Statistical upscaling of ecosystem CO2 fluxes across the terrestrial tundra and boreal domain: Regional patterns and uncertainties.

The regional variability in tundra and boreal carbon dioxide (CO2 ) fluxes can be high, complicating efforts to quantify sink-source patterns across the entire region. Statistical models are increasingly used to predict (i.e., upscale) CO2 fluxes across large spatial domains, but the reliability of different modeling techniques, each with different specifications and assumptions, has not been assessed in detail. Here, we compile eddy covariance and chamber measurements of annual and growing season CO2 fluxes of gross primary productivity (GPP), ecosystem respiration (ER), and net ecosystem exchange (NEE) during 1990-2015 from 148 terrestrial high-latitude (i.e., tundra and boreal) sites to analyze the spatial patterns and drivers of CO2 fluxes and test the accuracy and uncertainty of different statistical models. CO2 fluxes were upscaled at relatively high spatial resolution (1 km2 ) across the high-latitude region using five commonly used statistical models and their ensemble, that is, the median of all five models, using climatic, vegetation, and soil predictors. We found the performance of machine learning and ensemble predictions to outperform traditional regression methods. We also found the predictive performance of NEE-focused models to be low, relative to models predicting GPP and ER. Our data compilation and ensemble predictions showed that CO2 sink strength was larger in the boreal biome (observed and predicted average annual NEE -46 and -29 g C m-2  yr-1 , respectively) compared to tundra (average annual NEE +10 and -2 g C m-2  yr-1 ). This pattern was associated with large spatial variability, reflecting local heterogeneity in soil organic carbon stocks, climate, and vegetation productivity. The terrestrial ecosystem CO2 budget, estimated using the annual NEE ensemble prediction, suggests the high-latitude region was on average an annual CO2 sink during 1990-2015, although uncertainty remains high.

Author Correction: Neglecting diurnal variations leads to uncertainties in terrestrial nitrous oxide emissions.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Lakes as nitrous oxide sources in the boreal landscape.

Estimates of regional and global freshwater N2 O emissions have remained inaccurate due to scarce data and complexity of the multiple processes driving N2 O fluxes the focus predominantly being on summer time measurements from emission hot spots, agricultural streams. Here, we present four-season data of N2 O concentrations in the water columns of randomly selected boreal lakes covering a large variation in latitude, lake type, area, depth, water chemistry, and land use cover. Nitrate was the key driver for N2 O dynamics, explaining as much as 78% of the variation of the seasonal mean N2 O concentrations across all lakes. Nitrate concentrations varied among seasons being highest in winter and lowest in summer. Of the surface water samples, 71% were oversaturated with N2 O relative to the atmosphere. Largest oversaturation was measured in winter and lowest in summer stressing the importance to include full year N2 O measurements in annual emission estimates. Including winter data resulted in fourfold annual N2 O emission estimates compared to summer only measurements. Nutrient-rich calcareous and large humic lakes had the highest annual N2 O emissions. Our emission estimates for Finnish and boreal lakes are 0.6 and 29 Gg N2 O-N/year, respectively. The global warming potential of N2 O from lakes cannot be neglected in the boreal landscape, being 35% of that of diffusive CH4 emission in Finnish lakes.

Ecosystem carbon response of an Arctic peatland to simulated permafrost thaw.

Permafrost peatlands are biogeochemical hot spots in the Arctic as they store vast amounts of carbon. Permafrost thaw could release part of these long-term immobile carbon stocks as the greenhouse gases (GHGs) carbon dioxide (CO2 ) and methane (CH4 ) to the atmosphere, but how much, at which time-span and as which gaseous carbon species is still highly uncertain. Here we assess the effect of permafrost thaw on GHG dynamics under different moisture and vegetation scenarios in a permafrost peatland. A novel experimental approach using intact plant-soil systems (mesocosms) allowed us to simulate permafrost thaw under near-natural conditions. We monitored GHG flux dynamics via high-resolution flow-through gas measurements, combined with detailed monitoring of soil GHG concentration dynamics, yielding insights into GHG production and consumption potential of individual soil layers. Thawing the upper 10-15 cm of permafrost under dry conditions increased CO2 emissions to the atmosphere (without vegetation: 0.74 ± 0.49 vs. 0.84 ± 0.60 g CO2 -C m-2  day-1 ; with vegetation: 1.20 ± 0.50 vs. 1.32 ± 0.60 g CO2 -C m-2  day-1 , mean ± SD, pre- and post-thaw, respectively). Radiocarbon dating (14 C) of respired CO2 , supported by an independent curve-fitting approach, showed a clear contribution (9%-27%) of old carbon to this enhanced post-thaw CO2 flux. Elevated concentrations of CO2 , CH4 , and dissolved organic carbon at depth indicated not just pulse emissions during the thawing process, but sustained decomposition and GHG production from thawed permafrost. Oxidation of CH4 in the peat column, however, prevented CH4 release to the atmosphere. Importantly, we show here that, under dry conditions, peatlands strengthen the permafrost-carbon feedback by adding to the atmospheric CO2 burden post-thaw. However, as long as the water table remains low, our results reveal a strong CH4 sink capacity in these types of Arctic ecosystems pre- and post-thaw, with the potential to compensate part of the permafrost CO2 losses over longer timescales.

A plant-microbe interaction framework explaining nutrient effects on primary production.

In most terrestrial ecosystems, plant growth is limited by nitrogen and phosphorus. Adding either nutrient to soil usually affects primary production, but their effects can be positive or negative. Here we provide a general stoichiometric framework for interpreting these contrasting effects. First, we identify nitrogen and phosphorus limitations on plants and soil microorganisms using their respective nitrogen to phosphorus critical ratios. Second, we use these ratios to show how soil microorganisms mediate the response of primary production to limiting and non-limiting nutrient addition along a wide gradient of soil nutrient availability. Using a meta-analysis of 51 factorial nitrogen-phosphorus fertilization experiments conducted across multiple ecosystems, we demonstrate that the response of primary production to nitrogen and phosphorus additions is accurately predicted by our stoichiometric framework. The only pattern that could not be predicted by our original framework suggests that nitrogen has not only a structural function in growing organisms, but also a key role in promoting plant and microbial nutrient acquisition. We conclude that this stoichiometric framework offers the most parsimonious way to interpret contrasting and, until now, unresolved responses of primary production to nutrient addition in terrestrial ecosystems.

Tundra landscape heterogeneity, not interannual variability, controls the decadal regional carbon balance in the Western Russian Arctic.

Across the Arctic, the net ecosystem carbon (C) balance of tundra ecosystems is highly uncertain due to substantial temporal variability of C fluxes and to landscape heterogeneity. We modeled both carbon dioxide (CO2 ) and methane (CH4 ) fluxes for the dominant land cover types in a ~100-km2 sub-Arctic tundra region in northeast European Russia for the period of 2006-2015 using process-based biogeochemical models. Modeled net annual CO2 fluxes ranged from -300 g C m-2  year-1 [net uptake] in a willow fen to 3 g C m-2  year-1 [net source] in dry lichen tundra. Modeled annual CH4 emissions ranged from -0.2 to 22.3 g C m-2  year-1 at a peat plateau site and a willow fen site, respectively. Interannual variability over the decade was relatively small (20%-25%) in comparison with variability among the land cover types (150%). Using high-resolution land cover classification, the region was a net sink of atmospheric CO2 across most land cover types but a net source of CH4 to the atmosphere due to high emissions from permafrost-free fens. Using a lower resolution for land cover classification resulted in a 20%-65% underestimation of regional CH4 flux relative to high-resolution classification and smaller (10%) overestimation of regional CO2 uptake due to the underestimation of wetland area by 60%. The relative fraction of uplands versus wetlands was key to determining the net regional C balance at this and other Arctic tundra sites because wetlands were hot spots for C cycling in Arctic tundra ecosystems.

Increased nitrous oxide emissions from Arctic peatlands after permafrost thaw.

Permafrost in the Arctic is thawing, exposing large carbon and nitrogen stocks for decomposition. Gaseous carbon release from Arctic soils due to permafrost thawing is known to be substantial, but growing evidence suggests that Arctic soils may also be relevant sources of nitrous oxide (N2O). Here we show that N2O emissions from subarctic peatlands increase as the permafrost thaws. In our study, the highest postthaw emissions occurred from bare peat surfaces, a typical landform in permafrost peatlands, where permafrost thaw caused a fivefold increase in emissions (0.56 ± 0.11 vs. 2.81 ± 0.6 mg N2O m-2 d-1). These emission rates match those from tropical forest soils, the world's largest natural terrestrial N2O source. The presence of vegetation, known to limit N2O emissions in tundra, did decrease (by ∼90%) but did not prevent thaw-induced N2O release, whereas waterlogged conditions suppressed the emissions. We show that regions with high probability for N2O emissions cover one-fourth of the Arctic. Our results imply that the Arctic N2O budget will depend strongly on moisture changes, and that a gradual deepening of the active layer will create a strong noncarbon climate change feedback.

Degradation potentials of dissolved organic carbon (DOC) from thawed permafrost peat.

Global warming can substantially affect the export of dissolved organic carbon (DOC) from peat-permafrost to aquatic systems. The direct degradability of such peat-derived DOC, however, is poorly constrained because previous permafrost thaw studies have mainly addressed mineral soil catchments or DOC pools that have already been processed in surface waters. We incubated peat cores from a palsa mire to compare an active layer and an experimentally thawed permafrost layer with regard to DOC composition and degradation potentials of pore water DOC. Our results show that DOC from the thawed permafrost layer had high initial degradation potentials compared with DOC from the active layer. In fact, the DOC that showed the highest bio- and photo-degradability, respectively, originated in the thawed permafrost layer. Our study sheds new light on the DOC composition of peat-permafrost directly upon thaw and suggests that past estimates of carbon-dioxide emissions from thawed peat permafrost may be biased as they have overlooked the initial mineralization potential of the exported DOC.

Warming of subarctic tundra increases emissions of all three important greenhouse gases - carbon dioxide, methane, and nitrous oxide.

Rapidly rising temperatures in the Arctic might cause a greater release of greenhouse gases (GHGs) to the atmosphere. To study the effect of warming on GHG dynamics, we deployed open-top chambers in a subarctic tundra site in Northeast European Russia. We determined carbon dioxide (CO2 ), methane (CH4 ), and nitrous oxide (N2 O) fluxes as well as the concentration of those gases, inorganic nitrogen (N) and dissolved organic carbon (DOC) along the soil profile. Studied tundra surfaces ranged from mineral to organic soils and from vegetated to unvegetated areas. As a result of air warming, the seasonal GHG budget of the vegetated tundra surfaces shifted from a GHG sink of -300 to -198 g CO2 -eq m-2 to a source of 105 to 144 g CO2 -eq m-2 . At bare peat surfaces, we observed increased release of all three GHGs. While the positive warming response was dominated by CO2 , we provide here the first in situ evidence of increasing N2 O emissions from tundra soils with warming. Warming promoted N2 O release not only from bare peat, previously identified as a strong N2 O source, but also from the abundant, vegetated peat surfaces that do not emit N2 O under present climate. At these surfaces, elevated temperatures had an adverse effect on plant growth, resulting in lower plant N uptake and, consequently, better N availability for soil microbes. Although the warming was limited to the soil surface and did not alter thaw depth, it increased concentrations of DOC, CO2, and CH4 in the soil down to the permafrost table. This can be attributed to downward DOC leaching, fueling microbial activity at depth. Taken together, our results emphasize the tight linkages between plant and soil processes, and different soil layers, which need to be taken into account when predicting the climate change feedback of the Arctic.

Heterogeneity of carbon loss and its temperature sensitivity in East-European subarctic tundra soils.

Arctic peatlands store large stocks of organic carbon which are vulnerable to the climate change but their fate is uncertain. There is increasing evidence that a part of it will be lost as a result of faster microbial mineralization. We studied the vulnerability of 3500-5900 years old bare peat uplifted from permafrost layers by cryogenic processes to the surface of an arctic peat plateau. We aimed to find biotic and abiotic drivers of CLOSS from old peat and compare them with those of adjacent, young vegetated soils of the peat plateau and mineral tundra. The soils were incubated in laboratory at three temperatures (4°C, 12°C and 20°C) and two oxygen levels (aerobic, anaerobic). CLOSS was monitored and soil parameters (organic carbon quality, nutrient availability, microbial activity, biomass and stoichiometry, and extracellular oxidative and hydrolytic enzyme pools) were determined. We found that CLOSS from the old peat was constrained by low microbial biomass representing only 0.22% of organic carbon. CLOSS was only slightly reduced by the absence of oxygen and exponentially increased with temperature, showing the same temperature sensitivity under both aerobic and anaerobic conditions. We conclude that carbon in the old bare peat is stabilized by a combination of physical, chemical and biological controls including soil compaction, organic carbon quality, low microbial biomass and the absence of plants.

Neglecting diurnal variations leads to uncertainties in terrestrial nitrous oxide emissions.

Nitrous oxide (N2O) is an important greenhouse gas produced in soil and aquatic ecosystems. Its warming potential is 296 times higher than that of CO2. Most N2O emission measurements made so far are limited in temporal and spatial resolution causing uncertainties in the global N2O budget. Recent advances in laser spectroscopic techniques provide an excellent tool for area-integrated, direct and continuous field measurements of N2O fluxes using the eddy covariance method. By employing this technique on an agricultural site with four laser-based analysers, we show here that N2O exchange exhibits contrasting diurnal behaviour depending upon soil nitrogen availability. When soil N was high due to fertilizer application, N2O emissions were higher during daytime than during the night. However, when soil N became limited, emissions were higher during the night than during the day. These reverse diurnal patterns supported by isotopic analyses may indicate a dominant role of plants on microbial processes associated with N2O exchange. This study highlights the potential of new technologies in improving estimates of global N2O sources.

Environmental and microbial factors influencing methane and nitrous oxide fluxes in Mediterranean cork oak woodlands: trees make a difference.

Cork oak woodlands (montado) are agroforestry systems distributed all over the Mediterranean basin with a very important social, economic and ecological value. A generalized cork oak decline has been occurring in the last decades jeopardizing its future sustainability. It is unknown how loss of tree cover affects microbial processes that are consuming greenhouse gases in the montado ecosystem. The study was conducted under two different conditions in the natural understory of a cork oak woodland in center Portugal: under tree canopy (UC) and open areas without trees (OA). Fluxes of methane and nitrous oxide were measured with a static chamber technique. In order to quantify methanotrophs and bacteria capable of nitrous oxide consumption, we used quantitative real-time PCR targeting the pmoA and nosZ genes encoding the subunit of particulate methane mono-oxygenase and catalytic subunit of the nitrous oxide reductase, respectively. A significant seasonal effect was found on CH4 and N2O fluxes and pmoA and nosZ gene abundance. Tree cover had no effect on methane fluxes; conversely, whereas the UC plots were net emitters of nitrous oxide, the loss of tree cover resulted in a shift in the emission pattern such that the OA plots were a net sink for nitrous oxide. In a seasonal time scale, the UC had higher gene abundance of Type I methanotrophs. Methane flux correlated negatively with abundance of Type I methanotrophs in the UC plots. Nitrous oxide flux correlated negatively with nosZ gene abundance at the OA plots in contrast to that at the UC plots. In the UC soil, soil organic matter had a positive effect on soil extracellular enzyme activities, which correlated positively with the N2O flux. Our results demonstrated that tree cover affects soil properties, key enzyme activities and abundance of microorganisms and, consequently net CH4 and N2O exchange.

The uncertain climate footprint of wetlands under human pressure.

Significant climate risks are associated with a positive carbon-temperature feedback in northern latitude carbon-rich ecosystems, making an accurate analysis of human impacts on the net greenhouse gas balance of wetlands a priority. Here, we provide a coherent assessment of the climate footprint of a network of wetland sites based on simultaneous and quasi-continuous ecosystem observations of CO2 and CH4 fluxes. Experimental areas are located both in natural and in managed wetlands and cover a wide range of climatic regions, ecosystem types, and management practices. Based on direct observations we predict that sustained CH4 emissions in natural ecosystems are in the long term (i.e., several centuries) typically offset by CO2 uptake, although with large spatiotemporal variability. Using a space-for-time analogy across ecological and climatic gradients, we represent the chronosequence from natural to managed conditions to quantify the "cost" of CH4 emissions for the benefit of net carbon sequestration. With a sustained pulse-response radiative forcing model, we found a significant increase in atmospheric forcing due to land management, in particular for wetland converted to cropland. Our results quantify the role of human activities on the climate footprint of northern wetlands and call for development of active mitigation strategies for managed wetlands and new guidelines of the Intergovernmental Panel on Climate Change (IPCC) accounting for both sustained CH4 emissions and cumulative CO2 exchange.

Trophic transfer of polychlorinated biphenyls (PCB) in a boreal lake ecosystem: testing of bioaccumulation models.

Understanding the fate of persistent organic chemicals in the environment is fundamental information for the successful protection of ecosystems and humans. A common dilemma in risk assessment is that monitoring data reveals contaminant concentrations in wildlife, while the source concentrations, route of uptake and acceptable source concentrations remain unsolved. To overcome this problem, different models have been developed in order to obtain more precise risk estimates for the food webs. However, there is still an urgent need for studies combining modelled and measured data in order to verify the functionality of the models. Studies utilising field-collected data covering entire food webs are particularly scarce. This study aims to contribute to tackling this problem by determining the validity of two bioaccumulation models, BIOv1.22 and AQUAWEBv1.2, for application to a multispecies aquatic food web. A small boreal lake, Lake Kernaalanjärvi, in Finland was investigated for its food web structure and concentrations of PCBs in all trophic levels. Trophic magnification factors (TMFs) were used to measure the bioaccumulation potential of PCBs, and the site-specific environmental parameters were used to compare predicted and observed concentrations. Site-specific concentrations in sediment pore water did not affect the modelling endpoints, but accurate site-specific measurements of freely dissolved concentrations in water turned out to be crucial for obtaining realistic model-predicted concentrations in biota. Numerous parameters and snapshot values affected the model performances, bringing uncertainty into the process and results, but overall, the models worked well for a small boreal lake ecosystem. We suggest that these models can be optimised for different ecosystems and can be useful tools for estimating the bioaccumulation and environmental fate of PCBs.

Seasonal variation in the function and diversity of methanotrophs in the littoral wetland of a boreal eutrophic lake.

Littoral wetlands are responsible for most of the total methane (CH(4) ) emissions from lake ecosystems. We show that seasonally variable hydrological and temperature conditions in the littoral wetland of a eutrophic boreal lake affect the community composition and gene transcription of methanotrophs measured by a particulate methane monooxygenase (pmoA) gene-targeted microarray. Type Ib freshwater-cluster methanotrophs were favoured by the high water level, and CH(4) oxidation was positively correlated with their pmoA gene transcripts. In the dry subsite of the wetland, the more stagnant hydrological conditions in summer and autumn induced the dominance of type II methanotrophs over type I methanotrophs (community composition and pmoA gene transcripts). The relative abundance of type II methanotrophs increased in winter. The results provide new insight into the variation of methanotroph communities across seasons in littoral wetlands.

Effects of nitrogen load on the function and diversity of methanotrophs in the littoral wetland of a boreal lake.

Methane is the second most abundant greenhouse gas in the atmosphere. A major part of the total methane emissions from lake ecosystems is emitted from littoral wetlands. Methane emissions are significantly reduced by methanotrophs, as they use methane as their sole energy and carbon source. Methanotrophic activity can be either activated or inhibited by nitrogen. However, the effects of nitrogen on methanotrophs in littoral wetlands are unknown. Here we report how nitrogen loading in situ affected the function and diversity of methanotrophs in a boreal littoral wetland. Methanotrophic community composition and functional diversity were analyzed with a particulate methane monooxygenase (pmoA) gene targeted microarray. Nitrogen load had no effects on methane oxidation potential and methane fluxes. Nitrogen load activated pmoA gene transcription of type I (Methylobacter, Methylomonas, and LW21-freshwater phylotypes) methanotrophs, but decreased the relative abundance of type II (Methylocystis, Methylosinus trichosporium, and Methylosinus phylotypes) methanotrophs. Hence, the overall activity of a methanotroph community in littoral wetlands is not affected by nitrogen leached from the catchment area.

Quantification of greenhouse gas emissions from sludge treatment wetlands.

Constructed wetlands are nowadays successfully employed as an alternative technology for wastewater and sewage sludge treatment. In these systems organic matter and nutrients are transformed and removed by a variety of microbial reaction and gaseous compounds such as methane (CH(4)) and nitrous oxide (N(2)O) may be released to the atmosphere. The aim of this work is to introduce a method to determine greenhouse gas emissions from sludge treatment wetlands (STW) and use the method in a full-scale system. Sampling and analysing techniques used to determine greenhouse gas emissions from croplands and natural wetlands were successfully adapted to the quantification of CH(4) and N(2)O emissions from an STW. Gas emissions were measured using the static chamber technique in 9 points of the STW during 13 days. The spatial variation in the emission along the wetland did not follow some specific pattern found for the temporal variation in the fluxes. Emissions ranged from 10 to 5400 mg CH(4)/m(2)d and from 20 to 950 mgN(2)O/m(2)d, depending on the feeding events. The comparison between the CH(4) and N(2)O emissions of different sludge management options shows that STW have the lowest atmospheric impact in terms of CO(2) equivalent emissions (Global warming potential with time horizon of 100 years): 17 kg CO(2) eq/PE y for STW, 36 kg CO(2) eq/PE y for centrifuge and 162 kg CO(2) eq/PE y for untreated sludge transport, PE means Population Equivalent.

Hydrology is reflected in the functioning and community composition of methanotrophs in the littoral wetland of a boreal lake.

In lake ecosystems a major proportion of methane (CH(4) ) emissions originate from the littoral zone, which can have a great spatial variability in hydrology, soil quality and vegetation. Hitherto, spatial heterogeneity and the effects it has on functioning and diversity of methanotrophs in littoral wetlands have been poorly understood. A diagnostic microarray based on the particulate methane monooxygenase gene coupled with geostatistics was used to analyse spatial patterns of methanotrophs in the littoral wetland of a eutrophic boreal lake (Lake Kevätön, Eastern Finland). The wetland had a hydrology gradient with a mean water table varying from -8 to -25 cm. The wettest area, comprising the highest CH(4) oxidation, had the highest abundance and species richness of methanotrophs. A high water table favoured the occurrence of type Ib methanotrophs, whereas types Ia and II were found under all moisture conditions. Thus the spatial heterogeneity in functioning and diversity of methanotrophs in littoral wetlands is highly dependent on the water table, which in turn varies spatially in relation to the geomorphology of the wetland. We suggest that changes in water levels resulting from regulation of lakes and/or global change will affect the abundance, activity and diversity of methanotrophs, and consequently CH(4) emissions from such systems.

The role of Sphagnum mosses in the methane cycling of a boreal mire.

Peatlands are a major natural source of atmospheric methane (CH4). Emissions from Sphagnum-dominated mires are lower than those measured from other mire types. This observation may partly be due to methanotrophic (i.e., methane-consuming) bacteria associated with Sphagnum. Twenty-three of the 41 Sphagnum species in Finland can be found in the peatland at Lakkasuo. To better understand the Sphagnum-methanotroph system, we tested the following hypotheses: (1) all these Sphagnum species support methanotrophic bacteria; (2) water level is the key environmental determinant for differences in methanotrophy across habitats; (3) under dry conditions, Sphagnum species will not host methanotrophic bacteria; and (4) methanotrophs can move from one Sphagnum shoot to another in an aquatic environment. To address hypotheses 1 and 2, we measured the water table and CH4 oxidation for all Sphagnum species at Lakkasuo in 1-5 replicates for each species. Using this systematic approach, we included Sphagnum spp. with narrow and broad ecological tolerances. To estimate the potential contribution of CH4 to moss carbon, we measured the uptake of delta13C supplied as CH4 or as carbon dioxide dissolved in water. To test hypotheses 2-4, we transplanted inactive moss patches to active sites and measured their methanotroph communities before and after transplantation. All 23 Sphagnum species showed methanotrophic activity, confirming hypothesis 1. We found that water level was the key environmental factor regulating methanotrophy in Sphagnum (hypothesis 2). Mosses that previously exhibited no CH4 oxidation became active when transplanted to an environment in which the microbes in the control mosses were actively oxidizing CH4 (hypothesis 4). Newly active transplants possessed a Methylocystis signature also found in the control Sphagnum spp. Inactive transplants also supported a Methylocystis signature in common with active transplants and control mosses, which rejects hypothesis 3. Our results imply a loose symbiosis between Sphagnum spp. and methanotrophic bacteria that accounts for potentially 10-30% of Sphagnum carbon.

Survival of Mycobacterium avium in drinking water biofilms as affected by water flow velocity, availability of phosphorus, and temperature.

Mycobacterium avium is a potential pathogen occurring in drinking water systems. It is a slowly growing bacterium producing a thick cell wall containing mycolic acids, and it is known to resist chlorine better than many other microbes. Several studies have shown that pathogenic bacteria survive better in biofilms than in water. By using Propella biofilm reactors, we studied how factors generally influencing the growth of biofilms (flow rate, phosphorus concentration, and temperature) influence the survival of M. avium in drinking water biofilms. The growth of biofilms was followed by culture and DAPI (4',6'-diamidino-2-phenylindole) staining, and concentrations of M. avium were determined by culture and fluorescence in situ hybridization methods. The spiked M. avium survived in biofilms for the 4-week study period without a dramatic decline in concentration. The addition of phosphorus (10 microg/liter) increased the number of heterotrophic bacteria in biofilms but decreased the culturability of M. avium. The reason for this result is probably that phosphorus increased competition with other microbes. An increase in flow velocity had no effect on the survival of M. avium, although it increased the growth of biofilms. A higher temperature (20 degrees C versus 7 degrees C) increased both the number of heterotrophic bacteria and the survival of M. avium in biofilms. In conclusion, the results show that in terms of affecting the survival of slowly growing M. avium in biofilms, temperature is a more important factor than the availability of nutrients like phosphorus.