How Can Litter Modify the Fluxes of CO 2 and CH 4 from Forest Soils? A Mini-Review

: Forests contribute strongly to global carbon (C) sequestration and the exchange of greenhouse gases (GHG) between the soil and the atmosphere. Whilst the microbial activity of forest soils is a major determinant of net GHG exchange, this may be modiﬁed by the presence of litter through a range of mechanisms. Litter may act as a physical barrier modifying gas exchange, water movement/retention and temperature/irradiance ﬂuctuations; provide a source of nutrients for microbes; enhance any priming effects, and facilitate macro-aggregate formation. Moreover, any effects are inﬂuenced by litter quality and regulated by tree species, climatic conditions (rainfall, temperature), and forest management (clear-cutting, fertilization, extensive deforestation). Based on climate change projections, the importance of the litter layer is likely to increase due to an litter increase and changes in quality. Future studies will therefore have to take into account the effects of litter on soil CO 2 and CH 4 ﬂuxes for various types of forests globally, including the impact of climate change, insect infestation, and shifts in tree species composition, as well as a better understanding of its role in monoterpene production, which requires the integration of microbiological studies conducted on soils in different climatic zones.


Introduction
Forest ecosystems are a critical component of the global carbon (C) budget through their ability to sequester and retain large amounts of CO 2 [1]. An elucidation of the functioning of forest ecosystems, including their contribution to greenhouse gas (GHG) exchange, is crucial for the development of adaptation and mitigation strategies. In Europe, forests cover about 35% of the total land area, corresponding to 227 million ha with a decreasing share of coniferous (46%), broadleaved (37%), and mixed (17%) tree species [2].
Forest soils influence the GHG balance, with carbon dioxide (CO 2 ) and methane (CH 4 ) as key elements of the global C cycle. Of the three major GHGs the CO 2 flux is quantitatively often the most important and its concentration has increased by 1.9 ppm yr −1 in the last 10 years [3]. The biogenic sources of CO 2 efflux from soil are root respiration, rhizomicrobial respiration, priming-related effects, and basal respiration associated with the microbial decomposition of organic matter in soils [4]. Forest ecosystems also significantly contribute to the increase in CO 2 emission through forest fires, deforestation [5][6][7], and CO 2 release by soil microorganisms colonizing dead trees [7]. For many forests, soil CH 4 is another important GHG with a global warming potential (GWP) 28 times greater than that of CO 2 [8]. Soils may be a source or sink for CH 4 depending on the balance between CH 4 production (methanogenesis in an anaerobic environment) and CH 4 oxidation (methanotrophy in aerobic conditions) [9,10]. The process of methanotrophy has significant mitigating potential since methanotrophs can contribute to the reduction in atmospheric CH 4 (high affinity methanotrophy) on the one hand, but, on the other hand, can also

Carbon Dioxide (CO2)
There are two essential processes that contribute to the total CO2 efflux from the soil: production and transport [33]. The processes of autotrophic respiration (related to roots and rhizospheric microorganisms) and heterotrophic respiration (from non-rhizospheric microorganisms, bacteria and fungi, including the litter layer) are the soil sources of CO2 [34][35][36]. Respiration is very sensitive to climatic changes and can be influenced by various environmental factors, such as soil moisture and temperature, microbial communities, the type of vegetation, or soil surface litter [13,[36][37][38]. It is documented in many studies that aboveground litter inputs have a significant effect on soil respiration in various types of forests (see [13,[39][40][41][42]). Moreover, modifications of soil-based CO2 fluxes and decomposed litter are also important CO2 sources [32].
Different mechanisms for the effects of litter on soil CO2 fluxes are presented in the literature. They can generally be classified based on the following impacts (Table 1): (i) An alteration in the availability of substrates for soil microbes; (ii) modifications in soil microbial communities; (iii) a priming effect; (iv) the creation of a physical barrier that decreases gas exchange and/or water movement/water retention and acts as an insulating layer that modifies soil temperature.
However, none of these effects are mutually exclusive and a litter-related increase in C substrate availability, for instance, could occur in agreement with an alteration in the microbial communities of the soil and/or a lower soil temperature, as indicated in Table  1.

Carbon Dioxide (CO 2 )
There are two essential processes that contribute to the total CO 2 efflux from the soil: production and transport [33]. The processes of autotrophic respiration (related to roots and rhizospheric microorganisms) and heterotrophic respiration (from non-rhizospheric microorganisms, bacteria and fungi, including the litter layer) are the soil sources of CO 2 [34][35][36]. Respiration is very sensitive to climatic changes and can be influenced by various environmental factors, such as soil moisture and temperature, microbial communities, the type of vegetation, or soil surface litter [13,[36][37][38]. It is documented in many studies that aboveground litter inputs have a significant effect on soil respiration in various types of forests (see [13,[39][40][41][42]). Moreover, modifications of soil-based CO 2 fluxes and decomposed litter are also important CO 2 sources [32].
Different mechanisms for the effects of litter on soil CO 2 fluxes are presented in the literature. They can generally be classified based on the following impacts (Table 1): (i) An alteration in the availability of substrates for soil microbes; (ii) modifications in soil microbial communities; (iii) a priming effect; (iv) the creation of a physical barrier that decreases gas exchange and/or water movement/water retention and acts as an insulating layer that modifies soil temperature.

Enhancement of anaerobic conditions by litter Plantation
Pl. orientalis n/a n/a n/a n/a n/a 15. Qr. agrifolia n/a n/a n/a n/a n/a 19 180 n/a Gravelly loam n/a n/a n/a n/a n/a n/a n/a 12.5 n/a n/a Acidic n/a n/a No effect ** n/a Ar. rubrum, Qr. rubra n/a n/a n/a n/a n/a 8. Old-growth semidecidous tropical forest n/a n/a n/a >35 n/a n/a n/a >2000 n/a n/a n/a n/a Increased ** Flat plateau (the planalto) Pará, Northern Brazil (3 • 0 37 S, 54 • 34 53 W) [49] Priming effect Old-growth moist lowland tropical forest n/a n/a n/a n/a n/a n/a 27 2600 Oxisol n/a n/a n/a Increased by 20% ** n/a Gigante Peninsula, central Panama (9 • 06 N, 79 • 54 W) [50] Undisturbed old-growth forest Ts. heterophylla, Ps. menziesii n/a n/a n/a n/a n/a 8.

Priming effect
Undisturbed old-growth forest P. menziesii, T. heterophylla n/a n/a n/a n/a n/a 8. Ar. pseudoplatanus, Fr. excelsior n/a n/a n/a n/a n/a 10 714 Qt. petraeae-cerris community n/a n/a n/a n/a 2754 ± 206 kg C ha −1 yr −1 10   However, none of these effects are mutually exclusive and a litter-related increase in C substrate availability, for instance, could occur in agreement with an alteration in the microbial communities of the soil and/or a lower soil temperature, as indicated in Table 1.
The litter layer in different forest types can act as a reservoir of organic (e.g., sugars, phenolics, hydrocarbons and glycerides) and inorganic compounds (e.g., N, P, Ca) that can be used as substrates for soil microbes and plants [16,18,41]. Removal of the litter layer in an eastern Amazonian regrowth forest resulted in a significantly lower soil CO 2 efflux due to a decrease in substrate availability [43]. The litter layer which accumulated during the drier season may be a source of C and N for soil microbes through decay in the wetter season. Therefore, the decrease in CO 2 flux was caused by the withdrawal of fresh substrates for heterotrophic respiration associated with the litter decomposition processes [44]. The removal of litter also resulted in a reduction in the dissolved organic carbon (DOC) concentration, one of the easily mineralizable substrates for microorganisms resulting in a decrease in CO 2 emission from subtropical forest soil [3]. The CO 2 fluxes from the soils of an Acacia mangium deciduous forest were linked with the amount of litter and the total soil C content both in the dry and wet seasons. It is also known that litter removal can significantly reduce microbial biomass C (MBC), as well as microbial biomass nitrogen (MBN), in (sub-)tropical forests [15]. In turn, it is shown that the addition of litter increases soil microbial biomass in a coniferous forest in China [53]. On the other hand, the depletion of C substrates for microorganisms because of litter removal did not affect the microbial biomass in a sclerophyll forest in Chile, which was explained by topsoil resilience in this type of ecosystem where litter removal is a common practice due to its use in gardens [45].
The structure of soil microorganisms determines the efficiency of C use, since fungi are more effective in assimilating C from litter than bacteria because they can store more C than is necessary for their metabolic processes [54,55]. These microbes can also decompose some recalcitrant compounds like lignin, as well as cellulose and hemicellulose, contained in fallen leaves and/or needles with the use of fungal polymer-degrading enzymes [56,57]. Fungi are more resistant to decomposition processes than bacteria, as their cell walls contain more durable compounds (i.e., carbon polymers such as chitin and melatin) than bacterial cell walls (i.e., phospholipids and peptidoglycan) [55]. The combination of N [58,59] with an increased fungal biomass was found to decrease the soil respiration rate [13,60], since when the litter layer is N-deficient, decomposing microbes are less capable of utilizing the available C [61]. The vegetation type also influences the structure of microbe populations in the litter. Fungi are more tolerant to tannins than bacteria, and it is known that needle litter contains greater amounts of these compounds than leaf litter [62,63]. Fernández-Alonso et al. [64] reported that leaf litter was dominated by Gram-negative bacteria in contrast to needle litter, which was dominated by Gram-positive bacteria and saprophytic fungi, and also had a higher ratio of fungi to bacteria. The community of microorganisms inhabiting beech forest litter was dominated by fungi [56], and litter removal in subtropical pine forest resulted in an increased ratio of fungi to bacteria [13], while the addition of litter in a nutrient-deficient Pinus sylvestris var. mongolica plantation significantly reduced the ratio of Gram-positive to Gram-negative bacteria [53]. Based on the information presented above, it can be concluded that changes in microbial communities (e.g., the reduction in microbial biomass and activities) caused by litter manipulations are one of the main factors which decreases soil respiration in coniferous forests [13,39,41,65].
Higher CO 2 emissions related to the addition of litter may have been caused by the so-called priming effect, which resulted in an increased mineralization of fresh organic C residues present in the litter [66][67][68][69][70]. During the priming effect, the contribution of fresh organic matter to the soil increased, the decomposition processes were stimulated, and the older stored C was released from the soil as CO 2 [52,71]. This effect occurred when the activity of microorganisms or the community composition was changed due to the impact of litter manipulations on the decomposition processes associated with organic matter in soil [52,64]. Leaf litter chemistry strongly influenced the magnitude of the soil priming effect, which was positively influenced by the concentrations of several cations (Ca, K, and Mg) in the later stage of litter decomposition [71]. Although the CO 2 emissions were higher in the tropical forest, the addition of litter due to the priming effect had a greater impact on soil respiration in the temperate deciduous forest. In subtropical and tropical forests, priming effects could occur when nutrients were limited and soil microorganisms had a sufficient substrate (C) which was a source of energy [72].
The litter layer acted as a physical barrier for O 2 diffusion and, especially under wet conditions (e.g., after rainfall), could result in an anaerobic environment for GHG production [32,41]. Generally, both very low and very high water contents resulted in lower soil CO 2 emissions by limiting the substrate supply to microorganisms (under conditions which were too dry) and the reduction in O 2 (under conditions which were too wet) [73]. In a study including different initial soil moisture contents, it was shown that, initially, CO 2 emissions were higher in flooded soil compared with wet soil. In contrast, after 5 days of the experiment, CO 2 emissions were lower since flooding conditions restricted microbial growth and activity. However, at day 20, CO 2 emissions from the flooded soil, which had become wet again, increased and then decreased gradually. The higher CO 2 emissions were explained by enhanced activities of microbes due to the higher DOC contents in the flooded, as opposed to, wet soil [74].
Water fluctuations regulated CH 4 fluxes by afffecting soil microorganisms involved in CH 4 balance. CH 4 emissions from soil increased with increased moisture and flood conditions which were preferential for methanogenes producing CH 4 , while such conditions limited the process of methanotrophs oxidizing CH 4 under O 2 availability [10]. But methanotrophs needed both O 2 and water, although the optimum moisture for methanotrophy was determined by soil properties [10]. The effect of water on the CH 4 cycle depended on the original soil moisture content. Periodic rainfall events could stimulate methanotrophy (when the soil is dry) and stimulate methanogenes (and CH 4 emission) when the soil was wet.

Methane (CH 4 )
The diffusion of CH 4 from the atmosphere into the soil strongly affects CH 4 consumption, since the upper-most well-aerated mineral soil located immediately underneath the organic layer exhibits the highest methanotrophic activity [28,75]. The mechanisms associated with the effects of litter atmospheric CH 4 uptake by soils are regulated by moisture [19,27,32], as CH 4 diffusion is 10 4 times slower in water than in air [76].
The litter layer may influence soil CH 4 uptake in opposing ways as summarized in Table 2: (i) decrease the uptake by acting as a physical barrier to gas diffusion and reduced aeration due to faster litter decomposition in wet conditions; (ii) increase the uptake through the maintenance of soil gas diffusivity under wetter/high rainfall conditions; (iii) influence the capability of the soil for oxidizing CH 4 ; (iv) provide a source of nutrients for methanotrophs; (v) improve the formation of macro-aggregates, which facilitates CH 4 transport for methanotrophs. n/a n/a n/a n/a 6 1200 Acid brown n/a n/a n/a * Reduction by 17% n/a Black Forest, Southwestern Germany (48 Deciduous forest Fg. sp., Qr. sp. n/a n/a n/a~600 1-2 cm/~570 g dm m −2 yr −1 n/a n/a Cambisol Sandy n/a n/a Reduction by 17% n/a South Central Germany, (49 Temperate deciduous forest Fg. sylvatica n/a n/a n/a n/a n/a 6.5 796 Pseudogleyic Cambisol n/a n/a n/a Reduction by 16% * n/a F. sylvatica Pc. abies n/a n/a n/a n/a n/a n/a n/a Haplic Cambisol n/a n/a n/a Reduction * n/a   S. superba, C. chinensis, Cb. kwangtungense n/a n/a n/a n/a n/a/4. n/a n/a n/a n/a n/a 0.2 337 Organic soil n/a n/a n/a Increased   The role of litter as a physical barrier that restricts the uptake of atmospheric CH 4 is confirmed in studies on forests with different tree species, e.g., pine forests (P. strobus L.) [78], Scots pine [79], evergreen slash pine [19], pure and mixed stands of Norway spruce and European beech [28,80], and beech and oak forests [75,81]. Since the litter layer may serve as a moisture-dependent bidirectional buffer for atmospheric CH 4 uptake, various effects of litter on CH 4 uptake by forest soils are observed in different ecoregions of China [19,31]. The increase in CH 4 oxidation after litter removal is the result of an effect on the availability of both CH 4 and O 2 for methanotrophs through an impact on gas diffusion, which may be limited, particularly under wet conditions [3,18,19,75,[80][81][82]. Since O 2 diffusivity in water (2.1 × 10 −5 cm 2 s −1 ) is lower compared to CH 4 (2.6 × 10 −5 cm 2 s −1 ), it is more likely that O 2 availability is the major factor influencing soil microbiological processes in wet soil. The diffusivity of O 2 in air is also lower than CH 4 , corresponding to 0.205 cm 2 s −1 and 0.250 cm 2 s −1 , respectively. The diffusion of CH 4 into soil was facilitated by litter removal, especially under the higher soil moisture contents found in autumn and winter in pure mature beech forests [32]. With a higher water content, litter may impede CH 4 and O 2 permeation into the soil [81]. Therefore, CH 4 oxidation can be independent of the presence of litter in the dry season, as reported by Wu et al. [27] and Yan et al. [82]. Furthermore, in its role as a physical barrier, the litter layer may facilitate gas diffusion in soil by increasing the formation of macropores [27] since the transformation of litter into organic matter affects soil porosity [85,86]. However, under higher moisture levels, particularly during rainfall events, the presence of litter may increase CH 4 uptake by storing water, which cannot penetrate into the soil profile, as suggested by Borken and Beese [28] in a study on loamy silt soil (under mixed and pure beech stands) and Wang et al. [19] in a study on sandy loam (under evergreen slash pine). The lower CH 4 oxidation in the wet season is a consequence of the lower aeration and the limited diffusion of atmospheric CH 4 due to the reduced diffusion of gases through the litter layer [82] as a consequence of the lower diffusivity of O 2 than CH 4 . On the other hand, there are reports showing an increased CH 4 uptake in litter-covered soils. A few studies demonstrate that the litter layer may show a substantial high affinity of CH 4 oxidation that occurs in the deepest organic layer overlying the mineral soil [87], which is partly supported by the results of Wu et al. [27]. In contrast, a study on the effect of litter from deciduous and coniferous forests shows that the litter itself is not a significant sink or source of CH 4 [88] but may be a source of labile C compounds that can be used by methanotrophs [27,[89][90][91].
Despite the suggested mechanisms for litter-related regulation of GHG exchange, there are also studies that show a litter-induced reduction in CH 4 uptake from the atmosphere, but they do not explain the mechanisms involved in this process [83,92], which underlines the need for further research in this area.
It is difficult to quantify the effects of litter or litter quality on GHG emissions because of the paucity of reliable data. Two examples (the Figure A1 in Appendix A) indicate that the effects of litter can vary depending on the GHG under consideration, with a positive effect of litter amount on CO 2 fluxes but a negative effect on CH 4 fluxes, assuming that litter thickness reflects the amount of litter present. The correlations are significant but could be influenced by all the other factors, e.g., variation in soil conditions, tree species, and climate zones.

Tree Species-Specific Mechanisms of the Litter Effect on GHG Fluxes
Tree species and age may also influence the effect of litter on GHG fluxes through differences in litter quality [42,93] that may modify both the soil characteristics and microbial activity. In tropical forests, the C/N ratio in leaf litter differed significantly among four sites with dominant Leuceana leucocephala, Acacia, Cereus, and Pilosocereus species varying among different forests [94]. In contrast, a study on litter from tree species common in temperate forests (Norway spruce, Sitka spruce, Douglas-fir, European beech, and common oak) reveals that the total litterfall and nutrient concentrations does not differ significantly among the tree species but is affected by site characteristics [95]. The production of leaf litter of a forest predominance of beech (Fagus sylvatica L.), with three deciduous-tree species (beech, ash (Fraxinus excelsior L.), and lime (Tilia cordata Mill. and/or T. platyphyllos Scop.) was also similar to which dominated by five deciduous-tree species (beech, ash, lime, hornbeam Carpinus betulus L.), and maple (Acer pseudoplatanus L. and/or A. platanoides L.). The increasing tree-species diversity (and decreasing the abundance of beech) is accompanied by an increase in the total quantity of Ca and Mg deposited on the soil surface by leaf litter [96]. The increased base status influences microbial populations, biomass and activity through changes in pH. The interpretation of the effect of litter on GHG exchange is improved when studies include accurate and detailed information on ground vegetation. Although studies to date focus mainly on dominant tree species, a few studies also report ground cover, e.g., natural growths and shrubs (Lantana camara, Eupatroim odoratum, Hyptis suavelons) [40], herbs, grasses, scattered patches of mosses [78] and vegetation (Camellia oleifera, Itea yunnanensis), and some low-rise herbs (Artemisia sp.) [18].

Carbon Dioxide (CO 2 )
The contribution of the litter layer to respiration ranged from 5% to 45% of the total CO 2 emissions from temperate forests soils [3]. The emissions of CO 2 were confirmed to be often lower in coniferous forest soils than in deciduous forests [97]. A meta-analysis revealed that natural and doubled litter inputs increased soil respiration in forest ecosystems by 36% and 55%, respectively. The effect of litter inputs on the increase in soil respiration in different types of forests assumed the following order: coniferous forests (50.7%) > broad-leaved forests (41.3%) > mixed forests (31.9%) [89]. In coniferous forests, the removal of litter caused a reduction in CO 2 emissions, ranging from 2.61% in a fir forest in Poland to 68% in a Pinus caribeae plantation in Puerto Rico [39]. After the litter layer removal in a pine forest, CO 2 emissions were reduced by 43%, while CH 4 uptake increased more than twofold under dry and warm soil conditions [78].
These differences may result from the characteristics of the litter layer. Coniferous litter often contains more phenolic acids (e.g., ferulic and p-coumaric acids) than deciduous litter, which can only be used in small concentrations by microbes as substrates for metabolism. At higher concentrations, phenolic acids may inhibit bacterial and fungal activity [62]. Monoterpenes are biogenic volatile organic compounds produced especially by coniferous trees and released through root and litter [98]. Moreover, monoterpenes as well as phenolic compounds belong to secondary plant metabolites [99]. The presence of monoterpenes may increase CO 2 production due to their microbial mineralization, as reported in a Norway spruce site, in contrast to European beech samples [77].
The effect of litter on CO 2 fluxes can also be influenced by tree age. The removal of litter in deciduous forests lowered CO 2 emissions by 5% and 16% in old-growth and 85-year-old forests, respectively [100]. The CO 2 fluxes were found to be higher in a newly established (one-year-old) and a mature (38-years-old) stand, than in young (12-years-old), P. radiate plantations. The new and mature stands had more ground vegetation cover (80-90%) than the young radiata pine stands (30%). Together with a lower soil ammonium content (and lower mineralization rate) in the young stand, poor vegetation coverage may have resulted in lower CO 2 emissions [25]. The newly established pine stand also showed significantly higher CO 2 emissions than the mature stand, due to clear-cut harvesting practices, which result in a lower C/N ratio in the top 10 cm of the soil and lower organic matter quantity. It is known that the maximum CO 2 flux occurs in the top 15 cm of soil depth [40]. A reduced amount of litter and a lower C/N ratio cause rapid C mineralization, which results in increased litter respiration [101].
Broadleaf forests were found to have a relatively higher mean annual litterfall and a higher litter quality compared to mixed or pine forests [102,103]. The removal of litter reduced CO 2 emissions, to varying extents, depended on the type of deciduous tree species. In hornbeam oak forests, soil respiration decreased only slightly: from 2.88 kg CO 2 m −2 year −1 (research point with litter) to 2.78 kg CO 2 m −2 year −1 (litterfree research point), but in the acidophilous beech forest, CO 2 emissions decreased from 2.18 kg CO 2 m −2 year −1 to 1.32 kg CO 2 m −2 year −1 after excluding litter. The amount of CO 2 emitted from forest soils also depended on the speed of litter decomposition, which differed in different types of forests [42]. A beech stand was found to have the slowest litter decomposition, and its accumulation was approximately two to three times higher than in mixed stands of deciduous tree species [104]. Similarly, in a hornbeam oak forest litter decomposition, processes were faster and CO 2 was released more rapidly than in a beech forest. It was estimated that the average decrease in soil respiration globally after litter removal was 27% for different types of forests. The litter decomposition rate, along with soil respiration decreased after litter removal, in a seasonally flooded gallery forest and an upland forest where litter removal resulted in a 10-20% reduction in soil respiration [105]. Therefore, it can be concluded that the rate of litter decomposition had a significant impact on many differences in soil CO 2 emissions between various ecosystems. Tropical forests were very important in this context [50], as may have contributed about 67% to the total annual global CO 2 efflux [105]. These forests could react differently to litter manipulations, since they differed from temperate forests in terms of soil age, biotic composition, erosion, and/or uplift rates [106,107]. Tropical and subtropical forests may also have varied significantly in soil abiotic (e.g., soil moisture and temperature) and biotic (e.g., litter quantity and quality, tree species) factors, which also influenced the impact of litter inputs on CO 2 emissions [39,106,108,109].

Methane (CH 4 )
Similar to CO 2 emission, forest types differ in soil CH 4 uptake ability. The high consumption of atmospheric CH 4 by forest soils confirms the involvement of high affinity methanotrophs [12] and the process is carried out by different groups of methanotrophs [110]. Among the most abundant methanotrophs, Methylocystis spp. and Methylococcus more often populate deciduous forest soils than mixed and coniferous forest soils [111]. A number of studies conducted under the same climatic conditions also indicate that tree species affect CH 4 uptake in forest soils, with deciduous forests consuming more CH 4 than coniferous forests [28,48,112,113]. One of the explanations for this is that it is due to vegetation and soil-related differences in the structure and activity of methanotroph communities [111,114]. Of the factors that could be important, litter and soils from coniferous forests have a lower pH than deciduous stands, typically ranging between 3 and 4 in pine-dominated forests [115][116][117][118]. Such conditions are below the optimum typical level for methanotrophs [119] and may result in a lower CH 4 uptake; however, some methanotrophs have adapted to such conditions in forest soils [120]. A study on different temperate tree species shows that soil under beech trees is more acidic and has lower inputs of Ca and Mg via litter in comparison with mixed stands of deciduous tree species [96,104].
The properties of the litter itself are also an important element. Litter in deciduous forests is characterized by a higher degradability than in coniferous forests, which results in higher soil N turnover rates [121]. Strong interactions between CH 4 oxidation and soil N have also been reported [122][123][124]. In temperate forests, N fertilization is reported to reduce the CH 4 sink [125] due to a salt effect [126] or a higher nitrification rate [127]. In subtropical forests, N deposition can suppress CH 4 uptake by altering methanotroph and methanogen abundance, diversity, and community structure [128].
The presence of monoterpenes may be a largely conifer-specific mechanism responsible for the negative effect of the litter layer on methanotroph activity [3,77]. The concentrations of monoterpenes are negligible in European beech leaves (as well as the roots and soil samples) compared with samples from Norway spruce [77]. Monoterpenes are highly volatile but are initially present at relatively high concentrations in recently fallen litter [99,129,130]. They are released during the long-term decomposition of needle litter (or released by roots), and involve rapid changes in their composition [99,129]. Although the rates of monoterpene loss from needle litter are found to vary initially between conifers, most of the monoterpene content is lost after 4-6 months [99] and decreases sharply by 40-85.4% after 90-120 days [129]. Some compounds are also detected in soils and may interact with soil microorganisms, inhibiting methanotrophs, and decreasing soil CH 4 oxidation [3,77,131,132]. Methanotrophs may differ in their response to monoterpenes at the same concentration [133]. Among the most abundant monoterpenes, α-pinene is the major inhibitor of CH 4 oxidation in soils from a pine, aspen, and mixed stand in Canada [131], while β-pinene has the greatest inhibitory effect on the process in Norway spruce soil [77]. Among conifers, different monoterpene compounds may dominate. For example, α-pinen is found to be the dominant form in needles of Pinus densiflora, Pinus rigida, and Pinus sylvestris, while β-pinene is dominant in needles of Pinus thunbergii, Picea sitchensis, and Picea abies [129]. The removal of the organic layer in a German coniferous mixed forest causes a significant increase in CH 4 oxidation rates by a factor of 1.4-2.5 [80]. In Chinese pine stands, the removal of the litter layer reduces soil CO 2 emission by 17% and increases CH 4 uptake by 24% [18]. This is even higher in a study reported by Fan et al. [3] where the CO 2 emissions decrease by 24-32%, while CH 4 uptake increases by 55%. Although the litter layer reduces the atmospheric CH 4 uptake by dry soil, the removal of the layer has a minimal effect (+0.7%) on the annual CH 4 budget, through compensating effects during the wet and dry seasons [19].

Environmental Controllers of the Impact of Litter on GHG Fluxes
Consideration of the litter effect on soil GHG fluxes should include the role of climatic conditions, mostly temperature and precipitation. Due to the dynamics of these parameters, the litter effect on GHG fluxes may show significant seasonal variability.
Rainfall in an oak forest is reported to increase CO 2 emissions, which results in the rapid reactivation of litter-associated microorganisms [46]. On the other hand, in a tropical montane cloud forest in Peru, lower moisture levels do not change the soil respiration after litter removal, which is explained by the fact that the litter and organic matter are decomposed by microorganisms with different moisture sensitivities [47]. Litter respiration also depends largely on moisture content, and the contribution of litter to soil respiration is influenced by the frequency and amount of precipitation [46,48,134,135]. Warm and humid climatic conditions accelerate organic matter decomposition, resulting in increased rates of soil respiration [100]. Changes in the water content of the litter layer are often transient, since litter is directly exposed to wind and solar radiation [136]; nevertheless, they can influence CO 2 emissions, and therefore the overall forest CO 2 budget [137]. Continuous cycles of wetting and drying of the litter layer led to transient CO 2 emission [136,138]. An increase in CO 2 emissions followed by an increase in the litter moisture content due to rainfall was also reported in a semi-deciduous old-growth tropical forest (with mostly evergreen species) [49] and in a forest dominated by Quercus serrata Thunb. [136]. Studies carried out in coniferous and deciduous forests in different locations in Europe (Italy, the Netherlands, and Finland) show that leaf litter is the main source of CO 2 and the emissions peak at the higher moisture contents for all types of litter, temperatures, or sites, while the optimum soil CO 2 emissions are achieved at intermediate moisture contents (40-70% WFPS) [88]. Litter removal causes a decrease in soil moisture [15] (compared to soil with litter) and when soil moisture is low, both the transport of nutrients and the metabolism of decomposing microbes are reduced [139,140].
Hydrological conditions are a strong regulator of the CH 4 cycle since saturated (flooded) soil can be a source of methane, while well-aerated soil can be a sink of this gas [10]. After a high rainfall in the temperate zone or during the wet season in a tropical climate, soil may also emit CH 4 , although the net exchange between soils and the atmosphere depends on how this impacts the balance between CH 4 production and consumption [19]. The strong dependence of CH 4 oxidation on water content is confirmed as litter addition during the dry season, as this does not significantly affect CH 4 uptake, while it decreases by 47.1 ± 4.9% during the wet season after doubling the litter level [27]). The litter effect may result from enhanced microbial activity and/or from changes in litter quality and decomposition rate [16]. It is reported that litter may store water during rainfall events. Since water cannot penetrate the mineral soil, a high soil diffusivity is main-tained [28]. However, a study on different types of needleleaf and broadleaf litters revealed that the rainfall interception storage capacity of the litter layer varied with physical features and rainfall characteristics [93]. The interception-related storage capacity of needle litters varied significantly with the litter type, while there were no significant differences in water storage across the broadleaf litter types. It was reported that a higher intensity or longer duration of rainfall events could increase the interception storage capacity in all broadleaf and needleleaf litters [93].
The litter layer was an effective insulator, isolating the soil from the effects of variations in irradiance, consequently lowering soil temperature [21]. In a deciduous forest, litter and soil temperatures were responsible for 68% to 81% of the variability in CO 2 emissions, respectively [100]. When there was no litter on the soil surface, the influence of temperature on soil respiration was higher, the activity of soil microbes and their enzymes increased, and the degradation of organic matter was greater. Thus, under these conditions, an increase in soil CO 2 emission could often be observed [21,[141][142][143]. After the removal of litter in a Quercetum petraeae-cerris forest in northeastern Hungary, the soil was found to reach higher temperatures in summer and lower temperatures in winter [37]. The greatest reductions in CO 2 emissions after litter exclusion were observed in a Cinnamomum camphora forest in China (39.2%) [144], in a beech (Fagus sylvatica) forest in Poland (about 39.45%) [42], and in a wet tropical forest dominated by Tabebuia heterophylla in Puerto Rico (54%) [39].

Forest Management
In addition to environmental regulators, forest management can also control the contribution of litter to GHG emissions, via their effect on litter inputs through management practices involving tree removal, grazing, extensive deforestation, or fertilization [13,[22][23][24]. Soil disturbance, as well as the mixing and movement of litter and organic material deeper into the mineral soil layer during the mechanical site preparation in clear-cut sites results in a decrease or no increase in CO 2 emissions, although litter decomposition accelerates [1,25,26]. It is reported that clear-cut harvesting results in an up to 50% decrease in soil C [1,25], but a low or insignificant impact on C stocks is also reported [145,146]. In spruce forests after clear-cutting, soil respiration is higher in undisturbed plots than in plots with a disturbed upper soil layer and this is influenced by the thickness of the litter layer, the degree of upper soil damage, and the presence of logging residues on the soil surface [147]. Different management practices modify the soil conditions for methanotrophic bacteria and may differently affect CH 4 uptake in forest soils. Clear-cutting is found to weaken the capability of methanotrophs to oxidize CH 4 through alterations in soil temperature, water content, and C decomposition, while selective cutting stimulates, suppresses, or has no effect on CH 4 uptake [148]. Similar to agricultural soils, fertilization inhibited soil CH 4 uptake in N-rich forest soils but promoted consumption in N-deficient forest soils [148]. Forest conversion, which has an important effect on all GHG fluxes, significantly affects CH 4 absorption capacity through plant allelopathy (including the release of monoterpenes from litter and roots), changes in soil properties (e.g., temperature, moisture, pH, C and N availability), and has an effect on soil microorganisms [132]. The conversion of a naturally grown beech forest into a pure spruce and a mixed stand in Germany reduces the mean CH 4 uptake rates by 58% and 22% during the growing season and by 54% and 20% during the dormant season. This is associated with the formation of a distinct O horizon during the succession and changes in the population density or activity of methanotrophs [28].

Effect of Climate Change
Climate change influences CH 4 and CO 2 fluxes and the net ecosystem C balance through its effect on abiotic and biotic drivers [149,150]. The predicted increases in CO 2 , N deposition, temperature, elevated ozone concentrations, drought, and acid deposition will have an impact on net primary production (NPP), which in turn will affect both the quality and quantity of aboveground litter [15]. An increase in atmospheric CO 2 concentration may stimulate plant growth and thus increase litterfall [20,21,51,144]. The increased litter input will influence soil nutrient content and substrate availability, thereby impacting on microbial community structure and activity with associated increases in CO 2 efflux [14,52,105,[151][152][153][154][155][156]. Natural and doubled litter inputs tend to increase soil microbial biomass C and soil DOC by 21.0-33.6% and 60.3-87.7%, respectively. Litter inputs also increase the total phospholipid fatty acid (PLFA) content, which are linearly related to soil respiration, whilst decreasing the fungal/bacterial PLFA ratio, which decreases soil respiration [89]. A doubling of needle litter input in old-growth western hemlock (Tsuga heterophylla (Rafinesque) and Douglas fir (Pseudotsua menziessi Mirb.) forests results in increased soil respiration [23] while it does not increase CO 2 emissions in a European deciduous forest due to low soil moisture contents, which impede the litter decomposition processes [37]. Higher litter amounts can promote soil C release through the priming effect [70]. It is found that litter addition has a stronger effect on soil respiration in subtropical broadleaf forests than in pine and mixed forests, which is a result of the quantity and quality of the litterfall [13] reflected in the foliar N content, soil C:N ratio, lignin:N ratio, and phenolic content [157]. Climatic factors can also change the chemistry of the litter and affect its susceptibility to decomposition. Elevated CO 2 levels decrease the quality of Populus tremuloides leaf litter by increasing the C:N ratio, the phenolic, and the lignin content [158]. In a study on Molinia caerulea litter, N fertilization increases the N and decreases the C concentration, whereas elevated CO 2 decreases litter quality (increases C:N) and increases the lignin and saturated fatty acid content [159]. As a consequence, CH 4 production in laboratory incubation experiments decreases when litter from N-fertilized plants is used, whereas the litter from elevated CO 2 treatments has no effect [159]. Litter addition, in combination with warming, significantly increases CH 4 uptake rates due to a soil drying effect and increases the availability of DOC [159]. In turn, the repeated addition of beech litter impedes the diffusion of CH 4 and significantly reduces the annual uptake rate; due to their shape, broad beech leaves reduce CH 4 uptake more efficiently than spruce needles [160]. Future increases in N deposition may modify monoterpene fluxes in subtropical forests, as demonstrated in a study on a Masson pine forest and a monsoon evergreen broad-leaved forest, where long-term N addition results in an increased monoterpene release from litter [98]. Higher monoterpene contents in the future may reduce methanotrophic activity, as reported previously [3,77].
An increase in the range and impact of alien pests, particularly in Northern and Eastern Europe, is another likely consequence of climate change that may decrease C sequestration in forests, reduce forest structural and functional redundancy, and affect their resilience to future climate change [161,162]. Insect defoliation increases the organic inputs into the soil in the form of insect feces, cadavers, litter (changed in quality), and other plant material, and may result in an increase in soil CO 2 emission via a transformation of tree biomass into fast decomposable organic matter [163]. Aside from defoliators, wood-boring infestations can shift forests from C sinks to sources [162].

Future Research
The complex and multi-level influences of litter on the soil-atmosphere GHG exchange in forests shows a strong relationship between above-and below-ground processes. Given the important contribution of forest ecosystems to climate change mitigation, an understanding of the, interactions between litter production, CO 2 emissions and CH 4 uptake, is essential for a more complete understanding of the underlying controls and drivers of the GHG budget in these ecosystems. In turn, this will provide information on how best to utilize forest ecosystems to sequester carbon and reduce GHG emissions. This approach also needs to consider the effects of human activity and associated management interventions and how these are impacted by climate change. Based on this review, we propose that future research should focus on the following areas: (i) Due to the sensitivity of soil CO 2 efflux to climatic factors, it is important to focus on the multiple effects of climate change (e.g., increased CO 2 and CH 4 concentrations, increased temperature, drought, extreme precipitation events) on GHG exchange between the atmosphere and the litter-covered soil (or soil without litter) in various types of forests, globally. The litter may have direct or indirect effects on GHG emissions and decomposition processes [144], and even small changes can alter the global C balance, and atmospheric CO 2 concentrations, or nutrient cycling, with the potential to exacerbate the effects of climate change [21,105]. (ii) Equally noteworthy is the need to have a better understanding of global change-driven forest succession, where broad-leaved trees have begun to appear in needle-leaved ecosystems. Due to the shifts in tree composition, a number of ecosystem (e.g., litterfall rates, litter quality and soil-related processes, soil organic matter decomposition, and GHGs production) are changing [64]. (iii) Climate change affects the primary productivity of forests, and elevated CO 2 concentrations in the atmosphere may result in increased litterfall and increased organic matter inputs into the soil, resulting in increased C sequestration [52,144]. Large amounts of aboveground litter can also lead to a priming effect-a complex but not fully understood (especially in situ) soil-plant interaction [68]. As a result of the increased contribution of fresh organic matter to the soil, the decomposition processes are stimulated and the older stored C is released from the soil as CO 2 [52]. A better understanding of this phenomenon is important in the context of future climatic scenarios, according to which litter inputs will be increased [52] and the occurrence of priming effects may intensify. (iv) The effect of monoterpenes on CH 4 uptake is a largely uninvestigated topic and we propose several research areas that require attention: (a) the better recognition and identification of monoterpenes dominant in litter from different tree species, combined with the recognition of methanotroph responses; (b) understanding the longevity of any effects of monoterpenes in litter since recently fallen litter has a higher content of monoterpenes [129,130,164]; (c) finding out whether there are inhibitors that reduce the emissions of terpenes from litter, since monoterpenes can decrease the activity of methanotrophs; (d) the verification of whether methanotrophs produce monoterpenes [165] would be worthwhile with the objective of stimulating CH 4 uptake in coniferous forests, as terpenes are generally widespread in both plant and bacterial metabolism [166]; (e) further work is also required on the effect of increasing the N deposition on monoterpene fluxes [98] in all climatic zones. (v) The effect of forest pests on GHG emissions, including their impact on litter quality and quantity, is poorly understood [162,163] and requires further investigation. (vi) The identification of the underlying processes through which litter influences soil processes requires research into the species composition of microbial consortia occurring in different forest types in a range of climatic zones. (vii) Finally, the characteristics of the investigated litter needs to be specified in more detail, including the thickness of the litter, its morphology, temperature, number of layers and the degree of decomposition.  Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the review; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the review.

Funding:
The research was partly financed by the Polish National Centre for Research and Development within of ERA-NET CO-FUND ERA-GAS Program (contract number ERA-GAS/I/GHG-MANAGE/01/2018) "GHG-Manage".
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the review; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the review.