Displays

Tropical peatlands of Southeast Asia are a globally important carbon reservoir, storing an enormous amount of soil organic carbon as peat. These ecosystems are complex and poorly understood with large unknown biogeochemical processes. Despite the huge carbon stocks in these ecosystems, data on ecosystem-scale carbon dioxide (CO₂) and methane (CH₄) fluxes are still limited in comparison with mid- and high-latitude peatland ecosystems. The recent increase in the intensity of climate anomaly such as El Niño may alter the hydrological regime of this ecosystem, thus affects its carbon cycling. It is crucial to quantify the CO₂ and CH₄ fluxes of the ecosystem and understand their responses to environmental changes to predict the role of peat swamp forest in global carbon cycles. To date, the application of the eddy covariance technique to measure the ecosystem-scale CO₂ and CH₄ fluxes in tropical peatlands is still limited to few studies in Malaysia and Indonesia.

In 2010, we established a long-term greenhouse gas fluxes monitoring using the eddy covariance technique over a peat swamp forest in Sarawak, Malaysia. Here, we present the net ecosystem exchange of CO₂ (NEE) and CH₄ (F_CH4) from February 2014 to January 2017 (3 years). We had quantified the NEE and F_CH4, the diurnal and seasonal variations of NEE and F_CH4, and the response of NEE and F_CH4 to GWL. The F_CH4 was determined half-hourly as the sum of eddy CH₄ flux and CH₄ storage change in an air column below the flux measurement height. We had determined the global warming potential of this ecosystem from annual NEE and F_CH4 using sustained-flux global warming potential (SGWP).  The annual F_CH4 was converted into a CO₂ equivalent unit using an SGWP factor of 45 which represents the SGWP for CH₄ over a timescale of 100 years. Our preliminary result showed that the CH₄ emission potentially offset the CO₂ sequestration, which was higher than those reported in other regions in the world.

How to cite: Kiew, F., Wong, G. X., Hirata, R., Tang, A., and Melling, L.: Ecosystem-scale measurements of CO2 and CH4 fluxes from a tropical peatland in Sarawak, Malaysia, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-21496, https://doi.org/10.5194/egusphere-egu2020-21496, 2020.

Tropical peat swamp forest (PSF) is a unique ecosystem rich in carbon and water, which is widely distributed in Southeast Asia’s coastal lowlands, mainly in Borneo, Sumatra and Malay Peninsular. The ecosystem has accumulated a huge amount of organic carbon in peat soil over millennia under the condition of high groundwater level. However, PSF has been reduced and degraded by logging, drainage and burning mainly because of land conversion to oil palm and pulp wood plantations during the last two decades. Such human disturbances potentially increase carbon dioxide (CO₂) emissions to the atmosphere through enhanced oxidative peat decomposition and the increased risk of peat fires. Thus, it is essentail to assess the current carbon status of tropical peatlands and quantify the effects of disturbance on the carbon balance to understand the role of tropical peatlands in the regional and global carbon balances. We have continuously measured ecosystem-scale eddy fluxes and soil fluxes of CO₂ and methane (CH₄) in different tropical peat ecosystems, including a little drained PSF, a drained PSF, a burned ex-PSF and an oil palm plantation, in Central Kalimantan, Indonesia, and Sarawak, Malaysia, in Borneo. Based on the monitoring data, I’ll talk about the carbon balance of tropical peat ecosystems, such as its seasonal variation and its relationship with groundwwater level, and the effect of disturbance due to human activities and ENSO drought on the carbon flux and balance.

How to cite: Hirano, T.: Carbon balance of tropical peat ecosystems in Borneo, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-6405, https://doi.org/10.5194/egusphere-egu2020-6405, 2020.

Smoke from peatland wildfires contributes significantly to global greenhouse gas (GHG) emissions, while reactive gases and particulates cause transboundary haze episodes. Haze is the large-scale accumulation of smoke at low altitudes, especially frequent in Southeast Asia during dry periods. Understanding emissions from peatland fires plays a vital role in calculating GHG budgets, forecasting haze events and modelling future climate change. However, only a handful of field studies or laboratory experiments on tropical peat fire smoke have been undertaken to date. Of the few studies that have investigated tropical peatland fire emissions, there exists substantial inter-study variabilities of emission factors (EFs) with some gas emission factors varying by a factor of 10 between studies. Explaining the nature of such variability remains a challenge. In August/September 2018 in Riau, Indonesia, we carried out the first field-scale experimental burn on a tropical peatland (the GAMBUT Workshop), aiming to understand how fires ignite, how they spread, and how emissions vary across the life-cycle of a peatland fire. Our site was a heavily degraded tropical peatland subjected to long-term drainage, logging, and agricultural conversion. Here we present the field measurements of gas emissions from the fire experiment. Open-path Fourier transform infrared spectroscopy (OP-FTIR) was used to retrieve mole fractions of 13 gas species. EFs from 40 measurement sessions over two weeks of burning during different fire stages (e.g., slash and burn ignition, smouldering spread or suppression) and weather events (e.g., wind or rainfall) were calculated and reported. We present field evidence to indicate that EFs vary significantly among fire stages and weather events. Heterogenous physicochemical properties of our peatland site (e.g. moisture content, inorganic content and bulk density) were also found to affect the EFs. We discuss the implications for air quality forecasting, suggesting the necessity for more complex mapping of peatland heterogeneity/land-use for emissions inventories and temporally variable emissions factors, depending on the time since the initiation of a fire event.

How to cite: Hu, Y., Smith, T. E. L., Santoso, M. A., Amin, H. M. F., Christensen, E. G., Cui, W., Purnomo, D. M. J., Palamba, P., Nugroho, Y. S., and Rein, G.: Temporal variability of greenhouse gas and reactive gas emission factors during a two-week-long tropical peatland experimental burn, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-19872, https://doi.org/10.5194/egusphere-egu2020-19872, 2020.

Drainage canal networks associated with agricultural land use are a major contributor to peatland degradation in Southeast Asia. These canals are used to control water table depth and make the soil suitable for planting, but their presence has the negative impact of drying out peat soils near the ground surface. Drier soils in turn cause elevated fire risk, increased carbon release to the atmosphere, and subsidence. Although canals directly impact local peat hydrology, the effect of drainage intensity (i.e. canal density) in peatlands has not been quantitatively investigated, due to a lack of reliable canal maps in the region.

In this study, we trained a machine learning model to identify drainage canals and map their density throughout Southeast Asian peatlands using remote sensing imagery. Specifically, a fully convolutional neural network was applied to RGB 5m resolution Basemap imagery from Planet. Training data was generated by hand-labeling canals from satellite images, and validation of canal density was performed via comparison to independently labeled maps. A map of canal density was then produced across ISEA peatlands using images from 2017. We compared canal density with land use type and found that mean canal density is highest in industrial plantations. We also compared canal density with fire occurrence and subsidence data. This new dataset has potential applications for studies of peatland hydrology, land use change, and fire risk.

How to cite: Dadap, N., Cobb, A., Hoyt, A., Rao, K., Harvey, C., and Konings, A.: Mapping Drainage Canals in Southeast Asian Peatlands and their Implications for Peatland Degradation, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-12505, https://doi.org/10.5194/egusphere-egu2020-12505, 2020.

The precise quantification of peat deposits at local to global scale is of key importance for the implementation of adequate conservation policies of peatlands. To this end, new remote sensing applications are needed, that provide high resolution data sets at regional scale. In this presentation, we present the results obtained using Airborne Electromagnetics (AEM) to estimate peat thickness and carbon content of a large peatland site located in Indonesia. The effectiveness of the AEM method for assessing peat thickness and volumes, and in turn carbon stocks, is tested by comparing the results to ground-truth measurements. Our results show that the AEM method can detect both the top and the bottom of a peatland profile over a clay substrate at high spatial resolution, allowing for an accurate three-dimensional morphological description of the peat body. The AEM method performs extremely well along the flight lines, where the instrument clearly detects the peat layer and differentiates it from the underlying mineral substrate. Moving away from the flight lines, the accuracy slightly decreases because the interpolation of the AEM data does not fully capture the highly variable morphology of the peat bottom. We conclude that in varying dome conditions, a high flight line density is preferable to describe the spatial distribution of the peat layer. Once the volume of the peatland is determined, the average organic carbon content by soil volume retrieved from field campaigns and laboratory analyses is used to estimate the total organic carbon stored in the peatland.

The results obtained with the AEM method are compared to those obtained with an empirical method that uses the soil topography to predict the thickness of the peatland. This empirical approach is based on the analyses of several previous studies available from the literature that show how it is common for some dome-shaped peatlands to present a linear correlation between peat thickness and soil topography. In this study, we show that the linear correlation is site-specific, and when used for prediction purposes, it provides incorrect peat volume estimates when it is extended to other sites or over large territories. When compared to the AEM method, our results show that the AEM method is superior in detecting the peat morphology and volume.

How to cite: Silvestri, S., Knight, R., Viezzoli, A., Richardson, C. J., Anshari, G. Z., Dewar, N., Flanagan, N., and Comas, X.: Quantification of the organic carbon pool of a large Indonesian peatland using an airborne geophysical method and comparison to an empirical topographic approach, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-3432, https://doi.org/10.5194/egusphere-egu2020-3432, 2020.

Southeast Asian peatlands, one-third of global tropical peatlands, have sequestered and preserved gigatons of carbon in the past thousands of year. Rainfall fluctuation on yearly and even hourly timescales plays an important role that defines peat carbon accumulation or loss from tropical peatlands. Notably, research related to the ecosystem-scale carbon exchange, including methane (CH₄), over tropical peatland ecosystems remains limited. Given their significant carbon stocks, the fate of natural tropical peatlands under current and future climate is unknown.

We performed a study in Kampar Peninsula, a coastal tropical peatland of around 700,000 ha, in Sumatra, Indonesia. This ombrotrophic (acidic and nutrient-poor) peatland largely formed within the past 8000 years. The peninsula is characterized by a large, relatively intact central forest area surrounded by a mosaic of smallholder agricultural land, and industrial fiber wood plantation, smaller secondary forest areas, and undeveloped open and degraded land. We measured the net ecosystem CO₂ and CH₄ exchanges between natural peatland and the atmosphere using the eddy covariance technique over two years (June 2017-May 2019). In addition, peat subsidence rates were measured using polyvinyl chloride poles at every 1 km along 35 km long transect across the natural forest in the peninsula. In the natural forest, groundwater level shows periodic sharp rises and steady decreases corresponding to rain events. The groundwater level can rise up to 20 cm above the peat surface in the wet season, and then in the late dry season can reach -70 cm.

Our measurements indicate that the natural tropical peatland functioned as a significant source of CO₂ (410±60 g CO₂-C m^-2 year^-1) and CH₄ (6.8±0.7 g CH₄-C m^-2 year^-1) to the atmosphere. If we follow IPCC global warming potential (GWP) accounting methodology and apply a 100-year GWP of 34 for CH₄, this implies that CH₄ emissions contributed ~35% of the 100-year net warming impact. Carbon emissions (due to oxidation of peat, litterfall and coarse wood debris) contributed ~30-35% of the observed subsidence rates. The CO₂ exchanges increased linearly as groundwater level declined. Lower groundwater level enhances peat aeration and potentially increases oxidative peat decomposition, which results in higher CO₂ emissions. The CH₄ exchanges decreased exponentially as groundwater level declined.

The results indicate that tropical peatland ecosystems are no longer a carbon sink under the current climate. Our results, which are among the first eddy covariance exchange data reported for any tropical peatland, should help to reduce the uncertainty in the estimation of CO₂ and CH₄ emissions from a globally important ecosystem and improve our understanding of the role of natural tropical peatlands under current and future climate.

How to cite: Deshmukh, C. S., Julius, D., Nardi, N., Putra Susanto, A., and Nurholis, N.: Significant carbon loss from a natural tropical peatland under current climate , EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-11941, https://doi.org/10.5194/egusphere-egu2020-11941, 2020.

Assessing the flux of carbon (C) from terrestrial ecosystems to the atmosphere represents a critical element of global carbon budgeting. In tropical peatlands this has been a fundamental part of assessing the impact of land use change on an ecosystem that represents a significant global carbon store, with peat accumulation being often many meters deep. These systems have formed over thousands of years as a function of incomplete decomposition of organic matter from water-logged swamp forests. However, intact tropical peat swamp forests (PSFs) are under increasing threat from agricultural conversion, deforestation, drainage practices and fires. The resultant alteration of the peat soil results in peat oxidation, increased rates of organic matter decomposition and greenhouse gas (GHG) emissions. Consequently, these peats are reverting from C stores to sources.

Radiocarbon (¹⁴C) abundance can be used to assess C cycling rates in varied ecosystems and identify rapid or slow C turnover rates from years to centuries, as well as shifts in cycling rates – for example with land use or hydrological alteration. Within intact peatlands, deep peats generally contain an increasing abundance of ¹⁴C depleted content due to radioactive decay, conversely, shallower peats are more abundant in recently produced organic litter enriched with “Bomb C”; derived from nuclear testing in the 1960s. Similarly, root derived organic matter and the associated root respiration (autotrophic respiration) also have signatures resembling recent atmospheres, whereas microbial respiration of soil organic matter (heterotrophic respiration) will resemble the mean age of the soil carbon being utilised by the microbial community, and as such can be a tracer for sources of carbon being decomposed. 

Yet while an increasing body of knowledge exists on tropical peatland carbon flux rates or net ecosystem respiration in association with land-use change, these approaches fail to delineate the sources of carbon being used within the soil profile and thus fully address questions linked to changing carbon cycling rates with land use change.

Here we provide what we believe to be the first data on ¹⁴CO₂ fluxes from tropical peatland soils in relation to varying land use classes with the aim of determining if peats which were previously long-terms C stores are being utilised within short, fast C cycles and thus contributing to modern GHG budgets. CO₂ flux rates were measured using soil chambers and emitted CO₂ was subsequently trapped on a zeolite molecular sieve cartridge. An aliquot of the recovered CO₂ was graphitised and analysed for ¹⁴C by accelerator mass spectrometry. Associated soil age profiles were also determined.

Results indicate significant fluxes of multi-millennia old carbon from peatlands under altered land use classes and clear evidence for a shift to C cycling speed, with previously long-term stored C contributing to modern C budgets. Result highlight the instability of the peat profile under altered land-use classes and minimal to no contribution of modern C from recently produced organic matter to these carbon budgets. Findings clearly indicate the unsustainability of these agricultural practices and the need for burn- and drain-free land-use strategies.

How to cite: Evers, S., Smith, T., Garnett, M., Dhandipani, S., and Lupascu, M.: Millennia-old carbon fluxes from degraded tropical peatland soils , EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-21851, https://doi.org/10.5194/egusphere-egu2020-21851, 2020.

Deforestation followed by draining of tropical peat swamp forests are the most common disturbance regimes before further land development takes place. The severity and extent depend on a number of drivers that dictate how restoration should be performed. The permanent plots were established to monitor total ecosystem carbon stocks and greenhouse gases (CO₂, CH₄, N₂O) emissions from peatland under different vegetation cover, namely forest tree species, oil pal, and rubber plantations, and evaluate the effect of rewetting by blocking the drainage canals on GHG fluxes. We found that the mean total ecosystem carbon stocks at a reforested area, rubber and oil palm were 3983 + 318 Mg C ha^-1, 3363 + 207 Mg C ha^-1 and 3523 + 253 Mg C ha^-1 respectively. The average total soil emission of CO₂ during conditions before canal blocking in reforested areas, oil palm and rubber plantations were 10.93 Mg CO₂ ha^-1yr^-1, 16.66 Mg CO₂ ha^-1yr^-1 and 23.70 Mg CO₂ ha^-1yr^-1. After the canals were blocked, the average total CO2 emissions were 3.57 Mg CO₂ ha^-1yr^-1 in reforested area, 10.47 Mg CO₂ ha^-1yr^-1 in oil palm and 15.27 57 Mg CO₂ ha^-1yr^-1 in rubber plantation.

Methane (CH4) flux before blocking were (-0.10 + 0.84), (0.34 + 4.52), and (0.50 + 2.70) mg m^-2 hr^-1 in reforested area, oil palm and rubber plantation respectively, while the fluxes after blocking were (8.02 + 3.28), (5.36 + 6.13), and (0.64 + 1.19) mg m^-2 hr^-1 respectively. The increasing trends after blocking suggests that methanogenic bacteria were active in anaerobic. On the other hand, N2O decreased from (0.40 + 0.84), (0.40 + 0.84), and 0.40 + 0.84) mg m^-2 hr^-1 in forested area, oil palm and rubber plantations to (-0.20 + 0.27), (-0.45 + 2.08), and (2.15 + 0.25) mg m^-2 hr^-1 respectively.

How to cite: Murdiyarso, D., Lestari, I. L., Taufik, M., and Santikayasa, P.: Effects of rewetting of peatlands on GHG fluxes from the soils with different land-cover types, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-22269, https://doi.org/10.5194/egusphere-egu2020-22269, 2020.

Selective logging is the practice of extracting selected commercial trees from natural production forests. The intensity of logging correlates with a reduction in biodiversity, wood production and biomass stocks. Less is known about the relationship of logging to soil organic carbon (SOC) and how it changes or recovers over time. Empirical measurements in Borneo provided SOC, soil respiration, aboveground and belowground net primary productivity (NPP) from intact old-growth forest (OGF) as well as from moderately to heavily logged (LOG) forest sites. Soil carbon (C) content and heterotrophic respiration (R_h) was higher in LOG sites than in OGF sites. Moderately logged forest (logged > 10 years ago) contained more SOC than heavily logged forest (logged approx. 7 years ago). NPP was used to estimate the C input to the soil. All these data were used to test the biochemical model ECOSSE (Estimating Carbon in Organic Soils – Sequestration and Emissions) to calculate SOC for the study sites. The model performed well in simulating the soil respiration of OGF and generated acceptable results for LOG sites in the validation process. The results for logged forests showed an increase in R_h over the first 15 years, with some sites showing either a further increase over the next 15 years or stabilizing at a higher level compared to pre-disturbance conditions for other sites. However, for all modelled cases, a break was observed after 30 years, when R_h decreased to a lower level (but not as low as for OGF) before reaching a new equilibrium. At the same time, SOC begins to increase. Spatial modelling showed the results for Borneo under logged conditions and the potential of storing C if logging was reduced. Only 22% of Borneo is under old-growth forest; the results show moderate to high C losses if this region is subjected to logging. Overall, the results show the disturbance of SOC and Rh through logging over periods longer than 30 years.

How to cite: Vetter, S. H., Teh, Y. A., Martin, M., Elias, D. M. O., Riutta, T., and Smith, P.: Impacts of logging on soil organic carbon and heterotrophic respiration in tropical forests in Borneo , EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-10276, https://doi.org/10.5194/egusphere-egu2020-10276, 2020.

We report on two years of continuous monitoring of methane (CH₄) and carbon dioxide (CO₂) emissions at two contrasting sites in the Okavango Delta, North-Western Botswana, an inland delta bordered by the Kalahari Desert. Approximately 60% of the annual water influx into the Okavango Delta results from seasonal river discharges originating in the Angolan Highlands, and the remainder comes from direct rainfall. 96-98% of the 16.1 billion m³ entering the Delta annually are lost through evapo-transpiration (1500 mm.year^-1). Flooding is gradual and it takes the pulsed influx ca. 4-5 months to travel the 250 km separating the inlet in Mohembo from the main outlet in Maun. The wetlands of the Okavango Delta are in pristine condition and can be separated into three categories: permanently flooded, seasonally flooded (3-6 months per year) and occasionally flooded (typically once per decade). 

Two eddy-covariance systems were set up in August 2017, one at Guma Lagoon (18°57'53.01" S;  22°22'16.20" E) at the edge of an extensive papyrus bed in the permanently-flooded section of the delta, and the second one at Nxaraga on the SW edge of Chief’s Island (19°32'53'' S; 23°10'45'' E) in the seasonal floodplain. In addition, monthly measurements of methane and carbon dioxide fluxes were taken using a clear dynamic chamber at the Nxaraga site along transects chosen to span the natural soil moisture gradient (very dry to waterlogged soils).

The emissions of methane exhibited contrasting spatial and temporal patterns between sites. At the seasonal wetland, very low fluxes of CH₄ were typically observed from January to June. Emissions increased abruptly from July-August onwards after flood waters rewetted the flooplain in that area of the Delta. Throughout the year, local emission hotspots of CH₄ were observed along the vegetated river channels within the flux footprint of the eddy-covariance system, whereas CH₄ oxidation was recorded in persistently dry areas where the soil is sandy and salt-crusted. The chamber measurements corroborated the findings of the eddy-covariance measurements and soil moisture is likely the dominant control of methane fluxes at the seasonal wetland.

The methane emissions from the floating papyrus mat in the permanent wetland exhibited a marked seasonal cycle, characterised by relatively high emissions (of the order of 250 nmol.m^-2.s^-1; 2.5 larger than peak emissions recorded at the seasonal wetland) in the summer months (November-March) and minimum emissions in winter (typically 50 nmol.m^-2.s^-1 in June-August). At the seasonal timescale, methane emissions were strongly correlated to the phenological cycle of papyrus (lowest emissions during the senescence phase), suggesting that plant-mediated transport is the dominant control. The annual budgets of CH₄ and CO₂ in the permanent wetland were estimated at 153.4 ± 27.9 tons.km^-2 (3835.0 ± 697.5 CO₂-eq) and -874.0 ± 200.4 tons.km^-2 respectively, making the permanent wetland a potent net source of carbon to the atmosphere.

How to cite: Helfter, C., Gondwe, M., Murray-Hudson, M., and Skiba, U.: Methane and carbon dioxide emissions from two contrasting wetlands in the Okavango Delta, Botswana., EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-6023, https://doi.org/10.5194/egusphere-egu2020-6023, 2020.

Greenhouse gas (GHG) emission estimates from tropical African rivers are underrepresented in global datasets, resulting in uncertainties in their contributions to global emissions. To better constrain the contribution of rivers and streams to GHG emissions from tropical landscapes and to determine possible underlying controlling processes, we implemented a monthly synoptic sampling program from January 2019 – December 2019, in which CO₂, CH₄ and N₂O concentrations and fluxes, along with water quality and sediment parameters were measured from 60 river sites in the upper and middle catchments of the Mara River in Kenya (~8450 km²).

Consistent with previous studies, Mara basin streams and rivers were mostly sources of GHGs, and were comparable to previous studies in tropical and temperate regions. Based on CO₂ equivalents, CO₂ accounted for >60% of the emissions, while CH₄ and N₂O (<35%) were minor contributors. There were higher mean values of CO₂ and N₂O fluxes in streams draining croplands (92±9 CO₂ mmol m^-2 d^-1 and 14±2 N₂O µmol m^-2 d^-1) compared to those draining forested areas (45±5 CO₂ mmol m^-2 d^-1 and 3±0.6 N₂O µmol m^-2 d^-1). CH₄ fluxes showed no significant variation with land use. CO₂ and CH₄ concentrations had a negative correlation with dissolved oxygen (DO) and a positive correlation with dissolved organic carbon (DOC) and fine benthic organic matter (FBOM), while N₂O was positively correlated to nitrate (NO₃-N) and negatively correlated to DO. Based on the significant relationships of all three gases with DO and their substrates, we inferred that GHG concentrations were mainly controlled by in-stream biogeochemical processes - i.e. methanogenesis for CH_4, net heterotrophy for CO₂ and coupled nitrification-denitrification for N₂O. Changes in discharge, driven by precipitation events, significantly accounted for the seasonal variation in GHGs concentration and flux, with clear differences between the driest months (March and April) and the wettest (October-December). During low-discharge periods, streams were characterized by lower DO, lower nitrate NO₃-N, higher DOC, and higher FBOM concentrations compared to the wet season. This resulted in significantly higher CH₄ and CO₂ concentrations, which could be attributed to increased in-stream production through the aforementioned processes as a result of increased water residence times. In contrast, N₂O concentrations in the dry season were lower than in the wet season, indicating that due to low DO and NO₃-N concentrations, produced N₂O may have been further reduced to N₂ during denitrification. However, as fluxes are a function of both concentration and the discharge-related gas transfer velocity (k), all GHG’s exhibited higher fluxes in the wet season compared to the dry season. Mean monthly CO₂ and N₂O concentrations also responded positively to discharge, suggesting that terrestrial inputs could also account for higher fluxes during the wet season.

In future studies, we therefore plan to incorporate process measurements (e.g. nitrification, denitrification and ecosystem metabolism) across seasons in conjunction with measurements of GHG fluxes and environmental parameters. This will allow to a) elucidate the importance of in-stream production versus terrestrial inputs as controls of fluxes of GHGs and to b) attribute observed fluxes to specific biogeochemical processes.

How to cite: Mwanake, R., Gettel, G., Butterbach-Bahl, K., and Kiese, R.: Seasonal variation of CO2, CH4 and N2O fluxes from tropical streams and rivers under forest and cropland landuses: A case study of the Mara river basin in Kenya, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-20112, https://doi.org/10.5194/egusphere-egu2020-20112, 2020.

Agro-silvo-pastoralism is a highly representative Land Use in Africa, often presented as a strategical option for ecological intensification of cropping systems towards food security and sovereignty.

We set up a new long-term observatory (“Faidherbia-Flux”) to monitor and model microclimate, energy and C balance in Niakhar (central Senegal, rainfall ~ 500 mm), dominated by the multipurpose tree Faidherbia albida (12.5 m high; 7 tree ha^-1; 5% canopy cover). Faidherbia is an attractive agroforestry tree species in order to partition fluxes, given that it is on leaf during the dry season (October-June) and defoliated during the wet season, just when crops take over. Pearl-millet and groundnut crops were conducted during the wet season, following annual rotation in a complex mixed mosaic of ca. 1 ha fields.

Early 2018, we installed an eddy-covariance (EC) tower above the whole mosaic (EC1: 20 m high). A second EC system was displayed above the crop (EC2: 4.5 m if pearl-millet, 2.5 m if groundnut) in order to partition ecosystem EC fluxes between tree layer and crop+soil layers. Sap-flow was monitored from April 2019 onwards in 5 faidherbia trees (37 sensors).

The ecosystem displayed moderate but significant daily CO₂ and H₂O fluxes during the dry season, when faidherbia (low canopy cover) was in leaf and the soil was evaporating. At the onset of the rainy season, the soil bursted a large amount of CO₂. Just after the growth of pearl-millet in 2018, CO₂ uptake by photosynthesis increased dramatically. However, this was largely compensated by high ecosystem respiration. Surprisingly in 2019, although the crop was turned to groundnut, the fluxes behaved pretty much the same as with pearl millet in 2018: comparing annual balances between 2018 and 2019 we obtained [454, 513] for rainfall (P: mm yr^-1), [3500, 3486] for potential evapotranspiration (ETo: mm yr^-1), [0.13, 0.15] for P/ETo, [470, 497] for actual evapotranspiration (E: mm yr^-1), [2809, 2785] for net radiation (R_n: MJ m^-2 yr^-1), [1686, 1645] for sensible heat flux (H: MJ m^-2 yr^-1),  [-3.2, -2.8] for net ecosystem exchange of C (NEE: tC ha^-1 yr^-1), [-11.8, -11.1] for gross primary productivity (GPP: tC ha^-1 yr^-1) and [8.6, 8.3] for ecosystem respiration (R_e: tC ha^-1 yr^-1). The energy balance (R_n-H-LE) was nearly nil indicating that the EC system behaved reasonably. E was very close to P, indicating that little or no water would recharge the deep soil layers.

Now comparing the dry (2/3 of the year) and wet (1/3) seasons: surprisingly, NEE was more effective during the dry season [-3.9, -1.7]. This was the result of R_e being much lower on a daily basis as well as cumulated over the entire seasons [57, 84], whereas GPP was similar [-10.8, -12.1].

We found a good match between E measured above the whole ecosystem (EC1), and the sum of tree transpiration (T, measured by sapflow) + E measured just above crops + soil (EC2) throughout the wet and dry seasons.

The “Faidherbia-Flux” observatory is registered in FLUXNET as SN-Nkr and is widely open for collaboration.

How to cite: Roupsard, O., Frederic, D., Alain, R., Christophe, J., Didier, O., Laure, T., Sidy, S., Waly, F., Djimm M.L., D., Khalisse, D., Yélognissé, A., Seydou, D., Serigne, F., Mame Sokhna, S., Diaminatou, S., Guerric, L. M., Remi, V., Josiane, S., and Laurent, C.: More C uptake during the dry season? The case of a semi-arid agro-silvo-pastoral ecosystem dominated by Faidherbia albida, a tree with reverse phenology (Senegal), EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-11203, https://doi.org/10.5194/egusphere-egu2020-11203, 2020.

Soil organic carbon (SOC) is a key component of terrestrial ecosystems. Experimental studies have shown that soil texture and geochemistry have a strong effect on carbon stocks. However, those findings primarily rely on data from temperate regions or use model approaches that are often based on limited data from tropical and sub-tropical regions.

Here, we evaluate the controls on soil carbon stocks in Africa, using a dataset of 1,580 samples. These were collected across Sub-Saharan Africa (SSA) within the framework of the Africa Soil Information Service (AfSIS) project, which was built on the well-established Land Degradation Surveillance Framework (LDSF). Samples were taken from two depths (0–20 cm and 20–50 cm) at 46 LDSF sites that were stratified according to Koeppen-Geiger climate zones. The different pH-values, clay content, exchangeable cations and extractable elements across various soils of the different climatic zones (i.e. from arid to humid (sub)tropical) allow us to identify different soil and climate parameters that best explain SOC variance across SSA.

We tested if these SOC predictors differed across climatological conditions, using the ratio of potential evapotranspiration (PET) to mean annual precipitation (MAP) as indicator. For water-limited regions (PET/MAP > 1), the best predictors were climatic variables, likely because of their effect on the quantity of carbon inputs. Geochemistry dominated SOC storage in energy-limited systems (PET/MAP < 1), reflecting its effect on carbon protection. On a continental scale, climate (e.g. PET) is key to predicting SOC content in topsoil, whereas geochemistry, particularly iron-oxyhydroxides and aluminum-oxides, is more important in subsoil. Clay content had little influence on SOC at both depths. These findings contribute to an improved understanding of the controls on SOC stocks in tropical and sub-tropical regions.

How to cite: von Fromm, S. F., Hoyt, A. M., Asefaw Berhe, A., Shepherd, K. D., Vågen, T.-G., Winowiecki, L. A., Desta, L. T., Tondoh, J. E., Sila, A. M., Towett, E. K., Weullow, E., Aynekulu, E., Six, J., Trumbore, S. E., and Doetterl, S.: Large-scale controls of soil organic carbon in (sub)tropical soils, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-13719, https://doi.org/10.5194/egusphere-egu2020-13719, 2020.

Highly weathered soils depleted in minerals and phosphorus (P) support large tracts of the tropical rainforests in the Central Amazon, which significantly contribute to the global carbon (C) sink. In these soils (oxisols and ferrasols), P is either occluded in Al/Fe-oxides, bound to the soil mineral matrix or in soil organic matter, and therefore not directly available for uptake as inorganic phosphate (Pi). To liberate Pi for plant or microbial uptake two processes are key: (i) changes of sorption-desorption equilibria of Pi with the soil matrix and (ii) the release of Pi from organic compounds (Po) catalyzed by enzymes, such as phosphatases. Plant roots and soil microbes have developed strategies to stimulate the release of P by accelerating P dissolution and desorption and by releasing extracellular phosphatases into the soil environment, which requires however C and energy investment. Because of P limitation in this ecosystem, the relative contributions of abiotic and biotic controls over P mineralization is of pivotal importance. Yet conclusive results are still scarce.

We therefore aimed to disentangle abiotic and biotic controls over P mineralization in tropical soils. To achieve this, we collected forest soils from the Amazon Basin covering a range of soil texture and P concentrations, determined soil mineralogy and measured gross P desorption and mineralization rates using a ³³P isotope pool dilution assay. Moreover, we determined acid phosphatase activity rates and microbial biomass C and P. We found significant differences between the studied sites in gross P influx and efflux rates into the Pi pool. Gross influx rates (i.e. the sum of Pi desorption and organic P mineralization) exceeded efflux (i.e. sorption or biotic Pi uptake rates) only in sandy and silty soils, while in clayey soils efflux rates dominated P fluxes indicating a very high Pi sorption capacity. However, gross influx and efflux rates were not related to total or dissolved P. Microbial biomass and acid phosphatase activity normalized to microbial biomass C were highest in sites with overall low total P microbial biomass P accounting for up to 40 % of total P in low P soils. We therefore conclude that in low P soils organic P turnover plays a major role in soil P cycling, and despite of the high P sorption capacity of clay rich soils, microbes can be strong competitors for plant available P.

How to cite: Fuchslueger, L., Zezula, D., Püspök, J., Van Langenhove, L., Margalef, O., Canarini, A., Ranits, C., Quesada, C. A., Salinas, N., Cosio, E., Penuelas, J., Wanek, W., Richter, A., and Janssens, I.: Controls over phosphorus mineralization and immobilization rates in different tropical soils, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-15473, https://doi.org/10.5194/egusphere-egu2020-15473, 2020.

Variability in precipitation and temperature are key markers of climate change. Extreme events like heat waves, droughts, frosts, wind storms, flooding rains and fires greatly affect ecosystem and terrestrial carbon balance. Tropical regions in particular make strong contributions to the global carbon cycle and are the focus of our research. Our initial analysis confirmed the long-known pattern of large variability in rainfall in the tropical southern hemisphere (i.e. between the Tropic of Capricorn and the Equator) w.r.t. the north, with less variation in temperature between the hemispheres. In the follow-up analysis, we focus on exchanges of carbon and water and water use efficiency, based on 39 eddy covariance flux sites which represent 25 years of data across the tropics. Our working hypothesis is that long-term increases in temperature and significant changes (+/-) in rainfall will be reflected in changes in water use efficiency and cropping period, albeit with greater spatial and temporal variation in the south than in the north. We are also investigating relationships between water use efficiency of tropical regions calculated using eddy covariance flux data, with that calculated using tree ring data. We seek to combine methodologies that can help drive our understanding of the impact of climate change on water use efficiency of tropical regions.

Keywords: Eddy covariance, Tropics, Water use efficiency, Carbon cycle, Tree ring data

How to cite: Kumari, S. and Adams, M. A.: Impact of Climate change on tropical terrestrial water use efficiency, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-12512, https://doi.org/10.5194/egusphere-egu2020-12512, 2020.

The Gross Primary Production (GPP) in tropical terrestrial ecosystems plays a critical role in the global carbon cycle and climate change. The strong 2015–2016 El Niño event offers a unique opportunity to investigate how GPP in the tropical terrestrial ecosystems responds to climatic forcing. This study uses two GPP products and concurrent climate data to investigate the GPP anomalies and their underlying causes. We find that both GPP products show an enhanced GPP in 2015 for the tropical terrestrial ecosystem as a whole relative to the multi-year mean of 2001–2015, and this enhancement is the net result of GPP increase in tropical forests and decrease in non-forests. We show that the increased GPP in tropical forests during the El Nino event is consistent with increased photosynthesis active radiation as a result of a reduction in clouds, while the decreased GPP in non-forests is consistent with increased water stress as a result of a reduction of precipitation and an increase of temperature. These results reveal the strong coupling of ecosystem and climate that is different in forest and non-forest ecosystems, and provide a test case for carbon cycle parameterization and carbon-climate feedback simulation in models.

How to cite: Zhu, J., Zhang, M., Zhang, Y., Zeng, X., and Xiao, X.: Response of tropical terrestrial gross primary production to the super El Niño event in 2015, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-13872, https://doi.org/10.5194/egusphere-egu2020-13872, 2020.

Tropical forest degradation through selective logging, fragmentation, and understory fires substantially changes forest structure and composition.  In the Amazon, degradation is as widespread as deforestation; however, studies addressing the effects of forest degradation on tropical ecosystem functions are scarce. Here, we integrate small-footprint airborne lidar over the Brazilian Amazon (> 250,000 ha), collected between 2016–2018, with recent ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) land surface temperature and evapotranspiration products (70-m resolution, data acquired in 2018–2019) to investigate the role of forest structure, forest fragmentation, and disturbance history on dry-season land surface temperature and evapotranspiration.  During the dry season, degraded forests, especially those affected by multiple degradation events, are significantly warmer (up to 9.3°C) and show reduced evapotranspiration (10% less than intact forests). Likewise, forest near the edges (< 350m) experience the greatest warming (up to 6.5°C) and the greatest reduction (9%) in evapotranspiration. We also used the airborne lidar dataset to initialize the Ecosystem Demography Model (ED-2.2) to investigate the impact of degradation on the gross primary production (GPP), evapotranspiration (ET), and sensible heat flux (H) under a broader range of climate conditions, including severe droughts. Consistent with ECOSTRESS, the simulations during the dry season in typical years showed that severely degraded forests experienced water-stress with declines in ET (34% reduction), GPP (35% reduction), and increases of H (43% increases) and daily mean ground temperatures (up to 6.5°C) relative to intact forests.  In the model, the simulated changes are mostly driven by increased below-ground water stress, which can be attributed to the shallower rooting profile of degraded forests. However, relative to intact forest, the impact of degradation on energy, water, and carbon cycles markedly diminishes under extreme droughts such as 2015–2016, when all forests experience severe stress. Our results indicate the potentially important role of tropical forest degradation changing the carbon, water, and energy cycles in the Amazon, and consequently a much broader influence of land use activities on functioning of tropical ecosystems.

How to cite: Longo, M., Saatchi, S., Keller, M., Bowman, K., Ferraz, A., Cawse-Nicholson, K., Fisher, J., Pinagé, E., Moorcroft, P., Ometto, J., and Morton, D.: The impacts of Amazon forest degradation and fragmentation on energy, water, and carbon cycles, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-4801, https://doi.org/10.5194/egusphere-egu2020-4801, 2020.

Land use and land cover change in the Amazon results in the loss and degradation of ecosystem services vital to human wellbeing. The land-use transitions from forest to grasslands and to croplands modify the hydrological cycle as the non-forest cover has lower evapotranspiration and increased runoff.
The high rates of evapotranspiration of the Amazon forest drive the atmospheric moisture recycling system, which not only supports the forest itself but provides atmospheric moisture for precipitation downwind, important for agriculture, human consumption and hydropower across central Brazil. While deforestation reduces overall precipitation,  deforestation has also been correlated with a delay in the wet season onset leading to a longer dry season. Therefore agriculture presents itself as an interesting conundrum, as it is the main driver of deforestation, it also acts as both the degrader and one of the main beneficiaries of the system.

Recent advances in soybean double-cropping have increased agricultural productivity. However, as sowing is tightly coupled to the wet season onset, this strategy is dependent on a stable wet season onset.

Here, we analyse the contribution of terrestrial evapotranspiration to precipitation during the early wet season. We employed a Lagrangian moisture transport model which connects moisture source (evapotranspiration) locations with moisture sink (precipitation) locations in the agriculture state of Mato Grosso, Brazil. We calculated the fraction of precipitation derived from moisture recycling as well as estimate the delay in wet season precipitation under a scenario without moisture recycling. Finally, using this moisture transport model we identified and mapped source areas that contribute to two existing double-cropping locations, one in the Amazon biome (North) and one in the Cerrado biome (South).

We found that during the wet season transition, roughly 35% of the precipitation across Mato Grosso originates from moisture recycling. The fraction of moisture recycled precipitation is spatially correlated with latitude and longitude with the lowest fraction in the Northeast ≈20% and highest in the Southwest ≈60%. Both cropping locations showed a highly dispersed source area of precipitation. With 30% of recycled rainfall generated within 250 km of the precipitation location. The two cropping locations we analyzed shared a number of forest source areas highlighting their importance for moisture recycling. We found a 10-day delay in accumulated precipitation in our scenario without moisture recycling. This implies that double-cropping systems would become more uncertain as the sowing of soybean would most likely be delayed further into the year.

How to cite: O'Connor, J., T. Rebel, K., J. Santos, M., C. Dekker, S., and A. Tuinenburg, O.: Double cropping in the Amazon: its relation with moisture recycling, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-13755, https://doi.org/10.5194/egusphere-egu2020-13755, 2020.

Forest play a major role in the global carbon cycle storing large amounts of carbon in both living and dead organic matter. Forests can be either a sink or source of carbon depending on the net of far larger fluxes of carbon into (photosynthesis) and out of (mortality, decomposition and disturbance) forest ecosystems. Due to the potential for substantial accumulation of carbon in forests, has led to nationally determined commitments (NDCs) by Governments across the world to protect existing and plant large areas of new forest. However, significant uncertainty remains in our understanding of current forest carbon cycling, especially mortality and decomposition processes, and how carbon cycling will change under climate change. These uncertainties present two connected challenges to effective forest protection and new planting; (i) which existing forests are under the greatest risk to climate change and (ii) where are the most climate safe locations for new forest planting to maximise carbon accumulation.

Here we combine a terrestrial ecosystem model of intermediate complexity (DALEC) with Earth observation (e.g. leaf area, biomass, disturbance) and databased information (soil texture and carbon stocks) within a Bayesian model-data fusion framework (CARDAMOM) to retrieve location specific carbon cycle analyse (i.e. parameter retrievals) across Brazil at 0.5 x 0.5 degree spatial resolution between 2001 and 2015. CARDAMOM allows us to retrieve, independently for each location analysed, an ensemble of parameters for DALEC which are consistent with the location specific observational constraints and their uncertainties. These ensembles give us multiple potential, but observation consistent, realisations of forest carbon cycling and ecosystem traits. We directly quantify our uncertainty in forest carbon cycling and ecosystem traits from these ensembles. The DALEC parameterisations are then simulated into the future under a range of climate scenarios from the CMIP6 model dataset. From these simulations we will, with defined uncertainty, quantify the impact on forest carbon accumulation of existing forest and the potential accumulation of new planting. This information can feed into national planning identifying locations which have the greatest confidence of being a net sink of carbon under climate change highlighting forest areas which are most important to protect and suitable for new planting.

How to cite: Smallman, T., Milodowski, D., and Williams, M.: Quantifying forest growth and uncertainty across Brazil under potential future climates: combing models and Earth Observation, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-16550, https://doi.org/10.5194/egusphere-egu2020-16550, 2020.

Vegetation fires have a large impact on greenhouse gas emissions in Brazil. In the Brazilian Cerrado, fires occur as part of the natural dynamics of the leading savanna landscapes, but can have their frequency and extent greatly increased by changes in land cover and use, and climate. In addition to atmospheric composition, fires are also linked to deforestation and land degradation, so modeling tools to estimate their occurrence and effects are of significant interest in analyses of climate forcing and sustainable management of the study region.  In order to contribute to modeling tools that may support quantifying the impacts of vegetation fires and help sustainable management of the Cerrado in Brazil, we are working on improving the Integrated Model of Land-surface Processes (INLAND) to estimate fire emissions based on the model’s calculation of vegetation properties and burned area.  Based on data from remote sensing, we are calibrating INLAND burned area outputs, which in combination to modeled vegetation biomass will provide the basis for the emissions estimates, following the Intergovernmental Panel on Climate Change (IPCC) guidelines for greenhouse gas inventories. Our current estimates present correct timing and spatial patterns at regional scales, which we plan to improve to match information from other studies and databases.

How to cite: Cardoso, M., Castro, A., von Randow, C., and Sanches, M.: Modeling of greenhouse gas emissions from fires in the Brazilian Cerrado, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-12561, https://doi.org/10.5194/egusphere-egu2020-12561, 2020.

The potential of palm-oil biodiesel to reduce greenhouse gas (GHG) emissions compared to fossil fuels is increasingly questioned. So far, no measurement-based ecosystem GHG budgets were available, and plantation age was ignored in Life Cycle Analyses (LCA). We conducted LCA based on measured CO₂, CH₄ and N₂O fluxes in young and mature Indonesian oil palm plantations. CO₂ dominated the on-site GHG budgets in both the young and mature plantations. The young plantation was a carbon source (1012 ± 51 gC m^-2 yr^-1), while the mature plantation was a carbon sink (-754 ± 38 gC m^-2 yr^-1). LCA considering the measured fluxes showed higher GHG emissions for palm-oil biodiesel than traditional LCA assuming carbon neutrality. Plantation rotation-cycle extension and earlier-yielding varieties potentially decrease GHG emissions. Due to the high emissions associated with forest conversion to oil palm, our results indicate that no emission savings are achieved from biodiesel from first rotation-cycle oil palm plantations. Only biodiesel from second rotation-cycle plantations or plantations established on degraded land has the potential for pronounced GHG emission savings.

How to cite: Meijide, A., de la Rúa, C., Guillaume, T., Röll, A., Hassler, E., Steigler, C., Tjoa, A., June, T., Corre, M. D., Veldkamp, E., and Knohl, A.: Measured greenhouse gas budgets challenge emission savings from palm-oil biodiesel, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-11500, https://doi.org/10.5194/egusphere-egu2020-11500, 2020.

Drainage canals are a potentially important source of methane from tropical peatlands. Groundwater flow transports dissolved methane to canals, where it has the potential to escape to the atmosphere. However, these emissions are poorly characterized, and the extent to which methane is oxidised before being emitted to the atmosphere is unknown. In this study, we present preliminary data from a deforested tropical peatland in Brunei Darussalam. We use measurements of stable carbon isotopes to track methane production in the peatland. To determine the fraction of methane which is emitted vs. oxidized in a canal draining the site, we use measurements of the δ¹³C and δD of CH₄ in ditch water samples as well as surface gas samples. In addition, we monitor outflow and oxygen content in the ditch. Together, these measurements, in combination with a reactive transport hydrological model will enable us to estimate methane production, oxidation and fluvial export.

How to cite: Somers, L., Hoyt, A., Bte Isnin, S., Cobb, A., and Harvey, C.: Methane emission and oxidation in canals draining tropical peatlands, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-20747, https://doi.org/10.5194/egusphere-egu2020-20747, 2020.

Since 1990, intact forest cover in tropical peatlands of western insular Southeast Asia has dropped to less than 10%.  Most deforested and degraded areas are also affected by drainage, which modifies most important ecological and biogeochemical processes, including carbon dioxide fluxes, methane fluxes, fire risk, and vegetational succession.  Therefore, in this region, peatland ecosystem processes and their response to anthropogenic change occur against the background of long-term spatial impacts from changing drainage networks. We build on earlier work on tropical peatland morphology to develop spatial predictions of the long-term effects of drainage network configuration on tropical peatlands.  We apply this analysis to examine the impacts of anthropogenic drainage on the capacity for carbon storage within natural and artificial drainage networks in Southeast Asia. With a case study, we then show how this approach can be used to produce quantitative estimates of how much peat will be lost or gained in the long term, and where, after drainage or restoration projects.

How to cite: Cobb, A., Dommain, R., Tan, F., Heng, N., and Harvey, C.: Effects of changing drainage networks on carbon storage capacity of Southeast Asian peatlands, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-12299, https://doi.org/10.5194/egusphere-egu2020-12299, 2020.

Southeast Asia is a region where forest clearance, drainage of peatlands for agriculture, and ongoing use of fire to ‘manage’ land leads to extensive emissions of greenhouse gases to the atmosphere, and significant disturbance to peatland soils. While recent campaigns investigating tropical peatland fire emissions have improved our knowledge and understanding of ‘direct’ greenhouse gas emissions during fires, there remains a significant gap in our knowledge of the immediate post-fire impacts on peat respiration and methanogenesis. Ongoing research shows that peatland microbial communities (responsible for respiration), including methanogens and methanotrophs (responsible for controlling net methane emissions), are considerably altered following fire disturbance. As such, we hypothesise that peatland fires will lead to significant alterations to GHG emissions, compared to sites that have not burned. Further, we also hypothesise that the magnitude of this post-fire effect will be predictably interrelated to different forms of peatland degradation and land-use history.

Here we present results from seven fire locations (recently burnt) and their corresponding neighbouring control sites (not recently burnt), three of our fire locations were associated with forest clearance fires, while the other four locations were slash fires on oil palm plantations. We characterize the post-fire disturbance emissions of carbon dioxide (CO₂) and methane (CH₄) in situ, in the immediate aftermath of a fire (within days or weeks), and in the subsequent months following a fire at our burn sites. For comparison, we also measure CO₂ and CH₄ emissions from neighbouring control sites that remained unburnt. We find substantial, significant differences in CH₄ emissions between the burn sites and control sites for all seven of our measurement locations. We suggest a number of mechanisms responsible for this post-fire effect, including disturbance to the methanotroph microbial communities at the burn sites, as well as reduced elevation at the burn sites, leading to higher water tables.

How to cite: Smith, T., Evers, S., Lupascu, M., and Chiu, H.: How do tropical peatland greenhouse gas emissions respond in the immediate aftermath of a fire?, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-12567, https://doi.org/10.5194/egusphere-egu2020-12567, 2020.

Over the past few decades, tropical peatlands in Southeast Asia have been heavily degraded for multiple land uses, mainly by employing drainage and fire. More importantly, the extent of these degraded areas, primarily covered with ferns and sedges, have increased to almost 10% of the total peatland area in Southeast Asia. In particular, the role of sedges in plant-mediated gas transport to the atmosphere has been recognized as a significant CH₄ pathway in northern peatlands, however, in the Tropics this is still unknown. Within this context, we adopted an integrated approach using on-site measurements (CH₄, porewater physicochemical characteristics) with genomics to investigate the role of hydrology, vegetation structure, and microbiome on CH₄ emission from fire-degraded tropical peatland in Brunei.

          We found for the first time that in degraded tropical peatlands of Southeast Asia, sedges transported 70-80% of the total CH₄ emission and significantly varied with values ranging from 1.22±0.13 to 6.15±0.57 mg CH₄ m^-2 hr^-1, during dry and wet period, respectively. This variation was mainly attributed to water table position along with changes in sedge cover and porewater properties, which created more optimal methanogenesis conditions. Total emissions via this process might increase in the future as the extent of degraded tropical peatlands expands due to more frequent fire episodes and flooding.

          Further, we used 16S rRNA high-throughput sequencing to investigate the microbiomes in peat profile (above and below water table) as well as rhizo-compartments (Rhizosphere, Rhizoplane, Endosphere) of sedges. We found that the peat profile as well as rhizo-compartments of sedge harboured a higher number of methanogenic archaea in the order Methanomicrobiales and Methanobacteriales, compared to non-burnt and bulk soil, which further explains our findings of higher CH₄ emission from degraded tropical peatland areas covered with sedges. These insights into the impact of fire on hydrology, vegetation structure, and microbial community composition on CH₄ emissions provide an important basis for future studies on CH₄ dynamics in degraded tropical peatland areas.

How to cite: Akhtar, H., Lupascu, M., S. Kulkarni, O., Bandla, A., S. Sukri, R., R. Cobb, A., E. L. Smith, T., and Swarup, S.: Impact of fire on vegetation, soil microbes and CH4 emission from a degraded tropical peatland, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-12743, https://doi.org/10.5194/egusphere-egu2020-12743, 2020.

Tropical peat swamp forests hold about 15–19% of the global organic carbon (C) pool of which 77% is found in Southeast Asia. Nonetheless, these ecosystems have been drained, exploited for timber and land for agriculture, leading to frequent fires in the region. Fire alters the physico-chemical characteristics of peat as well as the hydrology, which may convert these ecosystems into a source of C for decades as C emissions to the atmosphere exceeds photosynthesis.

To understand the long-term impacts of fire on C cycling, we investigated C emissions in intact and degraded PSFs in Brunei Darussalam, which has experienced 7 fires over the last 40 years. We quantified the magnitude and patterns of C loss (CO₂, CH_4, and Dissolved Organic carbon) and soil-water quality characteristics along with continuous monitoring of soil temperature and water table level from June 2017 to January 2019. To investigate the age and sources of C contributing to ecosystem respiration (R_eco) and CH₄, we used natural tracers such as ¹⁴C.

We observed a major difference in the physico-chemical parameters, which in turn affected C dynamics, especially CH₄. In burnt areas (7.8±2.2 mg CH₄ m^-2 hr^-1) the CH₄ emission was approximately twice compared to the intact peat swamp forest (4.0±2.0 mg CH₄ m^-2 hr^-1) due to prolonged higher water table creating optimum methanogenesis conditions. On the contrary, R_eco did not show a significant difference between burnt (432±83 mg CO₂ m^-2 hr^-1) and intact areas (359±76 mg CO₂ m^-2 hr^-1). Further, radiocarbon (¹⁴C) analysis showed an overall modern signature for both CO₂ and CH₄ fluxes implying a microbial preference for the more labile C fraction in solution.

With frequent fires and more flooding in the future, these degraded tropical peat swamp forests areas may remain a hot spot of C emissions as suggested by our findings.

How to cite: Lupascu, M., Akhtar, H., Smith, T. E. L., and Sukri, R. S.: Post-fire carbon emissions from degraded tropical peat swamp forests in Brunei, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-6337, https://doi.org/10.5194/egusphere-egu2020-6337, 2020.

Tropical peatlands have the potential to be significant sources of methane (CH₄) to the atmosphere but their contribution to the global methane budget remains uncertain. Although much prior work has focused in Southeast Asia, other tropical regions, such as the Congo and the Amazon, have a much wider diversity of peatlands with more variable CH₄ emissions. Our work aims to better understand CH₄ production and emissions in these diverse peatlands, and how they are controlled by hydrology, geochemistry and vegetation. Using stable isotope and radiocarbon measurements, we assess the production pathway for methanogenesis and its carbon source at sites across the Pastaza-Marañon basin in Peru. As the largest peatland complex in the Amazon, this region is home to many peatland types, from palm swamps to open peatlands to pole forests. We find clear links between site geochemistry, hydrology, and CH₄ production. In rain-fed ombrotrophic sites (pH 3-4), we observe low emissions and highly depleted δ¹³CH₄ values (as low as -100‰). The lack of external nutrients and acidic conditions likely limit methanogenesis, and hydrogenotrophic methanogenesis dominates. In more minerotrophic sites (pH 5-6), more enriched methane (-75 to -60‰) suggests a contribution from acetoclastic methanogenesis. Emissions rates are also higher, likely fueled by external nutrient inputs from seasonal flooding. Across sites, modern, vegetation-derived inputs are the dominant carbon source for methanogenesis, with a limited contribution from old peat carbon in some ombrotrophic sites. The strong relationships we observe between peatland hydrology, vegetation, geochemistry and methane emissions will enable future work to upscale methane emissions across the region.

How to cite: Hoyt, A., Cadillo-Quiroz, H., Xu, X., Torn, M., Bazán Pacaya, A., Jacobs, M., Shapiama Peña, R., Ramirez Navarro, D., Urquiza-Muñoz, D., and Trumbore, S.: Isotopic Insights into Methane Production and Emission in Diverse Amazonian Peatlands, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-12960, https://doi.org/10.5194/egusphere-egu2020-12960, 2020.

Challenges to peatland restoration in Indonesia
By 
Nyoman Suryadiputra*) 

Tropical peat swamps in Indonesia are currently experiencing degradation at a very alarming rate. Degradation starts from the time of land clearing (generally burned / uses fire) for both private and community-owned plantations, then a very massive network of drainage canals is built (every 1 Ha of peat land cleared, about 120 m - 700 m long canals are needed). These drainage canals aim to reduce the surface water level of peat so that the land can be planted (especially for) oil palm or acacia. However, peat water release can go out of control, beyond the peatland water level threshold determined by government regulation No 71/2014 on Peatland Management, as a result peat becomes dry, flammable and emits large amount of GHGs. In the long run, if drainage and fires continue, peatlands will experience subsidence, form basins, peat even disappear, flooded during rain and eventually the land becomes unproductive (stranded) and difficult to restore. Such conditions will be more severe and difficult to overcome if in the landscape (peatland hydrology unit) there are various activities by various parties, each of whom has different interests and understandings of peatland use. Regarding the above, restoration of peatland that has been damaged has a very serious challenge. Damage that is getting heavier will have a high level of difficulty and a long recovery time. In addition, the success rate of restoration is determined by benchmarks or recovery criteria that have not been scientifically determined and adopted by the Indonesian government.

Keywords : peatland, degradation, landscape, restoration    

*)  Director of Wetlands International Indonesia

How to cite: Suryadiputra, N.: Challenges to peatland restoration in Indonesia, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-7695, https://doi.org/10.5194/egusphere-egu2020-7695, 2020.

Worldwide, peatlands are important sources of dissolved organic matter (DOM) and trace metals (TM) to surface waters and these fluxes may increase with peatland degradation. In Southeast Asia, tropical peatlands are being rapidly deforested and drained. The black rivers draining these peatland areas have high concentrations of DOM. However, the fate of this fluvial carbon export is uncertain, and its role as a trace metal carrier has never been investigated. This work aims to address these gaps in our understanding of tropical peatland DOM and associated elements in the context of degraded tropical peatlands of Indonesian Borneo. We quantified dissolved organic carbon and trace metals concentrations in the dissolved, fine colloidal and coarse colloidal fractions and determined the characteristics (optical and isotopic) of the peatland-derived DOM as it drains from peatland canals, flows along black river and eventually mixes with the Kapuas Kecil River before meeting the ocean near the city of Pontianak in West Kalimantan, Indonesia. Black rivers draining degraded peatlands show significantly higher concentrations of Al, Fe, Pb, As, Ni, and Cd, compared the white river. A strong association is observed between DOM, Fe, As, Cd and Zn in the dissolved and fine colloid fraction, while Al is associated to Pb and Ni and present in a higher proportion in the coarse colloidal fraction. We additionally measured the isotopic composition of lead released from degraded tropical peatlands for the first time and show that Pb originates from anthropogenic atmospheric deposition. Degraded tropical peatlands are important sources of DOM and trace metals to rivers and a secondary source of atmospherically deposited contaminants.

How to cite: Gandois, L., Hoyt, A. M., Mounier, S., Le Roux, G., Harvey, C. F., Claustres, A., Nuriman, M., and Anshari, G.: Degraded peatlands of South East Asia : a source of Trace Metal to surface waters, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-10859, https://doi.org/10.5194/egusphere-egu2020-10859, 2020.

Peatlands are a vast store of organic carbon and play a significant role in the global carbon cycle. The abundance of lipids, especially microbial hopanoids are, directly or indirectly, participating in the carbon cycle in peatlands. Although the hopanoids and their compound-specific carbon isotope composition have been applied to some extent in the study of paleoenvironment and paleoecology, it is still misty about the diagenetic transformation process and the main controlling factors of the early diagenetic transformation of hopanoids. The potential of paleo-ecological application of hopanoids is still lack of systematic understanding. Previously we investigated several typical peatlands in China and found pH plays an important role in the early diagenetic transformation of bacteriohopanepolyols (BHPs) into geohopanoids. We also found the pH value promotes the isomerization of geohopanoids. Here, we focus on the biogeochemical research of microbial hopanoids in Dajiuhu peatland. We  first carried out a series of modern process monitoring on a seasonal scale, such as monthly investigating the climate factors (air temperature, air humidity, rainfall), water, soil temperature, soil moisture, water chemistry (pH, ORP, conductivity, dissolved oxygen), the main nutrient (nitrate, phosphate, etc.), dissolved organic carbon (content, composition and isotopic composition). On this basis, we discussed the relationship between those compounds (composition, carbon and hydrogen isotopes) and the climate-environmental conditions as well as the carbon dynamics. Further on, we examined the response of the carbon cycle based on hopanoid index in a peat core (18ka BP) in Dajiuhu peatland. Our results showed that in acidic peat deposits in Dajiuhu peatland, the carbon isotopes of hopane are generally more than 5‰ positive compared with the carbon isotopes of n-alkanes from higher plants both the surface peat samples and core peat samples, which indicates that in acidic peat environment, the hopanoid-produced bacteria mainly take carbohydrate as carbon source. We also showed that the difference of the carbon isotopes between hopane and n-alkanes is not stable, suggesting that these hopanoids may use different carbon sources.

How to cite: Xue, J. and Huang, X.: The biogeochemical research of microbial hopanoids in Dajiuhu peatland, central China, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-8219, https://doi.org/10.5194/egusphere-egu2020-8219, 2020.

In tall vegetation canopies, such as forest or oil palm monoculture plantations, the below-canopy airflow can be influenced by the local topography and thereby cause horizontal exchange of the below-canopy air. Especially during night time, calm weather conditions may result in the formation of an isolated layer near the surface, which is decoupled from the above-canopy air layer. When decoupling and below-canopy horizontal air flow occurs, there is a high potential that above-canopy measured carbon dioxide (CO₂) fluxes based on eddy covariance measurements might not represent the true ecosystem CO₂ flux as below-canopy respiration might be undetected by the eddy covariance system. Nevertheless, eddy covariance data are frequently used as the reference for fluxes of tall vegetation ecosystems or for validation of modelling approaches estimating gross primary production (GPP) and net ecosystem exchange (NEE). It is therefore important to have accurate information on air mixing, decoupling and sub-canopy drainage flow to understand the complex CO₂ exchange behaviour in tall vegetation ecosystems.

In this context, we investigate wind and micrometeorological dynamics of a mature oil palm monoculture plantation (tropical lowland, Jambi Province, Sumatra, Indonesia). We use data from above- and below-canopy eddy covariance and micrometeorological measurements within the oil palm plantation to assess the wind dynamics and the strength of the turbulent mixing as an estimator for the degree of coupling. Further, we explore the potential implications of decoupling and horizontal below-canopy flow on the above-canopy derived NEE.

Preliminary results show that wind is generally weak in the oil palm plantation. Using a breakpoint analysis, the correlation of below- and above-canopy standard deviation of vertical wind speed (σ_w) derived from sonic eddy covariance measurements below (2.4 m height) and above the canopy (22 m height), we identified a site-specific σ_w threshold of 0.11 m s^-1 (below-canopy) and 0.26 m s^-1 (above-canopy) above which the atmospheric conditions are in fully coupled state. During the day, unstable conditions dominate over stable conditions while in the twilight hours and during the night, the reverse is the case. Below-canopy wind comes mostly from south-eastern directions during both day and night, and tends to blow independently from wind above the canopy for conditions with above-canopy u* < 0.3 m s^-1. Based on the above-canopy eddy covariance NEE measurements and on the direction difference (ΔWD) between above- and below-canopy wind, we observe a threshold of ~70° ΔWD above which the two layers might be decoupled. Below-canopy air flow might therefore influence the above-canopy NEE detections, biasing carbon balance estimates.

How to cite: Stiegler, C., June, T., and Knohl, A.: Air mixing and sub-canopy advection in an oil palm plantation in Indonesia, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-6796, https://doi.org/10.5194/egusphere-egu2020-6796, 2020.

The 2015/2016 El Niño Southern Oscillation (ENSO) was one of the most severe, as strong as in 1997/1998, and reached mainly the eastern Amazon. ENSO in the Amazon causes a decrease in precipitation and increase in temperature. Oil palm in dry conditions, low humidity, high temperatures and soil water deficit has its photosynthesis inhibited, decreased evapotranspiration and stomatal conductance and inflorescence abortion, for example. The objective of this study was to estimate the CO₂ gas exchange in interspecific hybrid oil palm plantation (Elaeis guineensis Jacq x Elaeis oleifera (Kunth) Cortés), relating to the effects of the meteorological variables in the 2015 ENOS in the eastern Amazon. The study area was a 12-year-old oil palm plantation (01º51’43.2’’S, 048º36’52.2’’W) in the municipality of Moju, Pará, Brazil, where a micrometeorological observation tower was installed. Were quantified the meteorological variables such as photosynthetically active radiation (PAR), vapor pressure deficit (VPD) and soil moisture. And the fluxes of CO₂ and H₂O for application of the eddy covariance method. Photosynthetic parameters were estimated using the light response curve (LCR) in the non-rectangular hyperbole model. The results were shown seasonally, the months of the wet season (December to June) presented precipitation greater than 150 mm/month and the dry season (July to November) with monthly precipitation less than 150 mm, being a threshold that influences the water deficit for the oil palm. The dry season presented a reduction of more than 57% in the precipitation, when compared to the climatological normal data of Belém. The daily average of net CO₂ exchange was higher in the wet season of -22.50 (± 0.40) µmol m-² s-¹ at 11:00 am and -22.14 (± 0.68) µmol m-² s-¹ at 10:30 am (local hour) in the dry season. In the wet season the parameters of LCR were lower quantum efficiency (0.0479 ± 0.0039 μmol CO₂ μmol-¹ photon absorbed), higher CO₂ assimilation rate (35.82 ± 1.92 µmol m-² s-¹) and lower ecosystem respiration (6.11 ± 0.39 μmol m-² s-¹). The dry season exhibited a quantum efficiency of 0,0494 (± 0,0063) μmol CO₂ μmol-¹ photon absorbed, CO₂ assimilation rate of 31,02 (± 1,93) µmol m-² s-¹ and higher ecosystem respiration (6.61 ± 0.65 μmol m-² s-¹). PAR and VPD preconditioned to net CO₂ exchange with a correlation coefficient of 0.75 and 0.72 and of determination of 0.56 and 0.52, in the wet and dry seasons, respectively. These results are important for a better understanding of oil palm behavior in the face of a severe weather event in eastern Amazonia.

How to cite: Silva, J. A. D. F., Araujo, A. C. D., von Randow, C., Manzi, A. O., and Oliveira, L. R. D.: CO2 gas exchange in oil palm plantation under 2015 ENSO conditions in eastern Amazonia , EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-21003, https://doi.org/10.5194/egusphere-egu2020-21003, 2020.

Tropical oil palm (OP) plantations are major emitters of greenhouse gases (GHGs), but there are management options, which may reduce these emissions, including increasing understory biomass. Managing the vegetation within and around plantations could potentially minimise environmental damage and maximise co-benefits such as soil protection, pest control and diversity. Such practices include creating reserves, buffer strips and management of vegetation in the plantations themselves. The impact of these management practices is uncertain, and there is a real need for an evidence-base to guide improvements in the environmental sustainability of OP management.

The timing for research related to management options is critical for influencing current decision-making. In Indonesia, most OP plantations were established in the late 1980s and early 1990s and due to the 25 – 30-year life cycle of OP plantations, nearly half are due to be clear-cut for replanting in the near-future. Hence, it is vital to understand replanting and restoration options which simultaneously allow for high productivity as well as supporting biodiversity and minimising GHG emissions.

The scope and specific objectives of our study were:

• 1) To measure GHG emissions under different understory management techniques (with/without vegetation through use of herbicides).
• 2) To link GHG data to soil data to develop understanding of ecosystem function under different OP plantation management approaches.

We will present monthly static chamber measurements of GHG emissions for the duration of one year starting October 2018, established on an existing long-term experiment investigating the impact of diversifying understory vegetation on biodiversity, ecosystem functioning and yield in Sumatra, Indonesia (The Biodiversity and Ecosystem Function in Tropical Agriculture Project (BEFTA)). The three different understory management treatments were:

• 1) Normal biodiversity complexity: standard industry practice, intermediate level of herbicide use in harvest circles.
• 2) Reduced biodiversity complexity: spraying/removing all understory vegetation with herbicides.
• 3) Enhanced biodiversity complexity: reduced-input management with no herbicide application and limited understory cutting.

We measured the GHG fluxes of nitrous oxide (N₂O), methane (CH₄) and soil ecosystem respiration/carbon dioxide (CO₂) using static chambers and analysis by gas chromatography (GC-µECD/FID).

Preliminary results show little difference amongst the different understory treatments in terms of N₂O fluxes. Fluxes were generally low (0-0.1 µg m^-2 h^-1) with high variability. However, there is a trend towards slightly higher emissions during the wetter months (Oct-Dec 2018) of up to 0.2 µg m^-2 h^-1.

Methane (CH₄) fluxes were generally small and fluctuated around zero. During the wet months, (Oct to Dec 2018) small emission fluxes up to 3 µg m^-2 h^-1 were observed; whereas during the dry months uptake of methane, prevailed. No distinctive differences between the different treatments was observed.

Due to the age of the plantation and imminent replanting, none of the plots were being fertilised at the time of measurement – greater differences between vegetation treatments may be observed under fertilisation.

In conclusion, initial results showed that the presence or absence of understorey did not increase soil emissions of N₂O and CH₄. This suggests that the within-crop ecological benefits do not result in an increased GHG burden.

How to cite: Drewer, J., Sionita, R., Pujianto, P., White, S., Luke, S., Turner, E., Banin, L., Skiba, U., Dwi Advento, A., and Caliman, J.-P.: The impact of diversifying understory vegetation in oil palm plantations on greenhouse gas emissions, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-2860, https://doi.org/10.5194/egusphere-egu2020-2860, 2020.

Tropical forest ecosystems are important components of global biogeochemical cycling. Many tropical rainforests grow in old and highly weathered soils, depleted in phosphorus (P) and net primary productivity in tropical forests is often limited by P availability. It is unclear, however, if heterotrophic microbial communities in tropical soils are also limited by P or rather by carbon (C). Elemental limitations of microorganisms in soil have often been approached by measurements of respiration rates in response to additions of nutrients or carbon. However, it has been argued lately, that microbial growth rather than respiration should be used to assess limitations.

In this study we therefore ask the question whether the growth of heterotrophic microbial communities in tropical soil is limited by available phosphorus or by carbon. We collected soils from three sites along a topographic gradient (plateau, slope, bottom) differing in soil texture, total and available P concentrations from a well-studied, P-poor region in Nouragues, French Guiana. We incubated these soils in the laboratory with C in the form of cellulose, inorganic phosphorus and with a combination of both, and studied microbial growth by measuring the ¹⁸O incorporation from labelled water into microbial DNA. Moreover, we measured microbial respiration and determined microbial biomass C, N (nitrogen) and P.

Our results demonstrate that, although microbial biomass C and N was similar in soil collected from all three topographic sites, soil respiration rates were significantly higher in soils from the plateau indicating a more active microbial community. Microbial C and N did not respond to cellulose and inorganic P additions, only microbial P increased significantly when P was added in all soils. Although microbial biomass C was not increased, C and P additions stimulated microbial respiration in clay rich plateau soils. In slope soils microbial communities initially only increased respiration activity in response to P additions, however at the end of the incubation also C showed significant differences in respiration activity, with strongest increases when C and P were added in combination. In sandier bottom soils microorganisms responded with increased activity to C addition, but also here respiration showed strongest increases in response to combined carbon and phosphorus additions. We will discuss these findings in relation to the pattern of gross growth rates in these soils and evaluate the stoichiometric limitations of microbial activity and turnover.

How to cite: Ranits, C., Fuchslueger, L., Van Langenhove, L., Janssens, I., Peñuelas, J., and Richter, A.: What controls microbial growth in tropical soils? The role of carbon and phosphorus., EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-20992, https://doi.org/10.5194/egusphere-egu2020-20992, 2020.

The conversion of tropical forest for cassava cultivation is widely known to decrease the soil organic matter (OM) and nutrient contents of highly weathered soils in the tropics. Amazonian Dark Earth (ADE) might be affected less due to their historical anthropogenic amelioration with e.g. charcoal, ceramics and bones, leading to higher soil OM and nutrient concentrations. In this study, we analysed the effect of land use change on the OM dynamics and its composition under tropical conditions, using ADE and an adjacent Acrisol (ACR) as model systems. Soil samples were obtained south of Manaus (Brazil), from a secondary forest and an adjacently located 40-year-old cassava plantation. The land use change induced a severe decrease of organic carbon (OC) concentrations in ADE (from 35 to 15 g OC kg^‑1) while OC in the adjacent ACR was less affected (18 to 16 g OC kg^‑1). Soils were analysed by ¹³C NMR spectroscopy to obtain information on how the conversion of secondary forest to cassava affected the chemical composition of OM. Our results show that land use change induces differences in the OM composition: The OM in ADE changes to a more decomposed state (increase of alkyl:O/N-alkyl ratio) whereas the OM in ACR changes to a less decomposed state (decrease of alkyl:O/N-alkyl ratio). According to a molecular mixing model, land use change influenced mostly the proportion of lipids, which might be related with a change of the plant input. The incubation of the soils with ¹³C glucose enabled resolving how soil microorganisms were affected by land use change. In both soil types ADE and ACR, land use change caused a reduction of the total ¹³C glucose respiration by approximately one third in a 7-days incubation, implying lower microbial activity. Microorganisms in both soil types appear to be more readily active in soils under forest, since we observed a distinct lag time between ¹³C glucose addition and respiration under cassava planation. This indicated differences in microbial community structure, which we will assess further by determining the ¹³C label uptake by the microbial biomass and the microbial community structure using ¹³C PLFA analysis. Preliminary results from synchrotron-based STXM demonstrate a distinct arrangement of OM at fine-sized charcoal-particle interfaces. Samples of soils receiving ¹³C label will be further analysed by NanoSIMS with the hypothesis that charcoal interfaces foster nutrient dynamics at the microscale. Despite the high loss of OC in the ameliorated ADE through land use change, the remaining OM might improve the nutrient availability thanks to charcoal interactions compared to the ACR. Our results contribute to a better understanding of the sensitivity of OM upon land use change and how the microbial community is responding to land use change in highly weathered tropical soils.

How to cite: Jarosch, K., Colocho Hurtarte, L. C., Gavazov, K., Westphal Muniz, A., Müller, C., and Schweizer, S.: Land use change in Amazonian Dark Earth and Acrisol: Responses of organic carbon, organic matter composition and microbial carbon utilisation , EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-19752, https://doi.org/10.5194/egusphere-egu2020-19752, 2020.

Fighting climate change has never been so urgent as today. As geoscientific research advances, more realistic and appalling future climate scenarios are unraveled. Decreasing greenhouse gases (GHG) emissions is not sufficient to avoid a bad outcome; hence, mitigation actions are needed to reduce climate-related risks in the future. The depletion of the soil organic carbon (SOC) pool due to land use change and soil degradation have substantially contributed to the increase in the atmospheric CO₂ concentration. Likewise, the sequestration of C by the soil is crucial to reverse this issue. These are especially important processes in the tropics where the replacement of native vegetation by agriculture still occurs at a high rate. Brazil is one of the biggest agricultural producers in the world. In 2018, agriculture and land use change represented 70% of the total Brazilian GHG emissions. Fortunately, Brazil also has opportune mitigation options. Since 1965, the Brazilian Forest Code requires landowners to conserve native vegetation by means of Riparian Preservation Areas, among other categories. Riparian forests provides several ecosystem services like water protection, biodiversity conservation and carbon sequestration. Frequently, the discussion over the carbon sequestration potential of riparian forests focus on the aboveground carbon, nevertheless, SOC stocks are more stable and protected from natural and anthropogenic hazards. We consider that the mandatory reforestation of riparian zones is a significant mitigation strategy in Brazil, owing to the potential of SOC sequestration by Brazilian biomes and the extensive area to be reforested. The objective of this study was to assess the SOC stocks of the main land uses of an agricultural watershed located in the state of São Paulo, Southeastern Brazil, and estimate the change in the SOC stocks that would occur with the reforestation of the riparian areas of this watershed. In order to achieve this goal, we compared the SOC stocks of riparian forests with the two main agricultural uses of the region, namely pasture and sugarcane. The mean SOC stock at 30 cm for riparian forests was of 44 Mg.ha^-1, for pastures was of 26 Mg.ha^-1 and for sugarcane was of 27 Mg.ha^-1. Although the riparian forests of the region are often poorly preserved, they contained considerably more SOC at 30 cm than the agricultural uses. Based on the estimates of the SOC stocks of the main land uses and the extent of the riparian zones of the sampled sites, we could foresee an accretion of 20% of organic carbon in the 30 cm soil layer of those areas. We hope that this study highlight the importance of the riparian forests and the ecosystem services they provide, and the relevance of the Brazilian Forest Code in the mitigation of climate change.

How to cite: Galera, L. D. A. and Martinelli, L. A.: Soil organic carbon sequestration potential of reforesting riparian areas in an agricultural watershed in the state of São Paulo, Brazil, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-3658, https://doi.org/10.5194/egusphere-egu2020-3658, 2020.

Soil extracellular enzymes are crucial for belowground functioning and are sensitive to anthropogenic land use change. The potential effects of tree species on soil microbial and biochemical properties provide crucial feedbacks on mineralization, a key ecosystem function beneath the tree canopy. In the highlands of northern Ethiopia, remnants of the original Afromontane forests are largely restricted to church forests with indigenous tree species. However the impacts on potential soil enzymatic activity by conversion of those forests to monocultures for wood production is largely unknown. We investigated potential soil enzyme activities under four indigenous tree species and adjacent Eucalyptus globulus and Cupressus lusitanica plantations in Gelawdios, Amhara Regional State, Ethiopia. The potential activities of six enzymes associated with soil C, N and P cycling were measured following the fluorometrically labelled substrates techniques. All enzymes exhibited significantly higher activities in soils under the indigenous trees than the plantation species except, N-acetylglucosaminidase, that was the highest in Eucalyptus globulus soil due to the ectomycorrhizae, associated with the Eucalyptus root systems. Among the four indigenous species Apodytes dimidiata showed the lowest activitie for most of the enzymes. A stronger positive correlation was observed between enzyme activity and total N than with total C in the soil. Acid phosphatase had the highest activity followed by  β-Glucosidase (482 and 167 nmol mg^-1 microbial biomass respectively). The activities of leucine aminopeptidase, β-xylosidase, N-Acetylglucosaminidase and cellobiohydrolase in soils under indigenous trees ranged between 63-23 nmol mg^-1 microbial biomass. The species specific effects of trees on soil enzyme activities indicate strong influence of tree traits on mineralization processes.     

How to cite: Ahmed, I. U., Mengistie, H. K., Sandén, H., and Godbold, D.: Effects of tree species on soil enzyme activities in natural mixed forest and monoculture plantations in Ethiopia, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-21746, https://doi.org/10.5194/egusphere-egu2020-21746, 2020.

South African ecosystems are highly vulnerable to the effects of climate change, such as increasing temperatures, modifications in rainfall patterns, increasing frequency of extreme weather events and fire, and increased concentration of atmospheric carbon dioxide (CO₂). At the same time, ecosystems are impacted by livestock grazing, cultivation, fuelwood collection, urbanization and other types of human land use. Climatic and land management factors, such as water availability and grazing intensity, play a dominant role in influencing primary production and carbon fluxes. However, the relative role of those parameters still remains less known in many South African ecosystems. Investigation of the carbon inter-annual variability at dwarf shrub Karoo sites will assist in understanding savanna dynamics and in constraining climate change scenarios as basis for climate adaptation strategies. 

This research is part of the EMSAfrica (Ecosystem Management Support for Climate Change in Southern Africa) project, which aims at producing data and information relevant to land users and land managers such as South African National Parks (SANParks). A particular focus is given on the importance of carbon cycling in degraded vs. intact systems. We investigate the impacts of climate parameters and diverse land management on ecosystem-atmosphere variability of carbon fluxes, latent and sensible energy. Long-term measurements were collected and analyzed from two eddy-covariance towers in the Karoo, Eastern Cape, South Africa. Study areas had almost identical climatic conditions but differ in the intensity of livestock grazing. The first site represents controlled grazing and comprises a diverse balance of dwarf shrubs and grasses, while the second site is degraded through overgrazing in the past (rested for approximately 8 years) and mainly consists of unpalatable grasses and short-lived species. These ecosystems are generally characterized by alternating wet (December to May) and dry seasons (June to November) with the amount and distribution of rain (average 373 mm per year) and soil moisture as the main drivers of carbon fluxes. We observed peak CO₂ uptake occurring during the wet season (January to April) and a progressive decrease from wet to dry periods being highly controlled by the amount of precipitation. At the end of the observation period (November 2015 – November 2019), we found that both study sites were considerable carbon sources, but during wet periods 'overgrazed in the past' site had stronger carbon sequestration compared to 'controlled grazing' site. The higher carbon uptake could be an indication that resting of the highly degraded site for a long period may improve carbon uptake in the Karoo ecosystems. Our study shows that CO₂ dynamics in the Karoo are largely driven by water availability and the effects of grazing intensity on above-ground biomass.

How to cite: Rybchak, O., Mukwashi, K., Du Toit, J., Feig, G., Bieri, M., and Brümmer, C.: How land management and water availability control ecosystem-atmosphere carbon exchange in the Karoo, South Africa, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-13578, https://doi.org/10.5194/egusphere-egu2020-13578, 2020.

Semi-arid rangelands, common across Sub-Saharan Africa (SSA), are under increased anthropogenic pressure by a growing population and the necessity to produce sufficient amounts of food and nutrition. Rangelands in SSA are characterized by nutrient-poor soils and distinct wet and dry season(s). Due to the soil and climate combination, conventional crop farming is often not feasible without additional inputs in terms of water and mineral fertilizer. Instead, livestock keeping constitutes a valuable option to use these marginal lands and has been practiced in SSA for centuries. As a result, livestock and wildlife jointly feed on these pastures, a trait that distinguishes these systems from most western rangelands, where large herbivore herds are the exception rather than the norm. To date a thorough, climate-smart assessment that includes continuous greenhouse gas (GHG) exchange measurements in combined wildlife-livestock systems has not been undertaken. Here we provide eddy covariance (EC) measurements of CO₂/CH₄/H₂O from Kapiti Research Station in Kenya - a benchmark site for sustainable food production while also hosting wildlife. The GHG exchange measurements were complemented with wildlife camera traps to monitor both animal movement as well as plant phenology in the footprint of the EC tower. Our results show continuous CO₂ uptake during the wet seasons with considerable CO₂ emissions following distinct (>10mm) precipitation events after prolonged dry periods. Temporal dynamics of net ecosystem exchange of CO₂ was strongly correlated with canopy greenness (green chromatic coordinate) derived from field camera imagery. Methane flux measurements were highly variable and were particularly related to the presence of wildlife and/or livestock in the fetch of the EC tower. Our data suggest that these rangeland systems are accumulating carbon and thus compensate the methane emissions from livestock.

How to cite: Merbold, L., Dowling, T., Leitner, S., Wooster, M., Vrieling, A., Fava, F., and Gluecks, I.: Continuous observations of CO2, H2O and CH4 exchange in an East-African rangeland, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-12657, https://doi.org/10.5194/egusphere-egu2020-12657, 2020.

The Kavango Zambezi Transfrontier Conservation Area (KAZA) is the World’s largest conservation area with an enclosed area the size of Sweden (519,912 km²), and is characterized by savannah forest, woodland and protected lands. KAZA is situated at the heart of the area most vulnerable to climate change in Africa, and forest loss and degradation are major concerns which directly impact wildlife species distributions and a growing human populations. In particular, detailed knowledge of current vegetation density change and forest area estimates throughout the conservation area is sorely missing, which hampers all efforts to mitigate the threats against KAZA and its unique ecosystems. A combination of remotely sensed data and plot-based estimates can provide forest area estimates and above ground biomass (AGB). Previous AGB mapping efforts in Africa focused on tropical humid forests, with little attention on tropical and subtropical savannah forest. The aim of the current study was to establish a link between remote sensing spectral data derived from Landsat 8 and ground characteristics to improve precision of AGB and forest area estimates in savannah forest. We used 114 sample plots distributed on 6 clusters collected over the 2019 winter growing season in Chobe National Park of Botswana and Landsat 8 spectral variables.

Restricting analysis to sampling dates, before the onset of fire burning and leaf yellowing resulted in increased estimation accuracy. We found a linear relationship between above ground biomass and Landsat 8 derived spectral variables (p < 0.001 and p < 0.005). The normalized difference vegetation index (NDVI) and Green-Red Difference Index (GRVI) exhibited a strong correlation with AGB than other indices (R2 = 0.76) and (R² = 0.67), respectively. An improvement in the correlation is seen when AGB (t/ha) and variables relationship is performed in the woodland/forest cluster (n=74), excluding the shrubland and grassland. The AGB of savannah forest in the study area based on spatial analysis was 111.6 Mg/ha. A root-mean-square error (RMSE) value from predicted and observed AGB was 25.6 Mg/ha. The high total AGB value from savannah forest in the study area highlight the importance of the savannah-forest mosaic as a biomass storage and carbon pool. Overall, spectral variables and indices, particularly the NDVI and GRVI and Landsat 8 band 5 (NIR), would be useful predictors and suitable auxiliary information of AGB in the savannah forest. The results of this study show that taking into account stratification/clustering of different vegetation types and senescence period can greatly increase the accuracy of AGB estimation. This results will allow us to build new models to quantify savannah forest change and long-term trends using Landsat time series from 1980 to 2019. Time series modelling will help inform how changing climate threaten the biodiversity of the KAZA region and be able to respond to these threats with appropriate, evidence-based strategies and measures.

How to cite: David, R., Donoghue, D., and Rosser, N.: Earth Observation models help management of tropical dry savannah forests in the Okavango-Zambezi transfrontier conservation zone (KAZA) region of Southern Africa., EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-22135, https://doi.org/10.5194/egusphere-egu2020-22135, 2020.

Deforestation remains one of the largest contributors to global greenhouse emissions. Despite the efforts in monitoring forest change, there is still a lack of pan-tropical spatially-explicit data informing the subsequent land cover (LC) changes over deforested areas (also known as post-loss LC). Based on this premise, this research focuses on predicting post-loss LC over deforested areas as detected by Terra-i, an early warning system of pantropical forest change providing alerts every 16-days from 2004 to present at spatial resolution of 250 m. A supervised deep neural network model suited to extract spatio-temporal patterns from dense earth observation time series data was leveraged in this work by using 16-day MODIS images of 2015. The model was trained according to nine labelled datasets representing different number of LC classes and complexity. These datasets were generated from pre-existing global LC maps with a native spatial resolution ranging from 100 m to 500 m. The effectiveness of the trained models in producing accurate predictions of post-loss LC was assessed over the Amazon region, the largest continuous region of tropical forest in the world. A two-stage assessment approach was conducted to determine the most suitable labelled datasets to predict post-loss LC over Terra-i’s areas. For the first stage, traditional metrics for the assessment of the quality of LC thematic data — e.g. overall accuracy, per-class mapping accuracy, area (or quantity) disagreement and allocation disagreement — were computed according to the test partitions from the labelled datasets. A second stage consisted in using the trained models in 2015 to make predictions for all available years of MODIS satellite imagery, from 2001 to 2018, across seven representative areas distributed in the Amazon. The observed LC predictions were masked using annual aggregated data of Terra-i from 2004 to 2010. The post-LC data by trained model, which represents a given labelled dataset, was verified by i) visualising the temporal and spatial distribution of the most frequent subsequent LC changes; and ii) comparing with Mapbiomas Amazonia, a regional-tuned multi-temporal LC dataset from 2000 to 2017 for the whole Amazon. The results showed that one out of the nine labelled datasets allowed the supervised deep learning model to produce reasonable spatial predictions and classification accuracies (overall accuracy of 86.36±0.64, area disagreement of 5.34±0.39 and allocation disagreement of 8.31±0.64) according to the test partition data. Moreover, the trained model provided similar patterns of post-loss LC as informed by the Mapbiomas dataset. Due to the nature of the model (i.e. neural network) and input data (i.e. global), it is expected the model is scalable to other pantropical areas. The insights and products derived throughout this study are targeted to reduce current uncertainties and challenges in the calculation of global and regional drivers and impacts of deforestation in tropical forests and landscapes.

How to cite: Coca Castro, A., Reymondin, L., and Mulligan, M.: Multi-temporal mapping of pantropical post-loss land cover using dense earth-observation time series and global pre-existing maps, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-389, https://doi.org/10.5194/egusphere-egu2020-389, 2020.

Forest degradation is one of the least understood processes in the terrestrial biome. Information about both the extent and rate of degradation is limited; meaning that estimation of subsequent carbon emissions are poorly quantified. Countries are required to report emissions from forest degradation to the UNFCCC, therefore, increased efforts to improve mapping and monitoring of forest degradation are essential for national reporting. As forest degradation is related to more subtle changes within a forest, resulting from loss of trees, then earth observation technologies, which can quantify changes in Aboveground Biomass (AGB) offer great opportunities in mapping forest degradation.

We test a methodology for monitoring forest degradation, which combines time-series forest plot data with L-band Synthetic Aperture Radar (SAR) data (ALOS/ ALOS-2) to map changes in AGB over time. We test this method in two countries (Mexico and Ghana), to determine if a single methodology can be used to map forest degradation across the tropics. Our study-sites span two tropical continents, covering a range of precipitation, AGB values, forest types, and modes of degradation, from seasonally dry pine forests in Mexico to humid lowland forests in Ghana,

In lower biomass forest, we could map changes in AGB over time, including small AGB losses, associated with minor degradation events (<20% loss of AGB). These minor degradation events are far more widespread than major degradation events (>50% loss of AGB) and therefore are an important source of emissions from degradation. However, in high biomass forest there was some saturation of the radar backscatter signal (>200 Mg ha^-1).

The use of ground-based AGB change data is essential for calibration of SAR data, therefore well managed national forest plot networks, which include degraded forest, are essential for accurate monitoring of degradation. Additionally, using longer wavelength P-Band SAR data, alongside L-Band SAR, could help overcome some issues related to saturation in high biomass forest. If this method were adopted at the regional or national level, it would allow countries, particularly those with lower biomass forest, to quantify emissions from degradation more accurately.

How to cite: Wheeler, C., Mitchard, E., Nolasco Rayes, H., and Mohammed, Y.: Mapping Forest Degradation with ALOS PALSAR: Case studies from Mexico and Ghana, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-18104, https://doi.org/10.5194/egusphere-egu2020-18104, 2020.

The framework of isohydry or anisohydry, which is usually defined as the sensitivity of leaf water potential (Ψ_L) to soil water potential (Ψs), has been rapidly adopted to solve a range of eco-hydrologic problems. While its reliability to describe the impacts of land-atmosphere interaction and seasonal phenology on plants has been recently questioned. In this study, we propose an expansion of the modern isohydricity framework to coordinate the dynamics of Ψ_L derived from vapor pressure deficit (VPD) and leaf area index (A_L), respectively. The contributions of VPD and A_L to the sensitivity of Ψ_L to Ψs are calculated and further evaluated using the FLUXNET dataset, as to validate the applicability of the extended concept. Then, we suggested a new method to calculate transpiration based on the new framework to establish relationship between Ψ_L and Ψs at ecosystem scale. Our results illustrate that the new framework is reasonable for describing the dynamics of Ψ_L and provides a promising potential for transpiration estimation.

How to cite: Li, C. and Yang, H.: A new perspective for transpiration estimation using an extended framework of isohydricity linking to drought-driven changes in VPD and leaf area, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-1993, https://doi.org/10.5194/egusphere-egu2020-1993, 2020.

The Brazilian Savanna, known as Cerrado (Cerrado sensu lato (s.l.)), is the second largest biome in South America. It comprises different physiognomies due to variations of soil, topography and human impacts. The gradients of tree density, tree height, above ground biomass (AGB) and wood species cover vary according to the Cerrado formation, ranging from different grassland formations (Campo limpo, campo sujo), savanna intermediary formations (Campo cerrado and Cerrado sensu stricto - s.s) and forest formations (Cerradão, Mata ciliar, Mata de galeria and Mata Seca).

Although the carbon stock in Cerrado is lower than in the Brazilian Amazon, the conversion of this biome to other types of land use is occurring much faster. In the last ten years, the degradation of Cerrado forest was the second largest source of carbon emissions in Brazil. Therefore, effective methods for assessing and monitoring aboveground woody biomass and carbon stocks are needed. A multi-sensor Earth observation approach and machine learning techniques have shown potential for the large-scale characterization of Cerrado forest structure.The aim of this study is to present a method to estimate the AGB of an area of the Brazilian Cerrado using ALOS-PALSAR (L-band SAR), Landsat, LIDAR (LIght Detection And Ranging) and field datasets. Field data consisted of 15 plots of 1 ha area located in Rio Vermelho watershed in Goiás-State (Brazil). We used a 2-step AGB estimation (i) from the field AGB using LIDAR metrics and (ii) from LIDAR-AGB to satellite Earth Observation scales following a Random Forest regression algorithm.  The methodology to estimate ABG of Cerrado Stricto Sensu vegetation is part of the Forests 2020 project which is the largest investment by the UK Space Agency, as part of the International Partnerships Programme (IPP), to support in the improvement of the forest monitoring in six partner countries through advanced uses of satellite data.

How to cite: da Conceição Bispo, P., Rodriguez-Veiga, P., Zimbres, B., do Couto de Miranda, S., Henrique Giusti Cezare, C., Fleming, S., Baldacchino, F., Zanin Shimbo, J., Auxiliadora Costa Alencar, A., Roitman, I., Bustamante, M., Pacheco-Pascagaza, A. M., Gou, Y., Roberts, J., Louis, V., Barret, K., Woodhouse, I., Sousa-Neto, E., P.H.B. Ometto, J., and Balzter, H.: Estimating the Above Ground Biomass of Brazilian Savanna using multi-sensor approach, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-20531, https://doi.org/10.5194/egusphere-egu2020-20531, 2020.

Forest albedo changes with vegetation dynamics, ecosystem demography and succession as different plant species can be characterized by contrasting leaf traits, as well as different allocation strategies. In tropical ecosystems, lianas (woody vines) strongly impact the forest biogeochemical cycles by competing with co-occurring trees for above- and below-ground resources. In addition to the particular location of their foliage (often at the top of the canopy), lianas were shown to contrast with tropical trees in terms of biochemical and structural leaf properties, as well as allocation strategies. As a consequence, liana spectral signature differs in several regions of the leaf spectrum (visible, near infrared and shortwave infrared) from tree’s one. At larger scale, the forest canopy reflectance spectrum is also affected by the relative abundance of lianas.

To evaluate the impact of lianas on the radiative transfer and albedo of tropical forests, we collected all published reflectance spectra of liana and co-occurring tree leaves as well as the canopy reflectance spectra characterized by high and low liana coverage. We then calibrated both a leaf (PROSPECT-5) and a canopy (ED2-RTM) radiative transfer model on those data to reproduce the spectral signatures together with their differences, for each single study and site. The Bayesian framework that we used generated leaf biochemical and structural trait distributions, as well as allocation pattern strategies that could be compared between both growth forms.

Collected spectra could be fairly well reproduced at both leaf and canopy levels by the selected mechanistic models. In most studies, calibration led to significantly lower chlorophyll and carotenoid contents, higher relative water content, and larger specific leaf areas for liana leaves, which allowed reproducing observed higher leaf reflectance values in the visible and shortwave infrared and the lower ones in the near infrared.  We then validated our findings with independent field data on leaf chemistry and structure from the same studies.

These calibrated radiative transfer models are tools that can be used in future research to investigate the liana impact on the global energy budget of tropical forests, as well as to monitor spatial and temporal liana abundance.

How to cite: Meunier, F., Shiklomanov, A., Dietze, M., Visser, M., and Verbeeck, H.: The impact of lianas on radiative transfer and albedo of tropical forests, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-22268, https://doi.org/10.5194/egusphere-egu2020-22268, 2020.

Despite their low contribution to forest carbon stocks, lianas (woody vines) play an important role in the carbon dynamics of tropical forests where they compete with free-standing plants for below- and above-ground resources. Doing so, they negatively impact individual tree growth, as well as the net productivity and the long-term carbon storage of the ecosystem.

However, lianas remain largely ignored in field-scale studies as well as modelling forecasts. Therefore, their exact impact on tropical forest biogeochemical cycles is very uncertain. In particular, it is unclear which resource (light, water) is the most competed for between growth forms and so is is the future impact of lianas on forests in a global climate change context in which brighter, drier and CO2-enriched conditions are expected in the Tropics.

To answer those burning questions, we incorporated for the very first time a plant functional type accounting for the lianescent growth form into a dynamic global vegetation model (ED2). We implemented several liana-specific processes in the modelling framework (climbing, resprouting, height limitation due to lack of self-supporting tissues etc.), and integrated liana-specific parameters according to data from multiple studies in order to account for significant differences of functional and structural traits between lianas and trees. These parameters included (but were not limited to) leaf biochemical and photosynthesis properties, stem hydraulic traits, root distribution, and allometric relationships.

Baseline runs successfully reproduced ecosystem gas exchange fluxes (GPP and latent heat), forest structural features (LAI, AGB), and several other benchmarking observations in multiple tropical sites characterized by different rainfall regimes and levels of liana abundance. In those simulations, lianas negatively reduced forest productivity and total carbon storage, by increasing tree mortality (+ 30% on average) and decreasing tree growth (-35%). The inclusion of lianas in the simulations reduced the forest net productivity by up to 0.5 tC ha⁻¹ year⁻¹, which resulted in significantly reduced accumulated above‐ground biomass by up to 20 tC/ha in regrowth forests. The negative impact of lianas on carbon storage almost disappeared in wetter, old-growth forest sites. Model uncertainty analyses also revealed that water limitation was the dominant factor driving competition between trees and lianas, even in sites with a short dry season.

These two-key findings (higher impact in regrowth forests and water-dominated competition) are expected to lead to a reinforcement of the negative impact of lianas on forest productivity under future aggravated forest disturbance and warmer climate conditions. The modelling workflow also allowed to identify key liana traits (quantum efficiency, stomatal regulation parameters, allometric relationships) and processes (water use, respiration, climbing) driving the overall model uncertainty. They should be considered as priorities for future data acquisition and model development to improve predictions of liana-infested forest carbon dynamics.

How to cite: Meunier, F., Dietze, M., di Porcia e Brugnera, M., Longo, M., and Verbeeck, H.: Modelling the impact of lianas on the biogeochemical cycles of tropical forests, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-13740, https://doi.org/10.5194/egusphere-egu2020-13740, 2020.

Savannas are a major component of the world’s vegetation and cover a land surface of about 15 Mkm², accounting for about 30% of the terrestrial primary production. In the South America, the Brazilian Savanna (Cerrado) is the second largest biome (2 Mkm²), after the Amazon biome, and a hotspot of biodiversity. The Cerrado region is heterogeneous, with savanna vegetation ranging from open grassland, through a gradient of increasing tree density to nearly closed-canopy woodland. The cerrado vegetation is markedly seasonal in phenology and is often burned, either naturally or as part of a management cycle. Due its large occupation, Cerrado have the potential to influence the regional and possibly the global energy, water and carbon (C) balances. The allocation of the net primary productivity (NPP) of an ecosystem between canopy, woody tissue and fine roots is an important descriptor of the functioning of an ecosystem, and an important feature to correctly represent in terrestrial ecosystem models for carbon rates estimation, as well as their residence time, variation with climate and disturbance, and in order to make better forecasts. Such estimation in Cerrado regions remains still difficult given the lack of important soil and vegetation data. Previous studies have showed that the fluxes of water and C are closely related to each other, and to the diurnal cycle of solar radiation. However, there is no study clearly assessing the allocation of C through the different types of vegetation, either in the different types of physiognomies. To help estimating the C flows across the different C pools and types of vegetation, we are using Carbon Data Model Framework (CARDAMOM) which is a computer programme that retrieves terrestrial carbon (C) cycle variables by combining C cycle observations with a mass balance model. CARDAMOM produces global dynamic estimates of plant and soil C pools, their exchanges with each other and with the atmosphere, and C cycling variables for processes driving change. It also produces a C cycle analysis consistent with C measurements and climate, and it is suited for using with global-scale satellite observations such as aboveground biomass (ABG) or leaf area index (LAI). For that, we count on field data available (AGB, BGB) and satellite data (LAI, AGB, soil C), which will help to present robust analyses of C cycling across gradients of biomass in the Brazilian Cerrado.

How to cite: Sousa-Neto, E., Smallman, L., Ometto, J., and Williams, M.: Carbon dynamics in the Brazilian Cerrado: stocks and fluxes estimated by a model data fusion framework (CARDAMOM), EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-20772, https://doi.org/10.5194/egusphere-egu2020-20772, 2020.

Forest soils are storing large quantities of carbon, but their quantitative role in sequestering C is less certain. In principal, soils developed over millennia are assumed to be ‘in equilibrium’ with minimal C stock changes. This concept is challenged by forest soil inventories (in Germany and France) indicate a substantial increase in soil C storage. However, soil organic matter (SOM) storage is susceptible to recent changes in forests - climate warming and droughts, increasing forest disturbances, and a more intensive forest management are all potentially increasing SOM turnover which may turn forest soils into C sources. Here, I will critically discuss the role in Swiss forest soils as C sinks by presenting data from 1000 soil profiles across environmental gradients and from flux measurements in large scale ecosystem manipulation experiments.

Swiss forests soils are among the C-richest soils in Europe storing on average 140 t C/ha. Analysis of 1000 forest soils show that these SOM stocks are caused by their high contents in potential SOM sorbents (pH, Al+Fe-oxides, Ca, clay), but also by the cool temperatures and high amounts of precipitation. Climate manipulation experiments suggest Swiss forest soils are vulnerable to loose C with expected climatic changes. A six year long soil warming experiment at treeline revealed soil C losses, while a 15 year long irrigation experiment in a dry forest induced C gains in the mineral soil, implying that a warmer and more frequent droughts will lead to C losses.

Switzerland - as other European mountainous areas – is currently experiencing a major change in land-use due to land abandonment, with the forests expanding by 3 to 4% per decade. Forest expansion affects a multitude of factors driving SOM cycling and storage, including the quantity and quality of organic matter inputs above and below the ground, a cooler and drier microclimate, and change in microbial diversity and activity. In contrast to the intuitive assumption that forests expansion leads to C gains in soils, measurements along an afforestation chronosequence of alpine grassland show that forest expansion leads to minimal changes in SOM stocks but a strong change in SOM quality. Soils gains in particulate organic matter with increasing forest age but lose C in mineral-associated organic matter. In support, reconstructing forest cover ages of 850 soil profiles showed that forest age and hence time since conversion into forest (predominantly from grasslands) did not significantly affect total SOM stocks, while other factors, especially physico-chemical soil characteristics and climate were more important. Overall, these results show that the inherently C rich forest soils in Switzerland are unlikely to gain additional C but rather loose it in response to the ongoing changes in climate and land-use.

How to cite: Hagedorn, F., Gosheva, S., Zimmermann, S., and Gavazov, K.: Soil organic matter dynamics in changing forests: linking ecosystem manipulation experiments with soil inventories across Switzerland, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-18241, https://doi.org/10.5194/egusphere-egu2020-18241, 2020.