Ideas and perspectives: patterns of soil CO₂, CH₄, and N₂O fluxes along an altitudinal gradient – a pilot study from an Ecuadorian neotropical montane forest

Tropical forest soils are an important source and sink of greenhouse gases (GHGs), with tropical montane forests, in particular, having been poorly studied. The understanding of this ecosystem function is of vital importance for future climate change research. In this study, we explored soil fluxes of carbon dioxide (CO₂), methane (CH₄), and nitrous oxide (N₂O) in four tropical forest sites located on the western flanks of the Andes in northern Ecuador. The measurements were carried out during the dry season from August to September 2018 and along an altitudinal gradient from 400 to 3010 m a.s.l. (above sea level). During this short-term campaign, our measurements showed (1) an unusual but marked increase in CO₂ emissions at high altitude, possibly linked to changes in soil pH and/or root biomass, (2) a consistent atmospheric CH₄ sink over all altitudes with high temporal and spatial variability, and (3) a transition from a net N₂O source to sink along the altitudinal gradient. Our results provide arguments and insights for future and more detailed studies on tropical montane forests. Furthermore, they stress the relevance of using altitudinal transects as a biogeochemical open-air laboratory with a steep in situ environmental gradient over a limited spatial distance. Although short-term studies of temporal variations can improve our understanding of the mechanisms behind the production and consumption of soil GHGs, the inclusion of more rigorous sampling for forest management events, forest rotation cycles, soil type, hydrological conditions and drainage status, ground vegetation composition and cover, soil microclimate, and temporal (seasonality) and spatial (topographic positions) variability is needed in order to obtain more reliable estimates of the CO₂, CH₄, and N₂O source/sink strength of tropical montane forests.

1 The importance of tropical forests for greenhouse gas budgets

Soils play a vital role in the global greenhouse gas (GHG) budget. Tropical forest soils, in particular, represent a net sink of carbon (C) (Pan et al., 2011), but at the same time, they are the largest natural source of N₂O, with an estimated contribution of 14 %–23 % to the annual global N₂O budget (Werner et al., 2007). In general, soil CO₂ is produced mainly by root respiration, microbial respiration, litter decomposition, and the oxidation of soil organic matter (Dalal and Allen, 2008). CH₄ is consumed by methanotrophic bacteria (Jang et al., 2006); however, forest soils prone to inundation emit CH₄ by methanogenic microorganisms (Archaea domain). N₂O is emitted through denitrification or a number of alternative pathways (e.g., nitrification, nitrifier denitrification, chemodenitrification, etc.; Butterbach-Bahl et al., 2013; van Cleemput, 1998; Clough et al., 2017) but can also be consumed during complete denitrification (Butterbach-Bahl et al., 2013). Overall, tropical forest soils emit on average 12.1 t CO₂-C ha⁻¹ yr⁻¹ (heterotrophic and autotrophic respiration), slightly less than the net primary productivity (NPP) (12.5 t CO₂-C ha⁻¹ yr⁻¹); i.e., the net C sink (belowground and aboveground) of tropical forests is ∼ 0.4 t CO₂-C ha⁻¹ yr⁻¹ (Dalal and Allen, 2008; Grace et al., 2006). Under aerobic conditions, CH₄ fluxes vary from −0.7 to −30.0 kg CH₄-C ha⁻¹ yr⁻¹, with an average consumption of −3.0 kg CH₄-C ha⁻¹ yr⁻¹, while the mean rate of N₂O emissions from tropical forest soils is 3.03 ± 0.52 kg N₂O-N ha⁻¹ yr⁻¹ (Dalal and Allen, 2008), i.e., 2–3 times higher than the mean N₂O emissions from temperate forest soils (1.0 ± 0.36 kg N₂O-N ha⁻¹ yr⁻¹; Chapui-Lardy et al., 2007; Van Groenigen et al., 2015).

The understanding of the mechanisms and processes underlying GHG flux variability has greatly improved over the last decades (Butterbach-Bahl et al., 2013; Heil et al., 2016; Müller et al., 2015; Sousa Neto et al., 2011; Su et al., 2019; Teh et al., 2014). However, there is still (1) considerable uncertainty about the overall balances of many ecosystems (Castaldi et al., 2013; Heil et al., 2014; Kim et al., 2016b; Pan et al., 2011; Purbopuspito et al., 2006), (2) a strong imbalance in field observations, skewed towards the Northern Hemisphere (Jones et al., 2016; Montzka et al., 2011), and (3) a bias towards the quantification of emissions in lowland forests within the tropics (Müller et al., 2015; Purbopuspito et al., 2006; Wolf et al., 2011). For instance, based on a compilation made of CO₂, CH₄, and N₂O fluxes in South America (Table S1) from 1983 to 2019, there have only been six studies carried out on tropical montane forests (i.e., > 2000 m a.s.l., above sea level), while they represent more than 11 % of the world's tropical forests (Müller et al., 2015; Teh et al., 2014). In fact, Teh et al. (2014) and Spahni et al. (2011) have argued that tropical upland soils are one potentially important source of CH₄ and N₂O that has been overlooked in both bottom-up and top-down emission inventories; their sink/source strength might be comparable to or greater than their lowland counterparts and, therefore, quantitatively important in regional and global GHG budgets.

2 Altitudinal gradients as a biogeochemical open-air laboratory

To further improve our understanding of the role of tropical forest ecosystems in the global GHG balance, environmental gradients (altitudinal, latitudinal, etc.) can offer great opportunities to study the influence of abiotic factors on biogeochemical processes under field conditions (Bauters et al., 2017; Jobbágy and Jackson, 2000; Kahmen et al., 2011; Laughlin and Abella, 2007), which complements the knowledge on short-term responses from experimental approaches. In the case of altitudinal gradients, these responses are driven by abiotic variables that covary with elevation, which, amongst others, creates a distinctly strong climate gradient over a short spatial distance (Bubb et al., 2004; Killeen et al., 2007; Körner, 2007; Myers et al., 2000).

Moreover, since altitudinal gradients reflect long-term adaptations based on a broad range of factors, they provide valuable insights into the influence that climate change may have on ecosystem processes (Malhi et al., 2010). There is indeed a growing concern regarding the sensitivity of tropical forests to climate change mainly because species in the tropics have evolved with narrow thermal tolerances compared to their temperate counterparts; this makes them particularly vulnerable to changes in global climate (Fadrique et al., 2018; Perez et al., 2016). Therefore, the effects of global warming are expected to be severe in the tropics, and the understanding and integration of the magnitude of their feedbacks in the Earth system are important to come up with appropriate forest management options to mitigate climate change (Bonan, 2008; Li et al., 2020).

To address these knowledge gaps, we present a pilot study of the soil–atmosphere exchange of CO₂, CH₄, and N₂O along an altitudinal gradient in a neotropical montane forest located on the western flanks of the Andes in northern Ecuador. The sampling campaign took place from 6 August to 28 September 2018. Four study sites (Fig. S1) were selected: Río Silanche at 400 m a.s.l (hereinafter: S_400), Milpe at 1100 m a.s.l. (hereinafter: M_1100), El Cedral at 2200 m a.s.l. (hereinafter: C_2200), and Peribuela at 3010 m a.s.l. (hereinafter: P_3010). Gas samples were taken using a static flux chamber method once per day per stratum over 2 weeks. Samples of soil were collected once during the whole field campaign for the analysis of bulk density (ρ_b), pH, nitrate (NO${}_{\mathrm{3}}^{-}$) and ammonium (NH${}_{\mathrm{4}}^{+}$) content, C and nitrogen (N) concentrations, stable N isotope signatures (δ¹⁵N), and soil texture. Additionally, soil moisture (expressed as water-filled pore space, WFPS) and soil temperature were measured daily. Specifically, we aimed to determine the magnitude of the soil–atmosphere exchange of CO₂, CH₄, and N₂O during the dry season. By working along this altitudinal gradient, we wanted to explore the potential effect of altitude on the GHG fluxes of the forest soils. Findings from this research could provide insights for future and more detailed studies on tropical montane forests.

Figure 1 (a) Soil CO₂ (mg C m⁻² h⁻¹), (b) CH₄ (µg C m⁻² h⁻¹), and (c) N₂O (µg C m⁻² h⁻¹) fluxes per month at Río Silanche (400 m a.s.l.; S_400), Milpe (1100 m a.s.l.; M_1100), El Cedral (2200 m a.s.l.; C_2200), and Peribuela (3010 m a.s.l.; P_3010). Light gray boxplots indicate the fluxes in August 2018, whereas dark gray boxplots indicate the fluxes in September 2018. Light gray dots in each boxplot represent the measurements taken each day and black dots the outliers of the respective site. The dotted red line across the boxes indicates zero net flux.

Table 1 Average measurements plus or minus the standard deviations (SDs) of soil CO₂, CH₄, and N₂O fluxes at Río Silanche (400 m a.s.l.; S_400), Milpe (1100 m a.s.l.; M_1100), El Cedral (2200 m a.s.l.; C_2200), and Peribuela (3010 m a.s.l.; P_3010) per month.

Note: flux values represent the mean of five chambers per site and measurement week using four-point time series and considering the constraint set to evaluate linearity in each measurement cycle (R²>0.65).

3 What did we see in Ecuador?

Across our study sites, P_3010 (the highest stratum) exhibited the highest soil CO₂ emissions (Fig. 1a and Table 1) probably due to the dominant role of soil pH and shifts in C allocation patterns. The highest soil pH in water (pH_water) was observed at this site (Table 2), and under acid conditions, Sitaula et al. (1995) and Persson and Wiren (1989) reported a decrease in CO₂ emissions with decreasing pH_water. On the other hand, although not measured or estimated in this study, an increase in fine root biomass is expected in tropical mountain forests compared to lowland forests due to imbalances or limitations in resource (water and/or nutrients) availability at higher altitudes (Bauters et al., 2017; Leuschner et al., 2007). Therefore, the observed increase in CO₂ emissions at P_3010 might be further driven by an increase in root biomass as the latter has been shown to be positively correlated with soil respiration (Han et al., 2007; Luo and Zhou, 2006a; Reth et al., 2005; Silver et al., 2005).

Table 2 Physicochemical soil properties of the study areas Río Silanche (400 m a.s.l.; S_400), Milpe (1100 m a.s.l.; M_1100), El Cedral (2200 m a.s.l.; C_2200), and Peribuela (3010 m a.s.l.; P_3010) at 5 and 20 cm depth, including mean values plus or minus the standard deviations (SDs) of bulk density (ρ_b), porosity, pH in water (pH_water) and KCl suspension (pH_KCl), nitrate (NO${}_{\mathrm{3}}^{-}$) and ammonium (NH${}_{\mathrm{4}}^{+}$) concentrations, bulk nitrogen (N) and carbon (C) content, carbon-to-nitrogen ratio (C∕N), and δ¹⁵N signatures from samples of soil taken in August. Similar lowercase letters in superscript and next to some values within one row and per depth (5 and 20 cm) indicate no significant difference at P<0.05 between sites (S_400, M_1100, C_2200, and P_3010).

Note: mean values plus or minus the SD were calculated from soil samples taken adjacent to each soil chamber (n=5) except for soil texture for which composites for each site at 5 and 20 cm depth were made from the soil samples taken from each chamber. ¹ Commonly known as Andisol in the United States Department of Agriculture (USDA) soil taxonomy. ² Expressed per gram of dry soil. ³ Calculated by dividing C (%) by N (%) for each soil sample. ⁴ Expressed relative to the international standard AIR.

In contrast to P_3010, the low CO₂ emissions observed at C_2200 could be attributed to (1) the lower WFPS (Fig. S3), (2) the lower contents of C and N (Table 2), or (3) the higher bulk density (Table 2). The lowest soil water content was observed at this site in August at 5 cm depth, and exactly in this month, the lowest emissions of CO₂ were obtained. The low contents of C and N exhibited in C_2200 (indeed, the lowest from all the sites) could also have hampered the CO₂ emissions (Dalal and Allen, 2008; Luo and Zhou, 2006a; Oertel et al., 2016). Additionally, this site had the highest soil bulk density (i.e., lowest porosity), which could have led to a decrease in soil respiration either by a physical impediment for root growth or by a decrease in soil aeration for microbial activities (Dilustro et al., 2005; Luo and Zhou, 2006a, b).

All sites acted as net sinks for CH₄ (Fig. 1b and Table 1) (i.e., uptake of atmospheric CH₄ by soils). During the entire field campaign (10 d), only one chamber at one site (S_400) and on a specific date (08/09/2018) acted as a net source of CH₄ (43.2 µg CH₄-C m⁻² h⁻¹). However, there were no statistical differences between months, and all sites exhibited indeed a high temporal and spatial variability.

Only S_400 and M_1100 (both months) (i.e., plots located at the lower locations) acted as net sources of N₂O (Fig. 1c and Table 1), whereas the plots located at the highest stratum (P_3010 & C_2200) showed a general net N₂O consumption during August and September.

The N₂O emissions obtained at the lowest strata corroborate the literature data on lowland tropical forests (Butterbach-Bahl et al., 2004, 2013; Koehler et al., 2009) and could be mainly attributed to the soil water content, temperature, and N availability observed at these sites (Figs. S2 and S3 and Table 2). Firstly, N₂O emissions in tropical forest soils are predominantly governed by WFPS which influences microbial activity, soil aeration, and thus the diffusion of N₂O out of the soil (Davidson et al., 2006; Werner et al., 2007). Secondly, an increase in temperature leads to an increase in soil respiration and thus to a depletion of oxygen concentrations, which is indeed a major driver in N₂O emissions. In fact, rising temperatures lead to a positive feedback in microbial metabolism, in which the stimulation of mineralization and nitrification processes induces an increase in the availability of substrates for denitrification and thus to an increase in N₂O emissions (Butterbach-Bahl et al., 2013; Sousa Neto et al., 2011). Finally, the dependency of N₂O emissions on WFPS and temperature is affected by substrate availability (NO${}_{\mathrm{3}}^{-}$). High contents of NO${}_{\mathrm{3}}^{-}$ give an indication of an open or “leaky” N cycle with higher rates of mineralization, nitrification, and thus N₂O emissions (Davidson et al., 2006). Moreover, NO${}_{\mathrm{3}}^{-}$ is normally preferred as an electron acceptor over N₂O, and it can also inhibit the rate of N₂O consumption to N₂ (Dalal and Allen, 2008).

In contrast to the low elevation sites where net N₂O emissions were observed, P_3010 and C_2200 (Fig. 1c and Table 1) presented net consumption (negative values, i.e., fluxes from the atmosphere to the soil). From 35 valid measurements, only one resulted in net emission at P_3010 (range: −9.3 to 0.95 µg N₂O-N m⁻² h⁻¹), whereas from 36 valid measurements, 19 resulted in net emissions at C_2200 (range: −104.9 to 9.3 µg N₂O-N m⁻² h⁻¹). Net N₂O consumption is often related to N-limited ecosystems, and it is presumably the cause in our case. At low NO${}_{\mathrm{3}}^{-}$ concentrations, atmospheric and/or soil gaseous N₂O may be the only electron acceptor left for denitrification (Chapui-Lardy et al., 2007; Goossens et al., 2001). P_3010 had the lowest content of NO${}_{\mathrm{3}}^{-}$ along with the lowest soil δ¹⁵N (Table 2), which clearly reflects the shift towards a more closed N cycle at higher elevations (Bauters et al., 2017; Gerschlauer et al., 2019). In fact, studies performed by Teh et al. (2014) and Müller et al. (2015) in the southern Peruvian and Ecuadorian Andes, respectively, related the decrease in N₂O emissions and thus the potential for N₂ production in soils at high elevations to differences in NO${}_{\mathrm{3}}^{-}$ availability. Moreover, Wolf et al. (2011) and Martinson et al. (2013) have indicated that N availability was (1) a dominant control on N₂O fluxes and (2) inversely proportional to altitude. In addition, the low N₂O fluxes could also be supported by the high content of clay (Table 2) and CO₂ emissions (Fig. 1a) (i.e., development of microsites for N₂O reduction) along with the low soil water content (% of WFPS) (Fig. S3) (i.e., better diffusion of atmospheric N₂O into the soil) and higher soil pH value (Table 2) (i.e., less severe inhibition of the nitrous oxide reductase) observed at P_3010 (Chapui-Lardy et al., 2007).

It is important to mention that the region where these measurements were taken is characterized by a marked seasonality in rainfall. We measured at the end of the dry season; thus, it is expected that there will be fluctuations in net fluxes (sources vs sinks) depending on the season. Moreover, although our limited dataset did not allow us to corroborate the main drivers that controlled these fluxes, daily measurements like those carried out here reflect the importance of evaluating short-term variations. As such, the net N₂O consumption with increasing altitude might be overlooked in an annual analysis, but it is equally important to (1) understand the mechanisms behind the production and consumption of N₂O and (2) have reliable estimates of the N₂O source/sink strength of tropical forests for regional and even global GHG budgets. Moreover, in order to corroborate the net consumption observed at high altitudes and improve the understanding of N₂O dynamics in terrestrial ecosystems, disentangling gross N₂O production and consumption at field scale is needed. Although the most commonly used method to measure N₂O fluxes via static chambers only allows the quantification of net fluxes, stable isotope techniques would greatly contribute to our mechanistic understanding of gross fluxes. For instance, enrichment and natural abundance approaches (¹⁸O, ¹⁵N) can be used to identify and estimate the contribution of different microbial processes to N₂O production/consumption (Butterbach-Bahl et al., 2013; Yu et al., 2020). Nevertheless, (1) the coupling of isotope techniques with molecular analyses of functional genes is paramount to fully understand the complexity of the microbial processes present, and (2) the improvement of measuring techniques for N₂O reduction is needed to close N ecosystem balances (Butterbach-Bahl et al., 2013; Chapui-Lardy et al., 2007). In fact, microbial composition and diversity, as well as the presence or absence of important genes (e.g., N₂O reductase nosZ I and nosZ II; Van Groenigen et al., 2015b) can help to detect N₂O consumption. Similarly, analytical techniques such as Raman gas spectroscopy could be used to detect and quantify N₂ fluxes from denitrification (Frosch et al., 2016), which is indeed a novel and simple approach compared to previously widely used techniques that may have led to underestimations (Fang et al., 2015).

4 Conclusions and future directions

GHG fluxes from tropical montane forests in South America are particularly scarce with limited spatial coverage and seasonal fluctuation in rainfall but important for consideration in future field measurements and modeling research. Overall, we found an unusual but marked increase in CO₂ emissions at the highest altitude, which is probably explained by soil pH and root biomass even though the latter was not measured or estimated. Our CH₄ uptake fluxes exhibited a high temporal and spatial variability but reiterate the role of humid tropical forest soils as CH₄ sinks. Contrary to the net N₂O emissions observed in the lowest strata, the net consumption at higher elevations seems to be quite unique, and it might reflect the shift towards a more closed N cycle at higher altitudes that was reported previously in tropical regions. Our results highlight the importance of short-term variations in N₂O fluxes, but it calls for more and broader studies especially in tropical montane forests, including the impact of spatial and temporal variability, forest management events and forest rotation cycles, ground vegetation composition and cover, and soil microclimate and hydrological conditions, as well as the implementation of isotope techniques, the coupling of microbial analysis with N₂O fluxes, and the response of tropical forests to current and future changes in N content.

In terms of spatial variation, GHG fluxes may vary between lower slope, mid-slope, and/or ridge (see Table S1) (Courtois et al., 2018; Teh et al., 2014; Wolf et al., 2011, 2012). Fluctuations of net fluxes can be observed depending on the season and the transition between them (see Table S1) (Butterbach-Bahl et al., 2013; Kim et al., 2016a). Management events (e.g., thinning, clear cutting, fertilization, draining improvements) and/or forest composition and growth stage (e.g., young vs mature forest) may influence, e.g., forest vegetation, soil characteristics, hydrology, and nutrient management among others and ultimately lead to changes in soil GHG fluxes (Barrena et al., 2013; Jauhiainen et al., 2019; Kim et al., 2016a). Moreover, soil hydrology (runoff, evapotranspiration, soil moisture, etc.) may affect biogeochemical cycles (Kim et al., 2016a). Microbial composition and diversity could be a key to understand the variability in N₂O fluxes (Butterbach-Bahl et al., 2013). Changes in N content – due to, e.g., urban development and increasing use of agricultural land – could cause shifts in soil N cycling and thus CO₂, CH₄, and N₂O fluxes (Koehler et al., 2012). Besides this, the effects of climate change on tropical regions (e.g., increases in temperature and CO₂ concentrations, as well as changes in rainfall patterns and drought events) may also affect soil GHG fluxes. Therefore, a strategic plan must be implemented. Long-term data of at least 1 or 2 hydrological years is needed, with sampling intervals covering seasonal fluctuations and being appropriate for the type of land (i.e., spatial variability across different topographical positions). The effect of N content and climate change in tropical forests could be evaluated using laboratory (e.g., incubations under controlled conditions) and/or field experiments (e.g., see Koehler et al., 2009, 2012; Hall and Matson, 1999; Martinson et al., 2013) with the use of altitudinal gradients like biogeochemical open-air laboratories. Finally, although sampling conditions in tropical montane regions can be challenging, (1) establishing networks and collaborations with local communities (i.e., citizen science) could contribute not only in terms of data acquisition but also in the development of local knowledge (e.g., how climate and land use change might affect ecosystems and people), and (2) modeling approaches for C and N biogeochemistry in forest ecosystems (e.g., Forest-DeNitrification-DeComposition, DNDC, model; GRAMP, 2013) could help to upscale fluxes from site to regional level. Nevertheless, the cooperation and contribution between field researchers and scientific organizations, e.g., in South America and around the world, as well as the capacity building in the respective countries, are crucial for improving our understanding of soil GHG fluxes from tropical regions and paramount for getting tangible datasets of remote regions such as montane forests.