Methane cycling in temperate forests

Abstract

Temperate forest soils are considered significant methane (CH₄) sinks, but other methane sources and sinks within these forests, such as trees, litter, deadwood, and the production of volatile organic compounds are not well understood. Improved understanding of all CH₄ fluxes in temperate forests could help mitigate CH₄ emissions from other sources and improve the accuracy of global greenhouse gas budgets. This review highlights the characteristics of temperate forests that influence CH₄ flux and assesses the current understanding of the CH₄ cycle in temperate forests, with a focus on those managed for specific purposes. Methane fluxes from trees, litter, deadwood, and soil, as well as the interaction of canopy-released volatile organic compounds on atmospheric methane chemistry are quantified, the processes involved and factors (biological, climatic, management) affecting the magnitude and variance of these fluxes are discussed. Temperate forests are unique in that they are extremely variable due to strong seasonality and significant human intervention. These features control CH₄ flux and need to be considered in CH₄ budgets. The literature confirmed that temperate planted forest soils are a significant CH₄ sink, but tree stems are a small CH₄ source. CH₄ fluxes from foliage and deadwood vary, and litter fluxes are negligible. The production of volatile organic compounds could increase CH₄’s lifetime in the atmosphere, but current in-forest measurements are insufficient to determine the magnitude of any effect. For all sources and sinks more research is required into the mechanisms and microbial community driving CH₄ fluxes. The variability in CH₄ fluxes within each component of the forest, is also not well understood and has led to overestimation of CH₄ fluxes when scaling up measurements to a forest or global scale. A roadmap for sampling and scaling is required to ensure that all CH₄ sinks and sources within temperate forests are accurately accounted for and able to be included in CH₄ budgets and models to ensure accurate estimates of the contribution of temperate planted forests to the global CH₄ cycle.

Background

Methane and its role in global climate change

Methane (CH₄) is a powerful greenhouse gas (GHG). Each molecule has 28–35 times the warming potential equivalent of carbon dioxide (CO₂) over 100 years [1–3]. As such, methane is the second largest contributor to global warming after CO₂ and is responsible for ∼ 30% of the rise in global temperatures since industrialisation [4]. Furthermore, CH₄ oxidation in the troposphere produces ozone (O₃) [5]. Ozone has strong greenhouse gas potential and influence on climate regulation (211 × stronger than CO₂ over a 100 year period [6]), and also has complex effects on atmospheric chemistry and air quality through interaction with nitrogen oxides (NO_x), volatile organic compounds (VOCs), formation of smog and respiratory illness, and the absorption of ultraviolet radiation. As such, CH₄ emissions are regulated under international climate agreements, such as the United Nations Framework Convention on Climate Change and also the convention on Long-Range Transboundary Air Pollution established to address air pollution that travels across national boundaries [4].

Annual CH₄ emissions are estimated at 550–594 million tonnes (CH₄), 60% of which is produced by human activity [4]. These anthropogenic sources have driven a 260% increase in atmospheric CH₄ concentrations relative to pre-industrial times (sensu 1750), and concentrations continue to increase. The current emissions trajectory is noteworthy, as it tracks midway between the two warmest Intergovernmental Panel on Climate Change (IPCC) scenarios (RCP6 and 8.5) [7] with concomitant impacts expected on ocean acidification, sea level rise, occurrence of extreme weather, biodiversity loss, agriculture and food safety, and other areas. Rapid and deep reductions in all GHGs, including CH₄, are needed to meet the targets to limit global warming to 2 °C above pre-industrial times.

Although CH₄ absorbs much more infrared radiation than CO₂, it is also short-lived in the atmosphere. Consequently, most of the impact (radiative forcing/GHG effect) associated with CH₄ occurs in the first few decades after release. For example, CH₄ global warming potential (GWP) in the first 20 years of release is 80.8–82.5 × equivalent of CO₂ [8]. Accordingly, reducing methane emissions will result in significant short-term effects on climate change, constituting one of the most efficient measures to help meet short-term emissions targets [9]. Realistic pathways to keeping the planet within safe thermal boundaries include substantial, early reductions in atmospheric CH₄ concentrations [4].

Methane sources and sinks

Methane is produced from a wide range of natural and anthropogenic sources, including both biological and non-biological processes [10]. As such, the range of sources and sinks (Sensu biological and/or other processes that remove CH₄ from the atmosphere) are diverse. An important definition exists between natural and anthropogenic sources. Natural sources are defined as ‘pre-agricultural’ emissions, i.e. those before adoption of wide-spread agriculture (Neolithic/agricultural revolution ca. 10,000 years before present (BP)). These include wetlands (76% of natural CH₄ emissions), termites (11%), oceans (8%) and CH₄ hydrates (5%). Anthropogenic sources (currently 60% of total emissions; as before) include fossil fuel production (33% of manmade CH₄ emissions), livestock production (27%), rice cultivation (7%), wastewater treatment (7%), and landfills (11%) [11]. Accounting for these emissions involves quantification and estimation of emissions from natural and anthropogenic sources, as well as considering land-use changes, both historic (pre-agriculture) and contemporary.

Biogenic CH₄ emissions (natural or anthropogenic) are a result of microbial activity, primarily methanogenic archaea, during the decomposition of organic matter in anaerobic environments such as in animal gut/intestines or wetland, and peatland soils [12]. More recently, other CH₄-producing microbial groups have been discovered, including fungi [13] and bacteria [14]. There is also a growing literature regarding non-methanogenesis oxic CH₄ production at plant and litters surfaces generating emissions (reviewed by Liu, Xie [15] and Putkinen, Siljanen [16]).

Methane is highly reactive and contributes to the chemistry of the troposphere and stratosphere. Reactions with hydroxyl radicals (OH) comprises the primary sink of CH₄, accounting for 90% of total losses [17]. The other major sink is through biological CH₄ oxidation in soils. This biological sink occurs mainly in aerobic soils and is estimated to remove about 3–10% of annual CH₄ released to the atmosphere [18, 19]. Biological CH₄ oxidation is carried out by methanotrophic microorganisms. Until recently, these were considered to use CH₄ as a primary source of both carbon and energy source and be active under aerobic/oxic conditions [20]. It is now recognised a diverse range of microorganisms can oxidise CH₄ with or without oxygen [11, 21]. Methanotrophs are nearly ubiquitous across natural ecosystems including soils and sediments.

Different land uses affect the size of the soil CH₄ sink. Significant research has focussed on land use systems such as rice production, wetlands, and peatlands which are significant CH₄ sources at a global scale. Conversely, soils in other (typically oxic) systems are net sinks; for example, CH₄ uptake in temperate zones accounts for nearly half of the global soil sink (10.4 Tg CH₄ yr^− 1) [19]. Globally, soils under temperate forested land have the highest methane uptake compared with other land-uses [18, 22]. For example, in New Zealand higher CH₄ oxidation has been measured in soils from pine plantations compared with pasture [23, 24]. Saggar, Tate [22] also found that pine forest soils had higher methane uptake rates compared with pasture and cropping soils, with uptake rates of 4–6 kg CH₄ ha^− 1yr^− 1 and < 1–1.5 kg CH₄ ha^− 1yr^− 1, respectively. Considering the overall global CH₄ budget, temperate forest soils represented the largest terrestrial CH₄ sink, with an average uptake of 7.64 (4.55–10.73) Tg CH₄ yr^− 1, accounting for 51% of the total global forest soil CH₄ uptake [4, 25, 26].

The importance of methane cycling in temperate forests

Despite temperate forest soils being the largest terrestrial CH₄ sink, when compared with other components of the global CH₄ cycle/budget, the relative contribution of temperate planted forests to the global CH₄ budget is minor. Indeed, in a tabulation of sources and sinks the impact of temperate forests was considered insignificant relative to other land uses (e.g., wetlands) or activities (agriculture and waste) and was not recorded individually (Table 3, Saunois, Stavert [4]). However, this belies the importance of forest-related CH₄ cycling at regional to jurisdictional scales; i.e., those at which land management and carbon reporting often occur. For example, in New Zealand nearly 24% of the land area is temperate natural forest, and a further 5% is planted temperate forest [27]. As such, although New Zealand holds only a small amount of Earth’s temperate forest (∼ 0.95%; Fig. 1, under Temperate forests and forest soils), the total carbon budget of the country is highly sensitivite to any changes in carbon exchange in temperate forests. Similarly, Tasmania (Australia), British Colombia (Canada), Washington and Oregon (USA), through to Hokkaido and Honshu (Japan), parts of South Korea, Chile and so forth all have high proportions of temperate forest as part of the total land area. Understanding the carbon balance in these systems is, by extrapolation, an important part of regional carbon budgets and policy informing land use change and management.

Fig. 1.

Global distribution of temperate forests and main tree species, either naturally occurring or planted, in each region

Providing a more accurate understanding of the CH₄ flux in temperate forests is needed and will allow better total budgeting of GHG emissions, particularly for sensitive regions (i.e., as above). However, research on CH₄ emissions in forested ecosystem reveals a far more complex story than previously thought, with an interplay of productive-consumptive, aerobic-anaerobic, biotic-abiotic processes occurring between upland-wetland soils, trees, and atmosphere [4]. That is, the CH₄ cycle within forests is complex, with many individual pathways occurring; nett CH₄ balance at any point in time must integrate these.

The aims of this review are to (1) highlight the features that define temperate forests and have the potential to influence CH₄ flux and (2) assess the current state of knowledge around the CH₄ cycle in temperate forests with particular focus on those being managed for specific purposes. This review considers processes and magnitude of CH₄ fluxes from trees, litter and deadwood, and soil, as well as interaction of canopy-released volatile organic compounds on atmospheric CH₄ chemistry. Methane fluxes will be quantified and factors (biological, climatic, management) affecting the magnitude and variance of these fluxes will be discussed. Based on the finding’s recommendations will be made on where research should be focused to accurately account for CH₄ in greenhouse gas budgets.

Temperate forests and forest soils

Approximately 26% of the ice-free land mass is under forest, but historically this was much greater. About 8,000 years BP, for example, Earth held 6 billion ha of forest; this has declined to ∼ 4 billion ha today [28, 29]. These forests encompass the major biome types, spanning tropical, sub-tropical, temperate, to boreal systems. Within these, ‘primary forests’ are those that are largely free from human impact and have established and grown naturally; these ecosystems maintain high levels of biodiversity due to continuity and range of microhabitats resulting from the complex structure of the forests. They are more typical in tropical and boreal biomes, and most underrepresented in temperate biomes.

This underrepresentation is due to the impact of human activity in temperate forests, which is significant and persistent. These impacts range from entire alteration of landscapes and shift in land use types, through to the directed planting, management, and harvesting of forests themselves. This signature of human activity on temperate forest is evident globally and can be traced back to ancient times. From the Roman empire across Europe, agricultural expansion in China, to fire in North America and impacts elsewhere [30], the expansion of civilisation and alteration of temperate forest ecosystems is a defining characteristic of Earth’s recent history.

Temperate forests cover about ∼ 1.5 billion ha of Earth’s surface (∼ 16% of total forests). The extensive coverage of temperate forests underscores their importance in Earth system processes, encompassing biogeochemical cycles and the provision of ecosystem services. Approximately 17% of global net primary productivity is supported by temperate forests, and more than 15% of all Earth’s terrestrial carbon is held in these forest systems [31]. As such, the exchange of carbon held in temperate forests (spanning the living biomass, to deadwood, litter and the forest floor, as well as the extensive soil reserves) with the atmosphere comprises an important factor influencing the trajectory of climate change [32, 33].

Distribution

Temperate forests are mostly distributed between 30° − 60° latitude in both the northern and southern hemispheres [34] (see Fig. 1). This latitudinal band largely spans the range of climatic conditions that define temperate biomes; typically, a 4–6 month frost free growing season (a key difference to the shorter season in boreal forests) and mean annual temperature of ∼ 5–20 °C. Compared with tropical forests, temperate forests do experience periods of 0 °C or colder and precipitation typically exceeds potential evaporation. However, there are various approaches to defining ‘temperate’, and these are expanded on by de Gouvenain and Silander [31].

The distribution of temperate forest is weighted to the northern hemisphere (∼ 80%) due to the larger extent of land north of the equator. This includes eastern United States, the Pacific Northwest and into Canada. Northern Europe, Turkey, Iran, southern European Russia, southern parts of eastern China, Japan, and Korea (Fig. 1). In the southern hemisphere, temperate forests are distributed through South America including Chile and parts of Patagonia, Tasmania and south-eastern Australia, and the entire extent of New Zealand [30, 31, 34].

Forest structure

In temperate forests under natural conditions, a wide floristic range occurs and this is primarily driven by ecoregion-associated climatic variation [34]. Areas with warm to hot continental climate, and those with marine climates such as the Eastern US and parts of Europe, support mixed-deciduous forests. However, in rainy subtropical temperate zones and where ocean moisture is trapped by mountain ranges and caught by the canopy or forms meteoric rain (temperate rainforest conditions), mixed evergreen forests predominate.

Within ecoregions, landform influences overlay these broader (e.g., climatic) factors affecting tree suitability. Pinaceae and Eucalyptus spp., for example, are well suited to soils of low nutrient status, low organic matter, or dryer (xeric) sites, whereas in moderately wet sites (mesic), beech, oak, and other species are better adapted [28]. However, the drivers of forest structure and composition are complex; local macroclimatic and ecophysiological conditions are expressed against geologic and biogeographic backgrounds and paleoclimatic history [31]. Indicative temperate forest species, either naturally occurring or those planted, in different ecozones are given in Fig. 1. Pests and diseases, fire, windstorm, drought, elevated nitrogen deposition/eutrophication, and other disturbances also influence forest structure and function [35]. These latter factors now have a strong signal of human activity driving the frequency and severity of occurrence [36, 37].

The legacy of human influence on temperate forests is so extensive (likely more than any other forest type) it is difficult to determine what comprises a ‘natural’ or ‘pristine’ state [30]. Indeed, the evolution of these forest systems should be recognised as occurring alongside human activity. Humans imprint is seen in their former and current distributions, but also the tree species that are present. Ironically the rate of human induced climate change may actually necessitate that human intervention is needed to support temperate forests by way of assisted migration or climate conscious planting to increase forest resilience. Natural rate of recruitment/replacement cannot keep pace with change in the habitat conditions (sensu Hutchinsonian niche space [38]). Seedlings growing today may be unsuitable for the conditions of tomorrow. An example are the coniferous temperate forests in California’s Sierra Nevada biome [39], where climate change has led to a vegetation-climate mismatch, effectively stranding forests in locations unsuitable for their future regeneration (sometimes colloquially referred to as ‘zombie forests’). Trees are sessile and long-lived; they cannot migrate from climate change. Paradoxically, the change brought about by humans on Earth’s systems means that the fate of temperate forests will be more connected with human activity than ever before.

Soils

Soil types are typical of those distributed through the latitudinal range. These include five main orders: Alfisols, Entisols, Inceptisols, Spodosols, and Ultisols (Fig. 2), and range in fertility, age and extent of weathering, water retention and other primary properties. However, just as the floristic diversity of forests can exhibit strong local diversity due to spatiotemporal and other factors, soil types too are highly diverse, even at local scale. Similarly, whilst five soil orders support most temperate forests globally, other soils type can be important regionally. In New Zealand’s North Island, for example, ash and tephra deposited by volcanic activity has weathered to form extensive Andosols (Fig. 2). These soils support the growth of extensive natural and planted temperate forests and are therefore an important soil type in forested regions with a history of volcanic activity.

Fig. 2.

Sankey plot connecting the main soil types that support temperate forests and the regions they are found. The left-hand side has the main soil types and key features of each soil type. On the right-hand side, regions are listed along with the proportion of global temperate forests (%) and the area of temperate forests (km² × 10⁶) in each region. Classification based in USDA soil taxonomy.

Like forests themselves (see later), soils can be subject to significant modification from their natural state due to human management. In productive planted forests, this includes alterations to fundamental properties such as pH, inputs to redress nutrient deficiencies, site alteration to increase draining and so forth. More broadly, human intervention in the matching of tree species to sites has been one of the defining themes of temperate forests historically through to today. Many temperate forests are managed to some extent, including choice of tree for site or, at least, interventions such as thinning or selective removal which advantage different forests structures and compositions. Choices such as planting deciduous or conifer-based forests have clear impact on the soil. Podsols (order within Spodosols), for example, form under coniferous or mixed forest with unique properties related to organic matter accumulation, horizon formation, and mobilisation of iron and aluminium oxides or hydroxides. The types of trees planted can strongly influence soil pedogenesis.

Temperate forest soils typically hold much greater levels of organic matter than those in tropical regions. Much of this organic matter rests on the soil surface, comprising a ‘forest floor’ layer of varying thickness. The top of the forest floor comprises the most recently fallen litter such as leaves/needles and wood debris. As this extends downwards to the surface of mineral soil horizon, the state of this resource changes, becoming increasingly humified, with a wider ratio of carbon to nitrogen (C: N) and other elemental changes, and increased moisture content [40]. Microbial biomass in the lower regions of the forest floor are often rich in fungal mycelium, other microorganisms, invertebrates, and other life. Plant roots extend laterally into the lower portions of this layer (particularly the interface of mineral soil and organic matter layer), recapturing nutrients and energy fallen from the canopy and other parts of the forest. Inputs of fresh litter into the forest floor seasonally vary among different temperate forest systems. In the USA, needle fall of conifers and deciduous broad-leaved trees leaf-fall occurs in Autumn, while in New Zealand conifer needle fall occurs in Autumn, and for the native beech forests (Nothofagus spp.) leaf fall is primarily in Spring [41]. Litter-fall is a key process of the nutrient and energy recycling within such forest systems.

Summary of prominent features of temperate forests that influence methane flux

A defining characteristic of temperate forests is the significant influence of human activity on these ecosystems globally. In comparison to other forest systems, extensive areas of temperate forests have been and continue to be profoundly shaped by human intervention, resulting in alterations to their structure and function, including changes in location, species composition, and environmental conditions. Human influence of these forests spans a continuum, from historic influences of land use alteration, agricultural expansion, and global movement in trees species (and pests, diseases, weeds, and herbivores), through to harvest of trees from naturally regenerated stands, selection systems and plantation forests intensively managed for timber production [30]. With increasing global demand for wood and fibre, energy, carbon storage, and other materials/resources that forests can provide, temperate forests are increasingly managed for production of fast-growing tree species.

A further prominent feature of temperate forests is the occurrence of distinct seasonality. Temperate forests can be defined via their latitudinal zone which provides conditions for winter induced dormancy and a defined growth season during the warmer months (as before). This seasonal variability is important as it underpins strong phenological events that drive temperate forest ecology and functionality. As discussed above, litter fall is a key seasonally-triggered event. During this period large shifts in resource allocation occur from canopy to the forest floor and soil, shifting soil nutrient cycling and energy flow through the trophic food webs. Such events profoundly impact biogeochemical processes related to carbon flow and, therefore, are likely to impact exchange of trace gases such as CH₄. Other phenology events include timing of leaf emergence, tree sap flow (sometimes harvested; e.g., maple), fungal sporocarp emergence and senescence, emergence of invertebrates (e.g., cicadas) and so forth. These and other biological events triggered by environmental cues and changes play crucial roles in temperate forest ecosystem ecophysiology and dynamics.

These prominent features mean that processes such as production and consumption of CH₄ are particularly challenging to assess in temperate forests, as the ecophysiological conditions change dramatically over the course of each year. Strong seasonal cycles of temperature and moisture drive biogeochemical cycles and phenological events, such as episodic reallocation of nutrients and energy within the forest system (as described above); these alter carbon cycling dynamics among primary producers, decomposers, and the wider food web. These unique elements of temperate forests must be explicitly considered in CH₄ cycling and budgets. Furthermore, given the extent of human interaction with temperate forests, including extensive management of these for production of wood, fibre, and environmental products [42], there exists significant potential for intervention practices targeted towards influencing ecosystem CH₄ budgets.

Forest methane fluxes

The biological cycling of CH₄ within forests ecosystems comprises a range of processes from soils, canopy, stems and other components. These act as various sources or sinks of CH₄, with each process varying in relative magnitude and direction over time and as conditions change [43]. There are also indirect processes such as the production of volatile organic compounds by trees that can interfere with the natural atmospheric processes that degrade CH₄ [44, 45]. While the critical zones in CH₄ cycling interactions are known - i.e., soil, tree stem, canopy, deadwood, and VOCs – the relative size contribution of each of these to the overall cycle is not well characterised. Soil is the exception and has been reasonably well studied with soils in temperate/upland forests generally considered CH₄ sinks [4, 18, 46–49].

The most common method for measuring CH₄ flux of different forest compartments in situ is with chambers [22, 43, 50]. Chamber measurements are well suited to process-level studies of individual components within the ecosystem [51] and therefore the majority of studies mentioned in the below sections use this method. Another technique for measuring forest CH₄ flux is the eddy covariance method. This method integrates fluxes over a larger area, which results in measurements that are more representative of the ecosystem as a whole [52]. Wang, Murphy [51] used this method to measure CH₄ fluxes in a temperate forest in Ontario, Canada and found that over the measurement period (June to October 2011), the site was a net CH₄ sink. Although, this method and others like it (flux gradient, Eddy accumulation etc [53]). can measure net CH₄ flux at a forest level, this method is generally expensive and not commonly used to measure CH₄ flux in temperate forests.

Forest soils

Size of soil flux

Methane consumption by forest soils is considered the greatest among all land use types [46, 47]. Global annual mean soil CH₄ consumption in temperate forest soils is reported to be 3.6 kg CH₄ ha^− 1 y^− 1 [47]. However, this average belies considerable regional variation. For example, estimates range from 4 to 6 kg CH₄ ha^− 1 y^− 1 in single-species planted forests [22]. This large uncertainty existing within systems largely constrained for tree species and management practices, shows the importance of understanding local and even site-specific variation influencing forest CH₄ emissions and regional CH₄ budgets.

Factors that affect soil flux

Soil CH₄ flux is the net product of the activity of CH₄ producing archaea (methanogens) and CH₄ consuming bacteria (methanotrophs). Factors that influence CH₄ flux are those that affect either gas diffusion and/or microbial activity. For a forest soil to be a CH₄ sink, conditions need to be favourable for methanotrophs and CH₄ oxidation. The ability of methanotrophs to access CH₄ via diffusion is the main factor influencing CH₄ oxidation activity [54–56]. Methane flux in temperate planted forest soils is affected by both environmental conditions and management practices. These factors differ between and even within temperate forests resulting in variation in CH₄ flux. The environmental factors covered in this review are soil water content, soil bulk density, oxygen availability, Eh (redox potential) and pH, temperature, and soil physicochemical properties (e.g., fertility, organic carbon, C:N ratio and other soil nutrients). Management practices include stand age, stand density, use of nitrogen (N) + phosphorus (P) fertiliser, tree species and compaction of soil due to the use of heavy machinery at harvesting.

Environmental factors

Soil water content

Soil water content has been found to affect CH₄ uptake in temperate forests [57–62]. As soil moisture increases, soil aeration decreases and macropores become blocked limiting methanotrophic activity [18, 63]. In extremely wet conditions (e.g. saturated soil), the reduced supply of oxygen can also create anaerobic conditions suitable for methanogens resulting in CH₄ emissions [18, 64]. Drier soils allow for increased diffusion of gas through the soil [57, 59, 65]. However, if soil moisture reduces too much (e.g. below wilting point) CH₄ uptake is limited due to the microbes being physiologically stressed [57]. In forests, tree roots help increase CH₄ uptake by reducing soil moisture [66].

Ni and Groffman [67] found that CH₄ uptake by soils was declining in forests in the northern hemisphere (0 to 60 °N latitude) as precipitation increased. This reduction in the forest CH₄ uptake could mean the total forest sink has been overestimated in several regions across the globe. These results show the sensitivity of soils to changes in precipitation and, therein, CH₄ cycling. There is an urgent need for up to date regional CH₄ budgets, including for New Zealand nationally, to accurately quantify the forest soil CH₄ sink. There is also a need to understand the magnitude and direction change of forest CH₄ cycling in response to predicted future climate conditions to help improve the accuracy of future CH₄ budgeting.

Soil bulk density/soil structure

Methane uptake in forests soils is dependent on physical factors controlling gas diffusion through the soil such as soil moisture (discussed above) and soil bulk density/structure, and compaction (discussed in forest management and silviculture section). Soil bulk density and soil structure directly influence gas diffusion and, therefore, exchange of trace gases such as CH₄. In general, there is a strong association between increased soil porosity and uptake of atmospheric CH₄; soils with medium and fine texture have reduced gas transport [68]. In forest soils, Smith, Dobbie [69] found that CH₄ oxidation rates were greater in coarse-textured soils with well-developed soil structure and permeable surface organic layer. Tate, Saggar [62] also found that afforestation of pastures increased CH₄ oxidation and concluded that this increase was related to the higher proportion of macropores in soils growing pine trees. Tree roots help increase CH₄ uptake in soils by changing soil texture, increasing porosity and therefore increasing gas diffusion [70].

Oxygen availability, eh and pH

Gradients of dissolved oxygen concentration directly affect methanotrophs. Methane consumption occurs under oxic conditions. However, when oxygen is depleted (anoxic conditions, Eh + 750 - -220), microorganisms can utilise alternative electron acceptors (NO³⁻, NO₂, Fe(III), Mn(IV) and SO₄²⁻) for CH₄ oxidation [71]. Anoxic conditions establish when oxygen supply cannot meet oxygen demand. Oxygen supply in soils is predominantly controlled by soil water content and structure (as above). However, even in seemingly aerobic, well drained soils, the complex physical structure of soils results in an abundance of anoxic microsites/hotspots and associated metabolic gradients [72, 73]. For example, significant anoxic CH₄ oxidation has been identified at depths ranging from 0 to 60 cm in temperate forest soil profiles [73]. Methane oxidation occurring under anoxic conditions (microsites and spatial and temporal scales) is now recognised as important but ill-defined control on CH₄ cycling in soils [72, 73].

Although temperate forest soils are generally considered CH₄ sinks, CH₄ production could still be occurring within hotspots in the soil. Biological CH₄ production occurs under anaerobic, Eh -240 conditions but methanogens have also been found to be abundant and active in oxic soils and broadly represented in soil metagenome datasets [74]. This highlights the need for high resolution studies across sites to capture these hotspots [75] when measuring site CH₄ flux. Microbial studies are needed to better understand anaerobic CH₄ uptake and aerobic CH₄ production.

In forests, most soils exhibit soil pH levels 6 but range from pH > 3.5 to pH < 8.0 [46]. Although pH generally has a first-order influence on many soil biogeochemical cycles, its influence on CH₄ cycling is relatively minor. Methanotrophic microorganisms typically adapt to local forest soil pH conditions [56] and, as such, rates are relatively pH insensitive.

Temperature and season

The distinct seasonality of temperate forests is a prominent feature that influences CH₄ (discussed above). Seasonal variation in CH₄ uptake is driven by changing temperature and soil moisture throughout the year [22]. Despite this the effect of temperature alone is generally small within the ranges associated with temperate forests [69]. Methane oxidation has been found to vary little over a wide range of temperatures (1–30 °C) [18]. The optimum temperature for CH₄ oxidation is between 25 and 30 °C [76, 77]. However, at temperatures outside this range (e.g. -5 °C [78] and 40 °C [77]) CH₄ oxidation was reduced/or completely inhibited.

Soil fertility and soil organic carbon

Temperate forests are found on soils that range in fertility. High fertility soil can exhibit greater CH₄ uptake than low fertility soil [78, 79]. The processes responsible for this effect are unknown, but it may be that the greater availability of nutrients reduces restrictions on methanotroph maintenance and growth relative to low fertility soils.

Soil organic carbon in forest soils has been found to have a contrasting effect on CH₄ flux. Global studies incorporating all forest biomes by Feng, Guo [25] and Gatica, Fernández [80] found that increasing soil organic carbon reduced CH₄ uptake, or even caused greater CH₄ emissions. This was thought to be due to carbon stimulating methanogenesis [81]. In contrast, Borken, Xu [79] found no correlation between soil organic carbon and CH₄ uptake and Lee, Oh [82] found that forest soils with greater levels of soil organic carbon (0 to 20 g kg^− 1 vs. 60 to 100 g kg^− 1) had greater CH₄ uptake (study included all forest biomes). Along with soil carbon content, soil C:N ratio in forest soils have also been found to affect CH₄ flux, with greater C:N ratios associated with reduced CH₄ uptake [83, 84]. More research is required exploring the influence of soil organic carbon and C:N ratios on CH₄ flux as the mechanisms behind these variable responses are largely unknown.

Other soil nutrients

The availability of ferric ion (Fe³⁺), sulphate ion (SO₄²⁻) and Manganese (Mn⁴⁺) determine whether CH₄ can be oxidised under oxygen limiting conditions (see section on oxygen availability). This relationship was thought to be more relevant to tropical soils/rice paddies [84] but is now considered to be relevant in all soils due to their heterologous nature [73, 85]. It is important to note that the availability of these alternative electron acceptors depends on the presence of other microorganisms that can reoxidise/replenish/recycle these nutrients. Production and release of reactive intermediates by methanotrophs are strongly influenced by metals (copper, iron manganese, lanthanides) which in turn influence the dynamic fluxes of nitrous oxide and CH₄ to the atmosphere [86, 87].

Forest management and silviculture

Use of nitrogen and phosphorus fertilisers

The addition of N fertiliser has been found to inhibit CH₄ oxidation in temperate forest soils [88–92]. Annual applications of N fertiliser of 150 kg N ha^− 1 yr^− 1 as ammonium nitrate (NH₄NO₃) have been reported to decrease CH₄ uptake by soils relative to the control by 64% [78] to 86% [93]. Lower rates of 50 kg NH₄NO₃-N ha^− 1yr^− 1 decreased annual CH₄ uptake by 16% compared with unfertilised soils [78]. A meta-analysis by Xia, Du [94] found that N addition significantly decreased soil CH₄ uptake by 39% in temperate forests. However, this was dependent on N level with CH₄ uptake increasing at N additions below 27 kg N ha^− 1 yr^− 1.

There are also studies that show N input from fertilisation can either stimulate forest soil CH₄ uptake or have no effect on uptake at all [89]. A single application of N fertiliser to temperate forest soils has been shown to cause an initial decrease in CH₄ uptake compared with unfertilised soils but after 2 months [95] – 1 year [96] there was no difference in CH₄ uptake between the soils that received fertiliser and the control soils that did not. After an initial suppression in CH₄ uptake following N fertiliser addition some studies have shown an increase in CH₄ uptake [96–98].

Mechanisms for inhibition/stimulation of CH₄ uptake in response to N fertiliser addition are poorly understood. There are multiple proposed mechanisms for the inhibition of CH₄ uptake as a result on N fertiliser addition. The inhibiting effect of NH₄NO₃fertiliser could be due to ammonia oxidisers competing with methanotrophs [99] or due to competition between NH₄⁺ and CH₄ at the binding site of the catalysing enzyme CH₄ monooxygenase in the first step of CH₄ oxidation pathway [76]. The intermediates and end products of methanotrophic ammonia oxidation, i.e. hydroxylamine and nitrate, can be toxic to methanotrophic bacteria and lead to inhibition of CH₄ consumption [89].

The response may also be dependent on soil N status. Soil CH₄ flux is less sensitive to N addition in N-limited ecosystems [92] with less reduction or even an increase in CH₄ uptake [95, 97, 98]. One possible explanation is that the addition of N corrects a N limitation that was reducing methanotroph activity [98] or the addition of N corrects other soil properties such as C:N ratio that makes conditions more favourable for methanotrophs [97]. In contrast, N addition in forest ecosystems could also lead to increased litter accumulation on soil surface reducing the diffusion of CH₄ and oxygen into the soil reducing CH₄ uptake [100].

It has also been suggested that the recovery of CH₄ uptake after an initial period of decline following N fertilisation could be due to a shift in the methanotroph community to one that is tolerant to excess NH⁴⁺ [93]. For example, Mohanty, Bodelier [101] found that N addition had little effect on CH₄ oxidation in soils where type I methanotrophs predominated but could markedly affect oxidation in type II-dominated soils. Others have reported that increased N availability does not impact methanotrophic community structure but did reduce the abundance and activity of CH₄ consuming methanotrophs [102]. There are many different members of the microbial community contributing to cycling of N and CH₄ in soils [103]. Complex interactions and multiple factors mean that mixed results have been obtained from use of N fertilisers.

Type of N fertiliser may also influence the response of CH₄ uptake to N fertiliser addition. Urea has been found to reduce CH₄ uptake by 5 to 20 times compared with unfertilised soils [91]. However, urea has been found to cause the least suppression of CH₄ uptake in upland ecosystems (not specifically forests) compared with other forms of N [92].

Little is known about the effect of P addition on CH₄ oxidation, particularly in temperate forest soils. The few studies that have been done show contrasting results. Winsborough, Thomas [104] found that the addition of P in a high N soil caused a decrease in CH₄ uptake whereas Borken, Xu [79] found that total P content in the soil was not correlated with CH₄ uptake and Burke, Smemo [105] showed no effect of P on the methanotroph community. The review by Veraart, Steenbergh [106] concluded that a better mechanistic understanding is needed about the effect of P on methanotroph community structure and on CH₄ oxidation, not just in planted forests but in soils generally.

Globally, fertiliser use in plantation forests is low, with many receiving no fertiliser at all [107]. In New Zealand for example, the average application rate of fertiliser (N + P) across all planted forests in 2017, including those with no applications) was 8 kg ha^− 1 [108]. Fertiliser application is infrequent. Applications in the first year are common but after this application frequency ranges from once per year to not at all. For N, the per application rate is ∼ 200 kg N ha^− 1, applied in the form of urea, ammonium sulphate and di-ammonium sulphate [108]. For P, the per application rate is ∼ 75 kg P ha^− 1 for phosphorus [108].

When N fertiliser is applied to plantation forests the rate is generally at rates ∼ 200 kg N ha^− 1 yr^− 1 therefore it is likely that N fertiliser use in planted forests will reduce CH₄ uptake. However, if N fertilisation is infrequent the response may not be permanent and CH₄ uptake rates could recover to pre-fertilisation levels within 1 year. In N limited environments, the addition of N could even increase CH₄ uptake. However, more research is required to better understand the mechanisms behind these processes before site specific response predictions can be made.

Stand age

Soil CH₄ uptake increases with stand age [109, 110]. Smith, Dobbie [69] found that after afforestation CH₄ uptake increased over 100 years. This could be due to decreased soil moisture content resulting from increased tree transpiration and rainfall interception by canopy and litter cover. Using a 120-year afforestation chronosequence, Hiltbrunner, Zimmermann [111] found that soil CH₄ oxidation increased with stand age because atmospheric CH₄ diffusion into the soil was enhanced as soil water content decreased.

Stand density

Thinning is the management practice of reducing the stand density which leads to increased tree size and improved timber quality. The timing and intensity of thinning varies greatly across forest types and management objectives globally. In New Zealand, radiata pine is typically planted at about 1,000 to 1,250 stems per hectare and pruned or thinned to 200–500 stems per ha. However, stand density has not been found to affect soil CH₄ uptake. De Bernardi, Priano [112] found there was no difference in CH₄ uptake between a Pinus radiata stand in Argentina with a density of 727 trees ha^− 1 and one with 977 trees ha^− 1 despite differences in canopy cover and ground cover (the lower stocking rate had a grass understory). Forest thinning has also been found to have no significant effect on soil CH₄ uptake in ponderosa pine forests [113]. Despite these results, thinning does affect soil moisture due to decreased canopy cover and less rainfall interception and therefore it would be expected that thinning would have an impact on CH₄ uptake. There are limited studies on the effect of stand density and thinning in temperate planted forests on CH₄ flux, and this is an area where more research is required.

Tree species

A range of trees are grown in temperate forests. These include deciduous, coniferous and broadleaved evergreen trees. Tree species affects soil CH₄ uptake in planted temperate forests [25, 114]. Coniferous forests have been found to have lower CH₄ uptake rates compared with deciduous beech trees [79, 115] and mixed stands with both broadleaved and coniferous trees [116]. In contrast, Christiansen and Gundersen [117] reported that CH₄ uptake did not differ between deciduous (oak (Quercus robur)) and coniferous (Norway spruce (Picea L. Karst)) tree stands. Modelling of global CH₄ flux by Feng, Guo [25] estimated soil CH₄ uptake rate of coniferous forests to be much greater than broadleaf forests and similar to that of mixed broadleaf/conifer forest however, this comparison included tropical and boreal forests as well as temperate. How tree species effect CH₄ uptake is not well understood. Methanotrophic communities in deciduous forest soils differ from those of mixed and coniferous forest soils suggesting that tree species might affect the methanotrophic community structure and activity [56]. Borken, Xu [79] speculated that organic compounds produced by the spruce and pine trees may have diminished the activity or population of methanotrophs in the soil. Maurer, Kolb [44] found Norway Spruce (Picea abies) released monoterpenes into the soil at rates high enough to explain reduced CH₄ uptake. It could also be due to the different composition of the litter under the different trees or the greater atmospheric deposition of N in coniferous forests influencing soil factors such as pH, carbon content and C:N ratio and therefore influencing methanotroph growth and activity [115, 116]. Further studies investigating the mechanisms behind the effect of tree type on soil-atmosphere CH₄ exchange are needed.

Compaction

The use of heavy machinery during the harvesting of temperate planted forests leads to severe compaction of the soil. This results in a change in soil structure, increasing soil bulk density, decreasing soil porosity and air permeability, and increasing water retention and potential for waterlogging [118]. This affects gas diffusivity and the microbial community in the soil, reducing CH₄ uptake or even facilitating CH₄ emissions in planted forests. In three beech stands in Germany, compaction reduced CH₄ uptake by 90% in wheel tracks [119] and in a mixed beech-spruce forest in Switzerland compacted soil vehicle tracks was found to be a CH₄ source, also supporting greater abundances of methanogenic archaea [120]. The severity of compaction depends on factors such as harvesting equipment, operation conditions and site characteristics [118]. This is likely to result in site specific impacts of compaction on CH₄ flux. Reducing heavy machinery within forests is a key factor that could be used to minimize the impact at harvesting on CH₄ uptake, and this should be reflected in research priorities. An example is the development of suitable information to allow the incorporation of site conditions and characteristics into harvest plans to minimise negative impacts on CH₄ flux.

Stem methane fluxes

Tree stems are most commonly a source of CH₄ [43] although there have also been reports of CH₄ uptake from tree stems [16, 121, 122]. Tree stem CH₄ flux has been extensively reviewed by Covey and Megonigal [43]. Table 1 provides a summary of stem CH₄ emissions in dry, temperate, upland forests (adapted from Covey and Megonigal [43]), highlighting the relatively small contribution of this CH₄ source to atmospheric pools. Tree emissions are commonly measured per stem. Studies that have upscaled per stem measurements from upland temperate forests have estimates of stem emissions ranging from 0.2628 g CH₄ ha^− 1 yr^− 1 to 288 g CH₄ ha^− 1 yr^− 1 (references in Covey and Megonigal [43]). Stem emissions have been estimated to offset soil CH₄ uptake by roughly 3.5% by Warner, Villarreal [123], 1–6% by Pitz and Megonigal [121] and Wang, Gu [124] estimated a much higher 63% from a temperate poplar forest in Beijing, China. In contrast, a recent study by Gauci, Pangala [122] estimated woody surfaces in upland temperate forests to be a small CH₄ sink with a total annual CH₄ uptake of 1.96 Tg CH₄ yr^− 1. It is important to note that calculating stand level CH₄ flux from tree stem measurements taken from single stems is difficult due to the variability of stem emissions/uptake within a forest (discussed in section below). Currently this variability is not routinely considered in scaling up calculations leading to overestimation of CH₄ emissions from tree stems at a global, stand or even tree level.

Table 1.

Methane stem emissions from trees in upland temperate forests (adapted from Covey and Megonigal, 2019 [43]).

Study	Site Description	Source	Rate	Units	Notes
[133]	Upland	Stem-based	0.009 (± 0.006)	nmol m− 2 s− 1	Low stem surface emissions from young plantation oak trees
[129]	Upland and Wetland	Stem/soil-based	1.19–9.84	nmol m− 2 s− 1	Fluxes vary along a gradient with highest in wetland areas
[123]	Upland	Stem-based	0.11 (± 0.21)	nmol m− 2 s− 1	Tree species were mainly tulip, maple, beach, and birch
[121]	Upland	Stem/soil-based	0.44 (± 0.24)	nmol m − 2 s− 1	Looked at 17 trees, mostly beach, tulip, hickory. Stem emissions were 1–6% of soil sink
[134]	Upland	Stem-based	0.52 (± 0.92)	nmol m− 2 s− 1	Fagus sylvatica forests
[128]	Upland	Stem-based	0.003 (± 0.001)	nmol m− 2 s− 1	Tree species included pine, hemlock, oak, birch and maple. They found widespread microbial production in living stems producing CH4

The processes that regulate CH₄ flux from tree stems are not well understood. The emission of CH₄ from tree stems is thought to be driven by two possible mechanisms. Firstly, methanogenic microbes can colonise the stems of trees, resulting in CH₄ production and release from the tree [16, 125]. There is also some evidence that methanotrophs are also present in tree stems [16, 122, 126] suggesting that the direction of CH₄ flux of tree stems is driven by the balance of co-occurring methanogenesis and methanotrophy.

In addition, CH₄ produced in the soil can be absorbed by the roots and transported to the stem of the tree where it diffuses into the atmosphere [123, 127, 128]. This second pathway is common in wetland and tropical environments in which the anaerobic nature of the soil promotes CH₄ production, creating a large pool of CH₄ that can be absorbed by roots. Consequently, studies which report CH₄ fluxes in these wetter environments often provide greater CH₄ stem emission values than what would be produced in a temperate climate [129]. This trend is supported by Warner, Villarreal [123] who observed stem CH₄ fluxes 1–2 orders of magnitude less in temperate forests than those reported for wetland systems.

Stem emissions and uptake in temperate forests are extremely variable. This variation in CH₄ fluxes arises from multiple factors including tree species, age, tissue type, site characteristics and environmental conditions [43, 130]. Stem emissions have been found to vary between and within tree species. Epron, Mochidome [127] found that coniferous species emitted almost no CH₄ whereas four broadleaved species had high intraspecific variability (0-3.7 nmol m^− 2 s^− 1). Warner, Villarreal [123] also found significant differences in CH₄ fluxes between 6 different tree species. Gauci, Pangala [122] measured mean CH₄ flux of sycamore (Acer pseudoplatanus) and ash (Fraxinus excelsior) trees and found that on average sycamore trees were emitting CH₄ and ash trees were up taking CH₄. There was also a wide flux range for each tree species (sycamore: -50.41–27.69 µg CH₄ m^− 2 h^− 1, oak: -39.79–27.08 µg CH₄ m^− 2 h^− 1) showing within species variability. Even on the same tree CH₄ emissions vary by tissue type with CH₄ emissions decreasing from the main stem > branches > leaves [127] and differing between sapwood and hardwood [124]. With increasing height up the stem CH₄ emissions have been shown to decrease [121] and Gauci, Pangala [122] found that measurements taken above 1.3 m on ash and sycamore trees in a temperate forest, showed CH₄ uptake whereas measurements below this showed CH₄ emissions. Some studies have shown temporal (seasonal and/or diurnal) variation [121, 131], while others have not [123]. Size and/or age of the tree has also been shown to influence stem CH₄ emissions [129, 132].

To improve estimates of stem CH₄ flux at a tree, stand, regional or global level, research should focus on characterisation of the microorganisms and biochemical pathways associated with CH₄ production and investigation into the biogeochemical processes involved in CH₄ production and transport through stems [130]. There also needs to be long term studies with frequent measurements (i.e., monthly, daily, hourly) to capture temporal variability along with studies that take measurements across individuals, tree species and locations to understand other sources of variability.

Foliage methane fluxes

The potential for tree foliage to play a role in CH₄ fluxes is uncertain. The few studies that have measured tree foliage in situ were in boreal forests and suggest that foliage may be involved in non-trivial fluxes but no consistent patterns have been reported with studies showing positive, negative and nondetectable fluxes [16, 135–137]. Recently, a study by Gorgolewski, Caspersen [138], in an upland temperate forest in Ontario, found that tree foliage was consistently a CH₄ sink. This is the first in situ observation of consistent CH₄ uptake across a range of tree species and sizes in a temperate forest. The foliage consumed CH₄ at a rate of − 0.54 nmol m^− 2 s^− 1 in direct sunlight, which was approximately 62% of the soil consumption rate and represented 38% of nett daytime ecosystem CH₄ consumption. The mechanism for potential foliar CH₄ uptake is hypothesised to be consumption by endophytic methanotrophic bacteria. Photosynthetically active radiation was also found to influence foliar CH₄ uptake [135, 138].

Tree foliage has also been found to be a CH₄ source. Gorgolewski, Caspersen [138] also measured CH₄ flux at a lowland temperate forest site where soil was a CH₄ source. They found that foliage was a CH₄ source (6.06 nmol m^-2 s^-1 ± 2.47 SE in direct sunlight) with an emission rate about 8% that of the soil. A possible mechanism for this is that CH₄ produced by methanogens in the soil or in the tree stem was transported to foliage through the transpiration stream [137]. It is unlikely methanogens on the tree foliage played a significant role in the nett CH₄ flux as they need anoxic conditions to survive [16]. Although it should be noted that there are microbes that have been shown to produce CH₄ in oxic environments [15, 16]. Another possible mechanism is aerobic CH₄ production by plants [139]. This process is induced by cutting injuries, increasing temperature, ultraviolet radiation and reactive oxygen species [140]. However, the possibility of aerobic CH₄ emissions by plants has been a contentious topic in the literature [139–141] because studies have mostly been lab studies not representative of natural conditions [140]. Tree foliage fluxes appear to be important and could be a potential sink in temperate forests, but further research is required to confirm fluxes and mechanisms.

Litter and deadwood methane fluxes

Temperate forests can hold significant masses of litter and deadwood both as a consequence of natural events such as leaf fall, shedding of branches (self-pruning/autonomic abscission), windthrow, and management such as thinning and harvesting. In temperate forests, 13% carbon stock is stored in deadwood and litter [142]. Decomposition of organic matter can result in the release of CH₄, particularly in waterlogged or otherwise anaerobic environments. Despite this, the few studies that have explored CH₄ emissions from litter and deadwood indicate emissions are negligible, or less than that from other plant types.

For leaf litter, Schipper, Harfoot [143] reported that CH₄ release from decaying fresh and senescent radiata pine needles was detectable when incubated under anaerobic conditions favouring CH₄ production, however the extent of CH₄ production was substantially less than that associated with decomposition of biomass from other plants. In this regard, the lability of the plant-associated carbon was hypothesised as being more important than amount of carbon added per se. Gritsch, Egger [144] explored CH₄ emissions from coniferous and deciduous leaf litter under a range of conditions and determined that emissions were essentially negligible. Peichl, Brodeur [145] reported a similar outcome from a study conducted in temperate pine forests, determining the presence of tree litter had no impact on CH₄ emissions over multiple sampling dates. The presence of litter is comparable to a ‘blanket’ above the soil providing a buffer to temperature and moisture changes, as well as providing a range of oxic to anoxic conditions, so although litter has been shown to not emit or uptake CH₄ directly, litter on the forest floor can indirectly modify CH₄ flux of forest soils. This has been reviewed by Walkiewicz, Rafalska [146]. In summary, the litter layer may influence soil CH₄ uptake in opposing ways by: (1) decreasing uptake by acting as a physical barrier to gas diffusion and reduced aeration due to faster litter decomposition in wet conditions, (2) increasing uptake through maintenance of soil gas diffusivity under wetter/high rainfall condition, (3) influencing the capability of the soil for oxidising CH₄, (4) providing source of nutrients for methanotrophs, (5) improving formation of macro-aggregates, which facilitates CH₄ transport for methanotrophs and (6) the production monoterpenes during needle decomposition that can act to reduce CH₄ uptake.

For deadwood, a study incorporating a specific focus on CH₄ dynamics in coarse woody debris in an upland temperate North American forest found that fresh woody debris could produce small amounts of CH₄, whereas for more decayed debris variable outcomes were observed, ranging from a small amount of CH₄ emissions through to substantial CH₄ uptake [123]. Gorgolewski [147] also found course woody debris was a CH₄ source in early stages of decay and a sink when it was more decayed in a temperate forest in Ontario. CH₄ concentrations within deadwood were also found to be greatest in the least decayed wood [148]. In a temperate deciduous forest in the Upper Midwest USA a range of CH₄ fluxes from woody debris were also measured (-3.73–22.8 mg CH₄ kg downed woody debris^− 1 s^− 1) [149] and a recent study by Kipping, Gossner [150] also found that deadwood in early stages of decay was a source of CH₄ with tree species as an important driver. As deadwood decomposes, it becomes closer in physical and chemical properties to soil, until it eventually becomes organic matter in the soil. Therefore, it seems reasonable that CH₄ flux of deadwood would become more similar to that of soil as it decomposes. In contrast, Lagomarsino, De Meo [151] found that CH₄ emissions increased as black pine deadwood decayed. Deadwood also influences soil CH₄ flux. Perreault, Forrester [149] observed greater CH₄ uptake in the soil near highly decayed woody debris, possibly from increased labile carbon in the soil from log decomposition.

The proposed mechanism for CH₄ emissions from deadwood is through methanogenic bacteria and archaea that are active in the anaerobic conditions inside decaying wood, or that the tree stems, once cut, serve as a conduit to release CH₄ stored from the soil, in the stem [152]. Methane emissions from decomposing wood under aerobic conditions have also been shown by wood-decomposing fungi in association with archaea [153]. The mechanism for CH₄ uptake in deadwood is through methanotrophic bacteria inhabiting the material; Mäkipää, Leppänen [154] found the highest number of methanotrophs in Norway spruce logs in late stages of decay.

What determines if deadwood is a CH₄ source, or a sink is not well understood. It appears to be a combination of tree species [150], microbial community [154] and level of decay [123]. However, there is still considerable unexplained variability in measured deadwood CH₄ flux that needs to be further investigated.

For future research on litter/deadwood fluxes, field-based measurements are critical as they can account for how varying environmental moisture and temperature influence CH₄ production, moving away from assumptions related to steady state conditions and reaction kinetics. This will be challenging given variability in ecosystems and difficulty partitioning CH₄ from litter and deadwood from other sources, particularly given the extent in overlap in δ13-C-CH₄ values among plants and litter decomposers [155]. However, bottom-up modelling of litter and deadwood CH₄ emissions based on laboratory incubation studies cannot account for the variability in the field and leads to overestimation of CH₄ fluxes. Sources of variability that need to be considered in temperate planted forests include tree species, age and management practices. These factors influence the amount and composition of deadwood/litter entering the system. The more complex the forest the more complex the inputs of litter and deadwood. There is also spatial and temporal variability which will also influence CH₄ flux.

Although CH₄ uptake/emissions from litter and deadwood are small, given the size of deadwood/litter pool in forests, and the range of types of plant carbon entering the forest system at different times/seasons etc., and the scale of the temperate forests globally, these fluxes should not be overlooked in forest CH₄ budgets.

The atmospheric chemistry of methane in relation to the VOC/hydroxyl radicals

Within the atmosphere CH₄ is predominantly degraded through oxidisation by OH [156, 157]. This reaction is limited by the abundance of OH, which are depleted through various reactions, including reactions with VOCs. It is therefore possible that an increase in VOC emissions could lead to a reduction in OH and a subsequent increase in the lifetime of CH₄. Due to this VOCs have been listed in IPCC reports as species affecting the CH₄ sink via OH oxidation [158]. Volatile organic compounds are emitted from a range of temperate tree species. Aydin, Yaman [159] found that both conifers and broad-leaved tree species emitted VOCs, with isoprene the predominant compound for broad-leaved tree species and monoterpenes the predominant compound emitted by conifers. Volatile organic compound emissions from temperate forests could therefore be influencing OH CH₄ sinks. However, it is uncertain how large this effect is. In New Zealand, monoterpene emissions from planted pine forests could be influencing OH CH₄ sinks but VOC emissions from pine trees have not previously been measured in New Zealand.

There is some evidence to suggest this VOC/CH₄ effect could be occurring. VOC reactions have been shown to act as an OH sink in boreal forests, with monoterpenes having the greatest reactivity [160, 161] and declining global OH levels have been shown to be partially responsible for past increases in CH₄ levels [156]. However, the magnitude of this effect alongside increasing atmospheric CH₄ is uncertain [7].

One aspect that is unclear is how much this effect is happening over temperate forests. Since monoterpenes and isoprene, the most common VOCs emitted by forests, have a lifetime of minutes to hours, and OH has a lifetime around 1 s [162], any effect will be occurring locally. Currently, there is little measurement of CH₄ OH oxidation over temperate forests, and studies tend to use global modelling [156, 163, 164].

It is also unclear if the VOC oxidation occurs at a significant magnitude to affect CH₄ oxidation. Many aspects effect the abundance of VOCs and OH in the atmosphere. Biogenic VOC emissions from forests can vary with changing environmental conditions [165] and are oxidised by O₃ as well as OH [162] so the abundance of this could be important. OH radicals are produced via photolysis of O₃ and concentrations are temporally and spatially variable with local concentrations depending on highly variable factors such as temperature, cloud cover and surface albedo [157, 166]. OH is highly reactive and is depleted through multiple atmospheric reactions. Concentrations of carbon monoxide and NO_x have been shown to be important for changes in the OH oxidation of CH₄ as well as concentrations of VOCs [158].

In NO_x enriched environments an increase in OH sinks such as VOCs make less difference to overall OH levels, while in NO_x depleted environments an increase in OH sinks is thought to be more important [167]. There is still much uncertainty around this theory, however, due to the complexities of the environment, and measurements do not always match what is expected. While global atmospheric models show a strong OH depletion caused by VOC emissions, Lelieveld, Butler [168] measured high OH concentrations over the pristine Amazon forest. They suggest that VOC oxidation efficiently recycles OH in low NO_x environments, and OH depletion over forests is not as strong as previously thought.

Overall OH is an important CH₄ sink and OH depletion through VOC oxidation could be significant for increasing CH₄ lifetimes. The effect of this should be considered in relation to the high VOC emissions from forestry. However, more information is needed to determine the magnitude of the impact of VOCs on CH₄ lifetimes, particularly actual measurements taken in or above planted forests.

Conclusions

Temperate forests and methane fluxes

Features unique to temperate forests that influence CH₄ cycling are strong seasonal cycles of temperature and moisture, and the extent of human interaction. These unique elements must be considered when accounting for CH₄ flux in temperate forests.

Methane cycling in temperate forests is a complex process with different components acting as either sinks or sources. Figure 3 summarises the current state of knowledge of CH₄ cycling in temperate forests based on the findings of this review. The magnitude and direction of CH₄ flux for each component is variable and is influenced by many factors. Some components have been studied more than others (e.g. soils, ∼ 46% of references in this review were focused on soils) but overall, more research is required into the processes that drive CH₄ cycling, the sources of variability (spatial and temporal) and the factors that affect CH₄ cycling for all components of temperate forests. The specific findings for each component and recommendations for further research to accurately account for CH₄ flux in temperate forests are summarised in the bullet point list below.

Fig. 3.

Summary of the main sinks and sources of CH₄ in temperate forests. Arrow colour and direction indicates potential of each compartment to act as a source (red) or sink (green). Size of arrows indicates the estimated size of the source or sink. The text below the forest compartment labels indicate confidence in this assessment based on the literature.

Soils

• Temperate forest soils are significant CH₄ sinks (Fig. 3 (Soil)). However, there are many factors that affect the magnitude and direction of CH₄ flux in soil. The literature often showed contrasting responses to different factors in different environments. This highlights the importance of local/site specific measurements of CH₄ flux rather than trying to extrapolate from global data.

• The spatial and temporal variability of soils and climate within a temperate forest leads to changes in CH₄ fluxes with hotspots occurring where conditions may favour CH₄ emissions. This shows the need for high resolution of sampling within a forest to get a true measure of soil CH₄ flux.

• Methane flux in temperate forest soil is far more complex biogeochemically than previously thought. For example, CH₄ production and consumption has been measured under conditions not previously known to support these processes. There is a need for more research into the mechanisms and microbial community responsible for the changes to CH₄ flux.

• In planted temperate forests there is an opportunity to influence and encourage CH₄ uptake in forest soils through forest management, for example by reducing fertiliser use or selecting tree species based on the site. However, before these recommendations can be made more site-specific research needs to be done.

Tree stems

• Tree stems in temperate forests are a small source of CH₄ (Fig. 3 (Stem)).

• The processes and microbial communities that regulate CH₄ flux from trees are the least understood component of the global CH₄ cycle.

• There is a large amount of variability in stem CH₄ flux arising from multiple factors including tree species, age, tissue type, site characteristics and environmental conditions. This variation is not well characterized nor is it included when scaling up per stem emissions to a stand or global level. As a result, CH₄ emissions estimates at a global, stand or even tree level, calculated from per stem measurements, are likely overestimates.

• Understanding the processes, microbial community and sources of variation involved in stem CH₄ flux is needed to improve estimates of stem emissions from temperate forests for CH₄ budgeting.

Foliage

• The potential for tree foliage to play a role in CH₄ fluxes in temperate forests is uncertain. Studies have shown conflicting evidence of CH₄ fluxes from the foliage of trees in temperate forests (Fig. 3 (Foliage)). Further research is needed to confirm fluxes and mechanisms.

Litter/deadwood

• Emissions from litter are negligible (Fig. 3 (Leaf litter)). However, the litter layer can influence soil CH₄ uptake. Emissions from deadwood are small but variable depending on factors such as tree species, level of decay and microbial community present (Fig. 3 (Deadwood)).

• Most studies on litter/deadwood CH₄ flux are laboratory based. These studies do not take into account variability in the field and leads to overestimation of CH₄ fluxes. Forest based studies are needed for more accurate litter/deadwood CH₄ flux measurements for CH₄ budgeting.

Volatile organic compounds

• OH is an important CH₄ sink and OH depletion through VOC oxidation could be significant for increasing CH₄ lifetimes (Fig. 3 (Volatile organic compounds)). As forests emit high levels of VOCs it is possible that temperate planted forests are contributing to the depletion of OH and therefore increasing CH₄ lifetime in the atmosphere. However, measurements taken in or above planted forests are needed to determine the magnitude in which forests impact CH₄ in the atmosphere and if it is significant or not.

Recommendations to improve methane budgets for temperate forests

• Further research into mechanisms and microbial community involved in temperate forest CH₄ cycling is needed for all sources and sinks in temperate forests.

• There is huge variability in CH₄ flux from all components of the forest that is not well understood, nor is this variability taken into account in global CH₄ budgets and models. This means most estimates of CH₄ fluxes are likely inaccurate. High resolution (spatially and temporally) of sampling in forests (not laboratories) is needed to better understand variability.

• Methane budgets and modelling of CH₄ flux need to incorporate all sources and sinks CH₄ in temperate forests (i.e. not just soils). All sources of variability also need to be considered to ensure the most accurate estimates of CH₄.

• A roadmap for sampling and scaling should be developed similar to the one developed for biological nitrogen fixation [169]. This will ensure that global research efforts into CH₄ cycling in temperate planted forests is focused, consistent, robust and fit for purpose in order to accurately quantify the contribution of temperate planted forests to the global CH₄ cycle. Being able to accurately quantify CH₄ flux in temperate forests will aid in mitigating CH₄ emissions from other sources and allow better global greenhouse gas emissions budgeting.

• Once there is a better understanding of CH₄ cycling in temperate forests the next steps will be to investigate the effects of climate change on these processes and how it will influence global CH₄ budgets.

Abbreviations

BP

Before present

CH₄

Methane

C:N

Carbon to nitrogen

CO₂

Carbon dioxide

Eh

Redox potential

GHG

Greenhouse gas

GWP

Global warming potential

IPCC

Intergovernmental Panel on Climate Change

N

Nitrogen

NO_x

Nitrogen oxides

O₃

Ozone

OH

Hydroxyl radicals

P

Phosphorus

VOCs

Volatile organic compounds

Author contributions

K.W and S.A.W wrote the majority of the manuscript together. S.J.S and C.A contributed to the first draft of the manuscript. N.M.R contributed to the writing of the manuscript. All authors contributed to the literature search. All authors read, edited, and approved the final manuscript.

Funding

This work was funded by the New Zealand Ministry of Primary Industries through the Greenhouse gas inventory research fund.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics and Consent to participate

Not applicable.

Competing interests

The authors declare no competing interests.