in the longer‐term retention of N and thus a decrease in net downstream transport during the growing season compared to streams without macrophytes, through direct and indirect effects on the stream reach. Direct effects are high uptake efficiency, low turnover rate (partly due to no direct feeding on macrophytes) and high longevity. An indirect effect is increased sedimentation of FBOM in macrophytes compared to non‐macrophyte habitats and streams which possibly also increase denitrification. Increased retention with macrophyte presence would decrease downstream transport during the growing season and thus the N loading on downstream ecosystems.     

Effects of different Nitrogen forms and osmotic stress on water use efficiency of rice (Oryza sativa)

A hydroponic experiment with simulated water stress induced by polyethylene glycol (PEG) was conducted in greenhouse to study the effects of different nitrogen (N) forms (; and the mixture of and ) on water stress tolerance and water use efficiency (WUE and WUE_T) of different rice cultivars. Two rice cultivars (cv. ‘Shanyou 63’ hybrid indica and ‘Yangdao 6’ indica, China) were grown under non‐water‐ or water‐stressed condition [10% (w/v) PEG, molecular weight 6000] with different N forms for 3 weeks. Under non‐water stress, the biomass of Shanyou 63 was 50.0% and 64.3% and of Yangdao 6 was 6.9% and 87.8% higher under the supply of mixture of and than either under the sole supply of or , respectively; under water stress, the biomass of both rice cultivars decreased in all three nitrogen forms compared with non‐water stress; however, the inhibitory effect of water stress on biomass varied between and nutrition; the reduction of dry matter was significantly higher in than in nutrition. Compared with non‐water stress, under water stressed condition, WUE of both two rice cultivars significantly decreased in supply; WUE did not vary in and the mixture supply. It is concluded that (a) the resistance of water stress of rice seedlings is related to nitrogen form; (b) under water stress, could maintain a higher WUE compared with ; (c) hybrid indica rice seedlings have a higher water stress tolerance than indica rice seedlings.     

PHYCOBILIPROTEIN COMPONENTS AND CHARACTERISTICS OF THE PHYCOBILISOME FROM A THERMOPHILIC CYANOBACTERIUM MYXOSARCINA CONCINNA1

A phycocyanin (PC) and three allophycocyanin (AP) components (designated PC, AP1, AP2, and AP3) were prepared from Myxosarcina concinna Printz phycobilisomes by the native gradient PAGE performed in a neutral buffer system combined with the ion exchange column chromatography on DEAE‐DE52 cellulose. PC contained one β subunit () and two α ones ( and ), and it carried two rod linkers ( and ) and one rod‐core linker (). AP1 and AP3 were characterized as peripheral core APs, whereas AP2 was an inner‐core one. AP2 and AP3 were demonstrated to function as the terminal emitters. Each of the three APs contained two β subunits ( and ), two α subunits ( and ) and an inner‐core linker (). AP2 and AP3 had another subunit of the allophycocyanin B (AP‐B) type () belonging to the β subunit group, and AP1 and AP3 carried their individual specific core linkers ( and ), respectively. No AP component was shown to associate with the core‐membrane linker L_CM. The functions of the linker polypeptides in the phycobilisome (PBS) construction are discussed.     

Isolation by distance in a continuous population under stochastic demographic fluctuations

The local density of individuals is seldom uniform in space and time within natural populations. Yet, formal approaches to the process of isolation by distance in continuous populations have encountered analytical difficulties in describing genetic structuring with demographic heterogeneities, usually disregarding local correlations in the movement and reproduction of genes. We formulate exact recursions for probabilities of identity in continuous populations, from which we deduce definitions of effective dispersal () and effective density (D_e) that generalize results relating spatial genetic structure, dispersal and density in lattice models. The latter claim is checked in simulations where estimates of effective parameters obtained from demographic information are compared with estimates derived from spatial genetic patterns in a plant population evolving in a heterogeneous and dynamic habitat. The simulations further suggest that increasing spatio‐temporal correlations in local density reduce and generally decrease the product , with dispersal kurtosis influencing their sensitivity to density fluctuations. As in the lattice model, the expected relationship between the product and the genetic structure statistic a_r holds under fluctuating density, irrespective of dispersal kurtosis. The product D σ² between observed census density and the observed dispersal rate over one generation will generally be an upwardly biased (up to 400% in simulations) estimator of in populations distributed in spatially aggregated habitats.     

Cover Caption: Plant Vascular System

This issue focuses on the plant vascular system, with a comprehensive review article written by Lucas et al. (pp. 294–388). The cover drawing illustrates the phosphate‐stress signaling and response network (pp. 347–351). A Pi deficiency signal is generated in roots and transported to shoots via the xylem (blue lines). This signal is recognized by source leaves to activate the Pi stress response pathway and then to load the subsequent signals into the phloem (red lines). Phloemmobile RNAs move to roots to increase Pi uptake and alter root architecture . Different phloem‐mobile RNAs are also delivered from source leaves to developing leaves and the shoot apex where they regulate development under Pi‐stress conditions.     

Plant–microbial competition for nitrogen uncoupled from soil C:N ratios

A green house experiment was designed to test the idea that competition for inorganic nitrogen (N) between plants and heterotrophic microorganisms occurs in soils with high C:N ratios, qualifying for N limited microbial activity, but not at low C:N ratios. The short‐term (24 h) ¹⁵N uptake by the grass Festuca gigantea and microorganisms in planted and unplanted soils was determined, and the bacterial activity was measured by the ³H‐thymidine incorporation technique. Two deciduous forest soils, with C:N‐ratios of 20 and 31, and the 20 soil amended with litter to a C:N ratio of 34, were used. A novel and important part of the experimental design was the preparation of the unplanted reference soil with plants present until the competition assay started by the addition of ¹⁵N labelled ammonium ( ) or nitrate ). The results suggested that plants and soil microorganisms competed for mineral N but under influence of other factors than the soil C:N ratio. The plants reduced the microbial uptake of and in the soil with low C:N ratio, which also had the lowest bacterial activity. The plants had a larger N uptake than microorganisms in the two natural soils but smaller in the litter‐amended, and their presence enhanced the bacterial activity, especially in the latter soil. The litter‐amended soil with its high C:N ratio and easily decomposable C was the soil that best fulfilled the criteria for competition, including a net consumption of mineral N during the assay, the lowest plant uptake of mineral N due to the high N immobilization by microorganisms, and a reduced microbial ¹⁵N uptake‐to‐bacterial activity in the presence of plants. Thus, other factors, such as the decomposability of the soil C and the bacterial activity, were more important than the soil C:N ratio to the outcome of plant–microbial competition for N.     

Relative contribution of AtHAK5 and AtAKT1 to K+ uptake in the high‐affinity range of concentrations

The relative contribution of the high‐affinity K⁺ transporter AtHAK5 and the inward rectifier K⁺ channel AtAKT1 to K⁺ uptake in the high‐affinity range of concentrations was studied in Arabidopsis thaliana ecotype Columbia (Col‐0). The results obtained with wild‐type lines, with T‐DNA insertion in both genes and specific uptake inhibitors, show that AtHAK5 and AtAKT1 mediate the ‐sensitive and the Ba²⁺‐sensitive components of uptake, respectively, and that they are the two major contributors to uptake in the high‐affinity range of Rb⁺ concentrations. Using Rb⁺ as a K⁺ analogue, it was shown that AtHAK5 mediates absorption at lower Rb⁺ concentrations than AtAKT1 and depletes external Rb⁺ to values around 1 μM. Factors such as the presence of K⁺ or during plant growth determine the relative contribution of each system. The presence of in the growth solution inhibits the induction of AtHAK5 by K⁺ starvation. In K⁺‐starved plants grown without , both systems are operative, but when is present in the growth solution, AtAKT1 is probably the only system mediating Rb⁺ absorption, and the capacity of the roots to deplete Rb⁺ is reduced.     

Locomotor activity of mice divergently selected for basal metabolic rate: a test of hypotheses on the evolution of endothermy

The aerobic capacity model postulates that high basal metabolic rates (BMR) underlying endothermy evolved as a correlated response to the selection on maximal levels of oxygen consumption () associated with locomotor activity. The recent assimilation capacity model specifically assumes that high BMR evolved as a by‐product of the selection for effective parental care, which required long‐term locomotor activity fuelled by energy assimilated from food. To test both models, we compared metabolic and behavioural correlates in males of laboratory mice divergently selected on body mass‐corrected BMR. elicited by running on the treadmill did not differ between selection lines, which points to the lack of genetic correlation between BMR and . In contrast, there was a positive, genetic correlation between spontaneous long‐term locomotor activity, food intake and BMR. Our results therefore corroborate predictions of the assimilation capacity model of endothermy evolution.     

Water chemistry in the microenvironment of rainbow trout Oncorhynchus mykiss embryos is affected by development,the egg capsule and crowding

The hypothesis tested was that embryonic metabolism affects the water chemistry in the boundary layer. In addition, embryo crowding would further compound the metabolic effect on the water chemistry in the boundary layer. As development progressed, the magnitude of the boundary layer gradients for O₂ and pH, but not for NH, increased. The presence of the egg capsule hindered the diffusion of O₂ into and H⁺ and NH out of the embryo. The magnitude of the O₂, pH and NH boundary layer gradient was significantly increased when embryos were surrounded by either sham embryos or live embryos. The majority of this crowding effect on embryo boundary layers was due to changes in water flow rather than due to metabolism directly. These results clearly show that the microenvironment adjacent to the developing rainbow trout Oncorhynchus mykiss embryo becomes more stagnant as development progresses in the presence of the egg capsule and is further intensified with embryo crowding.     

Effects of moderate ammonium enrichment on three submersed macrophytes under contrasting light availability

1. Increased ammonium concentrations and decreased light availability in a water column have been reported to adversely affect submersed vegetation in eutrophic waters worldwide. 2. We studied the chronic effects of moderate enrichment (NH₄–N: 0.16–0.25 mg L^?1) on the growth and carbon and nitrogen metabolism of three macrophytes (Ceratophyllum demersum, Myriophyllum spicatum and Vallisneria natans) under contrasting light availability in a 2‐month experiment. 3. The enrichment greatly increased the contents of free amino acids and nitrogen in the shoot / leaf of the macrophytes. This indicates that was the dominant N source for the macrophytes. 4. Soluble carbohydrate contents remained relatively stable in the shoot / leaf of the macrophytes irrespective of the treatments. Under ambient light, the starch contents in the shoot / leaf of C. demersum and M. spicatum increased with enrichment, whereas V. natans did not exhibit any change. The starch contents decreased in C. demersum, increased in M. spicatum and remained unchanged in V. natans after the combined treatment of enrichment and reduced light. 5. The enrichment did not affect the growth of the three macrophytes under the ambient light. However, it did suppress the growth of C. demersum and M. spicatum under the reduced light. The results indicate that a moderate enrichment was not directly toxic to the macrophytes although it might change their viability in eutrophic lakes in terms of the carbon and nitrogen metabolism.     

Enhanced exoenzyme activities in sediments in the presence of deposit-feeding Chironomus riparius larvae

1. The combined effects of deposit‐feeding, bioturbation and bioirrigation by benthic macrofauna on the enzymatic hydrolysis of organic matter were studied in microcosms. Chironomus riparius larvae (Insecta, Diptera) served as model macrofauna and stinging nettle leaves (Urtica dioica) were used as a detrital food source. 2. In the upper 10 mm of the sediment (the habitat of C. riparius larvae), the activities of several exoenzymes, the contents of several fractions of particulate organic matter (POM), and the concentrations of dissolved oxidants (O₂, NO) were measured on a small scale. Fluorescent particles (luminophores) were used to quantify the vertical redistribution of particles within the same layer. 3. In control sediment, the addition of detrital food enhanced exoenzyme activities in the 0–2 mm layer only. In the presence of C. riparius larvae, exoenzyme activities increased to 10 mm depth. Further, the content of POM in the 0–2 mm layer was lower in the presence than in the absence of larvae, suggesting ingestion and subduction of the added detritus. After prolonged incubation without further food addition, exoenzyme activities returned close to background values in both treatments, whereas the vertical distribution of POM remained unchanged. 4. The overall penetration depth of O₂ and NO into the sediment was greater in the presence than the absence of C. riparius, the differences being more pronounced after prolonged incubation. Locally high O₂ and NO concentrations due to bioirrigation by C. riparius were measured deep in the sediment. Net downward transport of particles was observed only in the presence of C. riparius larvae and only at the beginning of the incubation. 5. I conclude that deposit‐feeding and bioturbation by macrofauna can quickly remove freshly deposited POM from the sediment surface and transfer it to less oxygenated sites (i.e. animal guts and deep sediment layers). Bioirrigation also increases the availability of oxidants deep in the sediment. The oscillation of oxidant supply to POM particles by ingestion–egestion, burial and re‐burial, and the intermittent bioirrigation of subsurface sediment, is probably the cause of the increased rate of organic matter hydrolysis, the rate‐limiting step in mineralization.     

Distribution of benthic diatoms in U.S. rivers in relation to conductivity and ionic composition

• 1 We quantified the relationships between diatom relative abundance and water conductivity and ionic composition, using a dataset of 3239 benthic diatom samples collected from 1109 river sites throughout the U.S.A. [U.S. Geological Survey National Water‐Quality Assessment (NAWQA) Program dataset]. This dataset provided a unique opportunity to explore the autecology of freshwater diatoms over a broad range of environmental conditions.
• 2 Conductivity ranged from 10 to 14 500 μS cm^?1, but most of the rivers had moderate conductivity (interquartile range 180–618 μS cm^?1). Calcium and bicarbonate were the dominant ions. Ionic composition, however, varied greatly because of the influence of natural and anthropogenic factors.
• 3 Canonical correspondence analysis (CCA) and Monte Carlo permutation tests showed that conductivity and abundances of major ions (HCO + CO, Cl^?, SO, Ca²⁺, Mg²⁺, Na⁺, K⁺) all explained a statistically significant amount of the variation in assemblage composition of benthic diatoms. Concentrations of HCO + CO and Ca²⁺ were the most significant sources of environmental variance.
• 4 The CCA showed that the gradient of ionic composition explaining most variation in diatom assemblage structure ranged from waters dominated by Ca²⁺ and HCO + CO to waters with higher proportions of Na⁺, K⁺, and Cl^?. The CCA also revealed that the distributions of some diatoms correlated strongly with proportions of individual cations and anions, and with the ratio of monovalent to divalent cations.
• 5 We present species indicator values (optima) for conductivity, major ions and proportions of those ions. We also identify diatom taxa characteristic of specific major‐ion chemistries. These species optima may be useful in future interpretations of diatom ecology and as indicator values in water‐quality assessment.

     

Functional linkage between N acquisition strategies and aeration capacities of hydrophytes for efficient oxygen consumption in roots

We evaluated the specific strategies of hydrophytes for root O₂ consumption in relation to N acquisition and investigated whether the strategies varied depending on the aeration capacity. Aeration capacity of roots is an important factor for determining hypoxia tolerance in plants. However, some hydrophytes possessing quite different aeration capacities often co‐occur in wetlands, suggesting that root O₂ consumption also strongly affects hypoxia tolerance. We cultivated Phragmites australis with high aeration capacity and Zizania latifolia with low aeration capacity in hypoxic conditions with NH or NO treatment and compared the growth, N uptake, N assimilation and root respiration between the two species. In Z. latifolia grown with NH treatment, high N uptake activity and restrained root growth led to sufficient N acquisition and decrease in whole‐root respiration rate. These characteristics consequently compensated for the low aeration capacity. In contrast, in P. australis, low N uptake activity was compensated by active root growth, but the whole‐root respiration rate was high. This high root respiration rate was allowed by the high aeration capacity. The O₂ consumption‐related traits of hydrophyte roots were closely correlated with N acquisition strategies, which consequently led to a compensational relationship with the root aeration capacity. It is likely that this functional linkage plays an important role as a core mechanism in the adaptation of plants to hypoxic soils.     

Submersed macrophyte growth at low pH: carbon source influences response to dissolved inorganic carbon enrichment

1. Twenty‐one submersed macrophyte species were grown in the greenhouse at low and high dissolved inorganic carbon (DIC) concentrations and at pH 6 to test the hypothesis that growth response to DIC enrichment in the water column would be greatest for species able to use both bicarbonate (HCO) and free CO₂ from the water column, intermediate for species restricted to free CO₂ in the water column, and least for those species able to use free CO₂ from both sediment porewater and the water column. 2. Relative growth rates (RGR) increased significantly with DIC enrichment for 12 of the 21 species. Bicarbonate users as a group averaged a 4.6‐fold increase in RGR with DIC enrichment, significantly greater than the 2.5‐ and 1.7‐fold increases of the water column CO₂ users and sediment CO₂ users, respectively. 3. The ability to use HCO, as measured by a pH drift technique, was positively correlated with the ratio of RGR at high DIC to RGR at low DIC for the 21 species. 4. Bicarbonate users also exhibited significantly lower belowground : total biomass (R : T) ratios than the other two groups, and alone showed a significant increase in R : T ratio with DIC enrichment. 5. Our results indicate that DIC availability may influence community structure in acidic and circumneutral lakes.     

Relative influence of bioturbation and predation on organic matter processing in river sediments: a microcosm experiment

1. Our objective was to measure the effects of bioturbation and predation on the physical characteristics and biogeochemical processes in river sediments. 2. We investigated the impacts of tubificid worms tested separately and together with an omnivore (Gammarus pulex), which does feed on tubificids, on sediment distribution, water flux, sediment organic carbon, biofilm biomass and microbial activities, and the concentrations of dissolved oxygen, dissolved organic carbon, PO, NO, NO and NH in slow filtration sand–gravel columns. We hypothesised that gammarids, which exploit the top 2–3 cm of the sediment, would modify the impact of worms at the sediment surface. 3. In experiments both with and without gammarids, bioturbation by the tubificids modified both the distribution of surface particles in the sediment column and water flux. In addition, microbial aerobic (oxygen consumption) and anaerobic (denitrification and fermentative decomposition of organic matter) processes in the sediment were stimulated in the presence of tubificid worms. However, G. pulex did not affect either the density or bioturbation activity of the tubificid worms. 4. Bioturbation by the benthos can be a major process in river habitats, contributing to the retention of organic matter in sediment dynamics. The presence of at least one predator had no effect on bioturbation in sediments. In such systems, physical heterogeneity may be sufficient for tubificids to escape from generalist predators, though more specialised ones might have more effect.     

Phosphorus limitation of the freshwater benthic diatom Didymosphenia geminata determined by the frequency of dividing cells

1. Unlike other nuisance algal species, the freshwater benthic diatom Didymosphenia geminata typically forms blooms in low‐nutrient rivers. The negative association between D. geminata blooming behaviour and nutrient levels appears at both catchment and smaller scales. We conducted a series of trials in streamside experimental channels colonised with D. geminata using water from the D. geminata‐affected, oligotrophic Waitaki River, South Island, New Zealand to determine how elevated nitrate and phosphate concentrations affected D. geminata cell division. Because D. geminata blooms are typically most pronounced in unshaded waters, we also investigated the growth response to shading. In all experiments, we used the frequency of dividing cells (FDC) as a metric of cell division. 2. Concentrations of nitrate and dissolved reactive phosphorus (DRP) in the Waitaki River were very low (4 mg m^?3‐N and <1 mg m^?3 DRP). In pilot trials, substrata colonised by D. geminata were subjected to enrichment by either switching the water source toN‐ and P‐rich spring water or by adding a stock solution. Both trials resulted in periods of rapid cell division lasting at least 8 days. 3. Experimental addition of alone triggered an initial cell division which was not sustained. However, addition of alone or together with resulted in prolonged elevation in cell division indicating that the cell division rate was P‐limited. 4. Reduced light levels resulted in decreased FDC in D. geminata in both ambient and N, N + P and P‐enriched river water. 5. Stimulation of D. geminata division rate by addition of above ambient levels confirms that, while blooming behaviour is often associated with oligotrophic rivers, the cells divide faster with greater levels of phosphorus enrichment.     

Microbial thiosulphate reaction arrays: the interactive roles of Fe(III), O2 and microbial strain on disproportionation and oxidation pathways

In this work, we experimentally evaluate pH and  dynamics associated with abiotic and microbial  oxidation under varying [O₂], [Fe(III)] and microbial strain/consortia (two pure strains, Acidithiobacillus ferrooxidans, Acidithiobacillus thiooxidans, their consortia, and two enrichments from an acidic environmental system, Moose Lake 2002 and Moose Lake 2003). Results of the batch experiments demonstrate highly active microbial processing of  while abiotic controls under identical experimental conditions remain static with no pH decrease. When abiotic controls were manually titrated with acid to achieve similar pH decreases to those occurring in the microbial treatments, different S pathways were involved. In particular, disproportionation is a substantial component of initial microbial  processing, and is accelerated by the presence of Fe(III), indicating that recycling of S through intermediate oxidation states is likely to be widespread in acidic mine environments where high [Fe(III)] is common. Furthermore, the microbially mediated S reaction pathways were dependent on both environmental conditions and microbial strain/consortia, indicating that microbial community structure also plays a key role. Collectively, these results highlight the importance of microbial activity, their poor representation by abiotic S models, the likelihood that Fe(III), rather than O₂, is a key control on microbial S processing in acid environments and the need to identify the microbial community/strain involved.     

Bayesian effect estimation accounting for adjustment uncertainty

Summary Model‐based estimation of the effect of an exposure on an outcome is generally sensitive to the choice of which confounding factors are included in the model. We propose a new approach, which we call Bayesian adjustment for confounding (BAC), to estimate the effect of an exposure of interest on the outcome, while accounting for the uncertainty in the choice of confounders. Our approach is based on specifying two models: (1) the outcome as a function of the exposure and the potential confounders (the outcome model); and (2) the exposure as a function of the potential confounders (the exposure model). We consider Bayesian variable selection on both models and link the two by introducing a dependence parameter, , denoting the prior odds of including a predictor in the outcome model, given that the same predictor is in the exposure model. In the absence of dependence (), BAC reduces to traditional Bayesian model averaging (BMA). In simulation studies, we show that BAC, with estimates the exposure effect with smaller bias than traditional BMA, and improved coverage. We, then, compare BAC, a recent approach of Crainiceanu, Dominici, and Parmigiani (2008 , Biometrika 95, 635–651), and traditional BMA in a time series data set of hospital admissions, air pollution levels, and weather variables in Nassau, NY for the period 1999–2005. Using each approach, we estimate the short‐term effects of on emergency admissions for cardiovascular diseases, accounting for confounding. This application illustrates the potentially significant pitfalls of misusing variable selection methods in the context of adjustment uncertainty.     

Effects of irradiance on diel and seasonal patterns of nutrient uptake by stream periphyton

1. Irradiance strongly affects the abundance of stream periphyton communities that in turn influence patterns of instream nutrient uptake. We examined relationships between irradiance and periphyton nutrient uptake taking into account diel and seasonal variation in ambient irradiance. 2. Uptake of dissolved N, P and C by periphyton as areal uptake (U) and demand (V_f) was determined under 11 irradiance levels (0–100% of ambient conditions) using shallow stream‐side experimental channels. Experiments were conducted once per season over one annual cycle with both day and night uptake rates assessed, together with periphyton biomass and autotrophic production rates. 3. No consistent diel variation in areal uptake or demand was detected for the predominant inorganic or total dissolved nutrients even at the highest irradiances. Lack of variation may indicate nutrient limitation, with photosynthetic sequestration and storage of C during the day for subsequent utilisation at night. Alternatively, oxygen consumption by photoautotrophs at night may stimulate compensatory heterotrophic uptake (e.g. denitrification). 4. In all seasons, release of dissolved organic N was detected during the day but to a lesser extent at night. This was not directly related to irradiance levels, indicating that heterotrophic metabolism (e.g. microbial decomposition) contributes to this phenomenon. 5. Areal uptake and demand for the predominant inorganic and total dissolved nutrients increased in response to increasing irradiance in some or all seasons, but rates were typically higher during the spring and summer. Saturation of areal uptake and demand at elevated irradiances was evident during the spring. demand was also saturated at higher irradiances in the summer and autumn. Maximum demand was comparable during spring and summer, but saturation occurred at lower irradiance in summer (24 h average 135–145 μmol m^?2 s^?1) relative to spring (312–424 μmol m^?2 s^?1), indicating more efficient nutrient uptake in summer. Higher total periphyton biomass in summer, but comparable autotrophic biomass (chlorophyll a), implies that heterotrophic metabolism may contribute to this greater efficiency. In spring, autotrophic biomass peaked at an irradiance level of 225 μmol m^?2 s^?1, also suggesting a role for heterotrophic metabolism in demand at higher irradiances. 6. The results of this study show that irradiance levels exert a strong influence on the nature and quantity of instream nutrient uptake with N demand saturated at elevated irradiance levels during the spring, summer and autumn. Our results also suggest that heterotrophic metabolism makes a measurable contribution to instream nutrient uptake even under higher irradiances that favour autotrophic activity.