The African Tropics are hotspots of modern-day land use change and are, at the same time, of great relevance for the cycling of carbon (C) and nutrients between plants, soils, and the atmosphere. However, the consequences of land conversion on biogeochemical cycles are still largely unknown as they are not studied in a landscape context that defines the geomorphic, geochemical, and pedological framework in which biological processes take place. Thus, the response of tropical soils to disturbance by erosion and land conversion is one of the great uncertainties in assessing the carrying capacity of tropical landscapes to grow food for future generations and in predicting greenhouse gas fluxes from soils to the atmosphere and, hence, future earth system dynamics.
Here we describe version 1.0 of an open-access database created as part of
the project “Tropical soil organic carbon dynamics along erosional
disturbance gradients in relation to variability in soil geochemistry and
land use” (TropSOC). TropSOC v1.0 (Doetterl et al., 2021,
The hierarchical and interdisciplinary structure of the TropSOC database allows linking of a wide range of parameters and observations on soil and vegetation dynamics along with other supporting information that may also be measured at one or more levels of the hierarchy. TropSOC's data mark a significant contribution to improve our understanding of the fate of biogeochemical cycles in dynamic and diverse tropical African (agro-)ecosystems. TropSOC v1.0 can be accessed through the Supplement provided as part of this paper or as a separate download via the websites of the Congo Biogeochemistry Observatory and GFZ Data Services where version updates to the database will be provided as the project develops.
Tropical ecosystems provide many services of global importance. Tropical forests are among the largest terrestrial carbon (C) reservoirs and show some of the highest levels of biodiversity (Losos and Leigh, 2004; Pan et al., 2011). At the same time, tropical landscapes are among the most dynamic regions worldwide and hotspots of modern day land use change (Hansen et al., 2013) as they have to provide food for some of the poorest yet fastest growing populations on the planet. In particular, the African continent is facing huge environmental and societal challenges with a projected population growth of 400 % by the end of this century (Gerland et al., 2014), much of it happening in (sub-)tropical sub-Saharan Africa. In consequence, forested landscapes in tropicalAfrica are currently facing unprecedented levels of land conversion and land degradation, accompanied by decreasing soil fertility (UNESCO and WHC, 2010). At the same time, unlike other tropical regions of the world where deforestation is driven by the extension of commodity plantations and commercial logging, much of the deforestation in tropical African countries is driven by smallholder farms that apply slash and burn practices for subsistence farming with little alternatives to provide food for their families (Curtis et al., 2018; Tyukavina et al., 2018). As a result, deforestation and soil degradation have accelerated greatly since the second half of the 20th century, with soil erosion in particular, emerging as the main driver of soil degradation.
Today, erosion rates of tropical agricultural land globally are estimated at approx. 10.4 billion tons of soil per year and 0.2 billion tons of C per year. Tropical agricultural soil erosion therefore represents about half of the annual agricultural erosion globally while only representing about one third of global cropland (Doetterl et al., 2012). An exemplary region to observe the consequences of land use change on soil resources and biogeochemical cycles in the tropical African regional context is the African Great Lakes region and, in particular, East African Rift Valley system along the borders between the Democratic Republic of the Congo (DRC), Burundi, Rwanda, and Uganda.
The region is a model for the complex interplay of socioeconomic factors and their consequences for environmental systems in the tropics. One of the highest human fertility rates globally (e.g., recent estimates for the last decade range from 7.3 to 7.7 children per woman in the province of South Kivu, eastern DRC) (Dumbaugh et al., 2018) is leading to massive population growth in the region, largely relying on local food and energy resources. Ridden by conflict and open warfare in the 1990s and early 2000s, population growth in the region is further aggravated by refugees from remote areas settling in nearby safer, larger cities in the region (Kujirakqinja et al., 2010). Consequently, massive deforestation of upland forests for firewood and cropland expansion is taking place (Hansen et al., 2013), leading to large erosional soil fluxes and consequential soil degradation threatening soil quality (Karamage et al., 2016). Once conversion to agricultural land has taken place, soil conservation measures could counteract the loss of soil quality (Veldkamp et al., 2020). However, these measures are rare in the eastern Congo Basin due to the poverty of subsistence farmers, socioeconomic instability, and a lack of governmental intervention (Heri-Kazi Bisimwa and Bielders, 2020). Soil tillage and harvesting further degrade the nutrient-containing litter and topsoil layers. Consequently, fields often have to be abandoned after only a decade of use and recover poorly (Carreño-Rocabado et al., 2016; Ewel et al., 1991; Hattori et al., 2019; Heinrich et al., 2020; Kleinman et al., 1996; Lawrence et al., 2010).
With the expansion of cropland into forested landscapes, soil erosion rates are expected to continue to increase. Soil erosion will undoubtedly have an impact on biogeochemical cycles and will change the input, storage, and exchange of C between soils and the atmosphere, as well as the flux of nutrients between plants and soils in tropical systems in the region. To understand how tropical soils and ecosystems respond to erosional disturbance, it is necessary to consider the combined effects of climate, geology, topography, soil formation, biological processes, and human disturbance. To date, no study on the interrelationship of these controls on biogeochemical cycles has been carried out in tropical ecosystems. However, studies carried out in other regions have shown that controls on soil C dynamics, for example, are highly interlinked (Doetterl et al., 2015; Hobley and Wilson, 2016; Nadeu et al., 2015).
Soil redistribution as a consequence of erosion also changes the functionality of landscape units. For example, soil degradation on hillslopes is matched by a buildup of sediment deposits in valley bottoms where C and nutrient-rich soil can be rapidly buried in subsoils under new sediments. While this consequence of deforestation can lead to an increase in the residence time of C due to slower microbial C turnover in buried soil (Van Oost et al., 2007; Doetterl et al., 2012; Alcántara et al., 2017), important nutrients are now lost to plants, leading to a decrease in biomass productivity (Veldkamp et al., 2020) but also to a general degradation of tropical forest soils, lowering also microbial activity in soils (Sahani and Behera, 2001). Soil redistribution is also known to change the temporal and spatial patterns of soil weathering and affects C stabilization. In agricultural systems, the effects of this pressure can be observed very clearly: erosion removes weathered soil from eroding slopes but also brings the soil weathering front into closer contact with the C cycle (which occurs primarily in topsoils), thereby affecting carbon, nitrogen, and phosphorus (CNP) cycling and the stabilization of C with minerals in these systems (e.g., Berhe et al., 2012; Park et al., 2014; Doetterl et al., 2016).
Concerning feedbacks on biogeochemical cycles between soil weathering, erosion will differ significantly not only between natural and disturbed systems but also between systems with differing soil mineral reactivity. Recent advances have shown that mineral reactivity, constrained predominantly by soil weathering and the mineralogy of the soil parent material, has direct control over soil organic carbon, with climate exerting only indirect control through its impact on biogeochemical processes and matter fluxes (Doetterl et al., 2015; Tang and Riley, 2015). However, the exact effects of mineralogy on the temperature sensitivity of microbial decomposer communities and the primary productivity of ecosystems have, to date, not been constrained (Hahm et al., 2014; Tang and Riley, 2015).
Tropical Africa is expected to experience great changes to both soil biogeochemical cycling and ecosystem-level carbon (C) fluxes between soil, plants, and the atmosphere with unknown consequences for biogeochemical cycles. Despite decades of recognizing their importance, tropical soils remain among the least studied in the world (Mohr and van Baren, 1954; Mohr et al., 1972; Ssali et al., 1986; Juo and Franzluebbers, 2003). Although a more complete understanding of soil–plant coupling in tropical environments is critical, most of our process understanding of biogeochemical cycling between plant and soil is still derived from temperate regions. However, due to differences in their environmental setting and soil forming history, many tropical soil systems will likely react very differently to soil disturbance and land conversion than temperate soil systems. For example, temperate ecosystems can differ fundamentally in the way nutrients cycle and in the dominating and limiting factors for plant growth (Du et al., 2020). In contrast to soils in the temperate zone, long-lasting chemical weathering has led to a massive depletion of mineral nutrients from soils in many tropical systems, although the remaining available nutrients are very efficiently recycled in natural tropical biospheres (Walker and Syers, 1976; Vitousek, 1984). Hence, any loss of nutrients is therefore a critical disturbance with direct effects on the functioning of tropical (agro-)ecosystems. Recent studies highlight the importance of soil degradation and the change in chemical soil properties that follows land conversion on plant communities in tropical systems (Bauters et al., 2021), organic matter turnover by microbial decomposers (Bukombe et al., 2021), and the stabilization of C and nutrients in soil of varying mineralogical properties (Reichenbach et al., 2021).
Improving our process understanding of the coupling between soil biogeochemistry and plant responses in the context of tropical land use changes will help to better constrain and define plant–soil interactions in ecosystem and land surface models. Furthermore, insight into plant–soil interactions can help to better inform policy makers and stakeholders in improving land management practices.
In the following we aim at providing an overview of the data collected by project TropSOC (Tropical soil organic carbon dynamics along erosional disturbance gradients in relation to variability in soil geochemistry and land use) which is now available to the research community as an open-access database. We give a brief description of the project's design before elaborating on the structure of the database and its content. Note that beyond the overview information presented here, more details on methods and sampling designs for each assessed parameter are explained in greater detail in the Supplement accompanying the paper and the database.
The main objective of project TropSOC was to develop a mechanistic understanding of plant and microbial process responses to changing soil properties in the African Tropics, exemplified along land use and erosional and soil geochemical gradients studied in the Congo and the Albertine Rift. Trying to understand biogeochemical cycling affected by human activities in tropical (agro-)ecosystems as a whole, TropSOC had two main foci:
investigate how nutrient fluxes and organic matter allocation between
tropical soils and plants differ in relation to the controlling factors of
geochemistry, topography, and land use; investigate how the geochemistry of soils and their parent material
control, interact with, or mediate the severity of erosional disturbance on C
cycling in tropical soils.
In order to address these objectives, project TropSOC investigates effects on tropical soil biogeochemical cycling and biological responses to variation in soil and environmental properties along three main vectors (Fig. 1): (i) mineralogy of parent material, since it may drive the geochemical features of developed soils which control soil fertility and the potential of soils to stabilize organic matter and nutrients; (ii) landform, since topography may influence water and soil fluxes, particularly erosional soil loss on slopes and soil deposition in valleys; and (iii) vegetation and land cover, since it may control the input to and extraction of organic matter from soil and respond to variation in soil properties and hydrology, as well as mediate the impact of rainfall to induce soil erosion.
Factorial design of project TropSOC studying biogeochemical cycles in Central African tropical forest and agricultural landscapes in relation to mineralogy, landform, and land cover types.
Conducted in one of the hotspots of global change – in the Central African Congo
Basin and African Great Lakes region – the database described here is the
foundation for several papers published as a part of the 2021 special
issue
The study area of TropSOC is located in the eastern part of the Democratic
Republic of the Congo, Rwanda, and Uganda, in the border region between the
Congo and the Nile Basin (Fig. 2). It is largely understudied (Schimel
et al., 2015) despite its great significance for the global climate system
(Jobbágy and Jackson, 2000; Amundson et al., 2015) and is being confronted
with rapid land conversion and forest degradation (Hansen et al., 2013).
The climate of the study region is classified as tropical humid with weak
monsoonal dynamics (Köppen Af–Am) and mean annual temperatures (MAT)
ranging between 15.3 and 19.3
Overview of the study region with respect to major
investigated factors: soil parent material geology and geochemical regions
As a part of the East African Rift mountain system, the active tectonism within the study region produced a hilly, patchy landscape with steep slopes up to 60 % and soil parent material ranging from volcanic ashes to mafic and felsic magmatic rocks, as well as a sedimentary rocks of varying geochemistry and texture (Schlüter and Trauth, 2006) (Fig. 2a, b).
The study area is dominated by agricultural land use, with larger patches of
protected, old-growth closed-canopy forest in highland areas (Fig. 2c).
Typical crops planted for subsistence farming are rotations of cassava
(
Within the study area three regions each representing a geochemically different parent material for soil formation were determined. The first region (Fig. 2a) is predominantly situated on mafic magmatic rocks, typically mafic alkali basalts (Schlüter and Trauth, 2006), resulting from extinct (Mount Kahuzi) and active (Mount Nyiragongo) volcanic activities between the cities of Bukavu and Goma, Kivu, DRC. The second region is situated on felsic magmatic and metamorphic rocks typically consisting of gneissic granites (Schlüter and Trauth, 2006) near the city of Fort Portal on the foothills of the Rwenzori Mountains, Uganda. The third region is situated on a mixture of sedimentary rocks of varying geochemistry consisting of alternate layers of quartz-rich sandstone, siltstone, and dark clay schists (Schlüter and Trauth, 2006) and spread across the western province of Rwanda in and around the district of Rusizi.
The dominant soil types of the study region are various forms of deeply weathered tropical soils (FAO, 2014). Potential ash deposition through the region's active volcanism occurs frequently, re-fertilizing soils to various degrees. Following World Reference Base (WRB) soil classification (IUSS, 2014), soils in the mafic region can be described as umbric, vetic, and geric Ferralsol and ferralic vetic Nitisol. Soils in the mixed sedimentary rock region and the felsic region can be described as geric and vetic Ferralsol. Soils in valley bottoms can locally show gleyic features, where the dominating soil types are variations of fluvic Gleysol.
Several striking differences in the elemental composition of the three
parent materials can be noted. In the mafic region, bedrock is characterized
by high iron (Fe) and aluminum (Al) content, as well as a comparably high
content of rock-derived nutrients such as base cations and phosphorus (P).
The felsic and sedimentary rock regions are characterized by lower
contents of Fe and Al, by lower rock-derived nutrient contents, and by
higher Si content (Fig. 3). A specific feature of the
sedimentary site is the presence of fossil organic C in the parent material
of soils ranging between 1.29 % and 4.03 % C. Fossil organic C in these
sediments is further characterized by a high C : N ratio (mean
Chemical composition of unweathered rock samples
representing the parent material for soil formation in three studied
geochemical regions (mean
Soil chemical composition of subsoil in stable, old-growth closed-canopy forests (no erosion) in the three investigated
geochemical regions (mean
In summary, the study region provides a unique combination of (i) near-pristine forest and agricultural land, (ii) steep terrain and heavy tropical precipitation with high erosion potential, and (iii) geologically diverse parent material for soil formation. These factors make the study region ideal for identifying the importance of various controls on tropical soil biogeochemical cycles.
Plots were established along geomorphic gradients in old-growth closed-canopy forest, as well as in cropland, in all three geochemical regions. Field
campaigns to collect soil and plant samples at 136 forest and cropland plots
along slope gradients (catena and stratified random approaches) and
additionally within several cropped nearby micro-catchments were carried out
between March 2018 and July 2020. A detailed description on data quantity
and quality can be found in the metadata files accompanying the database and
are briefly described in Sect. 4.1 of this publication. In order to cover
potentially stable, eroding, and depositional landforms, topographic
positions of plots ranged from plateaus (slope
Topographic information of TropSOC plots across different geochemical regions and land use. Slope and altitude are displayed as minimum and maximum values. Each topographic position per geochemical region contains the range between three and seven field replicate plots.
Sampling in forests followed a strict catena approach, and plots were
established following an international, standardized protocol for tropical
regions (Phillips et al., 2016). Within each geochemical region, three plots
covered by old-growth closed-canopy tropical forest vegetation (forest that
developed a complex structure characterized by large, living and dead trees)
were established from February to June 2018 per topographic position as field replicates representing
an area of 40 m
Overview of forest
At the time of plot installation, four replicate soil cores per plot (one in each subplot) were taken in a depth-explicit way in 10 cm increments up to 1 m soil depth and combined as composites per plot. In addition, one soil profile pit was dug to a depth of 100 cm in the center of one of three replicate plots (Fig. 5) per topographic position in each geochemical region. These soil pits were dug and described according to FAO guidelines (FAO, 2006).
Leaf litter (L horizon) and partially decomposed organic material in O
horizons were sampled at eight points along the border and in the center of
each forest plot (Fig. 5a). At each sampling
point, the thickness of the L and O horizon layer was measured with a ruler
and then sampled within a 5 cm
All collected composite samples were kept cooled until being brought to the
laboratory (usually within 48 h). In the laboratory, samples were
oven-dried at 40
In 2018, full inventories of the forest tree species and standing
aboveground biomass (AGB) were conducted on all forest plots. The forest
inventory followed an international, standardized protocol for tropical
regions (Matthews et al., 2012). First, we identified, to species level, all
living trees with a diameter at breast height (DBH; measured at 1.3 m above
ground) greater than 10 cm in each plot. Second, these identified trees were
classified into the following empirical DBH classes: 10–20, 20–30, 30–50, and
To assess plant functional traits (leaf nitrogen, phosphorus, potassium,
magnesium, and calcium content) of living canopy leaves (see Sect. 2.7), we
sampled, at the beginning of the weak dry season (December–February),
sun-exposed shoots from the outer canopy of selected tree species that
collectively make up 80 % of the standing basal area per plot with the
help of trained tree climbers and following a sampling protocol described in
Pérez-Harguindeguy et al. (2016). For every tree species, we selected at
least 3 individual trees, and a minimum of 5 and maximum of 17 trees per
plot were sampled for mature, healthy-looking (
Plots on cropland were established following a stratified random approach using the same slope classification and selection criteria as for forest sites. However, cropland plots belonging to the same geochemical region and topographic position were not connected along a hillslope catena. On cropland, only fields that were currently covered by cassava were sampled. Cassava fields were chosen since cassava is one of the most important food crops in the region and is harvested for both tubers and leaves. Rotations of cassava, maize, pulses, and vegetables are common throughout the area, and two harvests are possible per year. The main varieties of cassava on our sites were Mwabailon, Nabiombo, Mwamizinzi, Sawasawa (in eastern DRC), Bukalasa, Shayidire, Gitamisi, Amaduda (in Rwanda), Sambati, and Mubalaya (in Uganda). Only fields without soil protection measurements (i.e., terraced systems) were sampled. For an overview of forest plot sampling design, see Fig. 5b.
Soil sampling was carried out in the same way as for forest soils with the
exception that only two cores were combined per plot taken within a 3 m
As part of the regional stratified random sampling design for cropland plots
(see cropland plot installation), biomass from different cassava varieties
was collected for 65 plots out of the 100 sampled cropland plots. Biomass
was sampled shortly before harvest, approximately at the time of the plant
tuber's maximum development. The timing of harvest differed between 12 and 24
months after planting depending on the variety and season. Within each plot,
a 3 m
Farmers were sent a questionnaire to collect information on the land use and management history of sampled fields following McCarthy et al. (2018). This questionnaire was completed for a corresponding total count of 87 out of the 100 sampled cropland plots.
Three weather stations (ATMOS 41, Meter, Germany) were installed in August
2018, one in each geochemical region of project TropSOC close to the investigated
forest catenae (mafic: latitude
Litterfall was assessed following a standardized protocol to measure
tropical forest carbon allocation and cycling (Matthews et al., 2012). At
each of our 36 forest soil sampled plots, 10 litter traps were installed and
distributed evenly and systematically per plot. These had a diameter of 60 cm each and were installed at a height of 1.0 m above ground. Litter samples
were collected every two weeks for the period between August 2018 and
February 2020 and later aggregated to assess seasonal and annual
variability in litter productivity and quality (see Sect. 2.4). Collected
litter included all organic residues collected by the traps. Larger dead
animals and woody material
For all soil-sampled forest plots, standing root biomass and fine root
production were assessed from September 2018 to December 2019. Sampling took
place once per season within this period (one coring every 3 months), and
a total of three rainy seasons and three dry seasons in 2018 and 2019 were
covered. Each plot was divided into four equally sized subplots of 20 m
Belowground standing root biomass was sampled using a soil core sampler
(Vienna Scientific Instruments, Austria). Two cores were sampled per subplot
in which undisturbed soil cores were divided into five depth layers: one
organic soil layer (O horizon), and four mineral soil layers from 0–10, 10–20, 20–30, and 30–50 cm. After transport to the
laboratory, each sample was rinsed inside a 2 mm sieve; roots were separated
into fine roots (
Fine root net primary productivity was assessed using the ingrowth net
method following Ohashi et al. (2016). Two net sheets (polyester mesh
aperture size 2 mm, 10 cm wide, 20 cm high) were installed per subplot in a
regular pattern with a distance of approximately 1 m between the two nets.
Each net was vertically inserted in the top 20 cm of soil starting from the
surface of the mineral layer. Nets were sampled every 3 months after
installation and seasonally four times a year from September 2018 to
December 2019. Data are provided as grams per square meter (g m
A wide range of chemical and physical parameters were assessed for the sampled soil and plant material with the aim to characterize (i) indicators of soil redistribution, (ii) the degree of soil weathering, and (iii) the physical structure of soil, as well as (iv) soil fertility and (v) soil organic carbon characteristics, in order to link them to (vi) functional traits of the sampled biomass, (vii) biomass production, and (viii) land management. For a full overview of all assessed parameters including their assessment methods, please consult the metadata accompanying the database.
Among others, key measured parameters encompass the following:
Soil bulk density Soil texture Soil water holding capacity
Soil pH (KCl) Soil potential cation exchange capacity and its base saturation Soil effective cation exchange capacity and its base saturation Main elemental composition of bulk soil (Al, Fe, Mn, Si, Ti, Zr, P) and the
total reserve in base cations (Ca, Mg, Na, K) in rock parent material, soil,
litter, and vegetation samples Pedogenic oxide concentration (Al, Fe, Mn)
Dissolvable soil organic nitrogen and carbon Plant available phosphorus in soil
Total and organic carbon and nitrogen content in rock parent material, soil,
litter, and vegetation samples Bulk soil radiocarbon signature C : N ratio in soil, litter, and vegetation samples Soil carbon stabilization mechanisms
Heterotrophic soil respiration (including isotopic signature of respired
gas) Microbial biomass during incubation Extracellular enzyme activity during incubation
239
All of the parameters listed above have been measured in soil for three depth layers (0–10, 30–40, 60–70 cm) representing distinct sections of the soil profile. Physico-chemical key properties of the remainder of the soil samples in other soil layers have been assessed using mid-infrared spectroscopy and predicted following the workflow of Summerauer et al., 2021). An overview of chemical and physical key soil parameters is provided in Appendix Table A1. Note that all physico-chemical soil properties and the corresponding mid-infrared data are part of the central African spectral library (Summerauer et al., 2021), which minimizes the need for future traditional soil analyses.
Overall a total of approximately 2100 soil and rock samples were collected,
of which about 10 %–30 % were used for yet more detailed analyses in
different experiments by our group (see below). Additionally, 6000 above-
and belowground biomass and litter samples were taken during several
sampling and monitoring campaigns at forest and cropland sites. Several
thousand near-infrared and mid-infrared (NIR-MIR) spectra in the wavenumber range 600 to 7500 cm
Analyses conducted on collected samples so far contributed to scientific advances realized through
the creation of a data frame of reference samples for calibration used in
the newly developed soil spectral library for central Africa (Summerauer et
al., 2021) an investigation of the role of geochemistry and geomorphic position for
soil organic matter stabilization mechanisms and patterns of soil organic carbon (SOC) stocks in
tropical rainforests (Reichenbach et al., 2021) an investigation of the role of geochemistry and geomorphic position on the
heterotrophic soil respiration (Bukombe et al., 2021) an assessment of the suitability and the application of radioisotope
soil fractionation and incubation experiments encompassing cropland soils
along geomorphic and geochemical gradients (unpublished).
As part of this paper, the entirety of TropSOC's data is available as an open-access database with extensive metadata documenting experimental approaches, framing of the analyses, data quality, and methodology. An overview of all datasets presented in this database is given in Appendix Table A2.
In summary, TropSOC's first results demonstrate that even in deeply weathered tropical soils, parent material has a long-lasting effect on soil chemistry that can influence and control microbial activity, the size of subsoil C stocks, and the turnover of C in soil. Soil parent material and the resulting soil chemistry need to be taken into account in understanding and predicting C stabilization and turnover in tropical forest soils. Given the investigated rates of erosion on cropland, our findings confirm the threat of large losses of organic matter leading to a sharp decline in soil fertility with little potential for soils to recover from nutrient losses naturally on decadal or centennial timescales. TropSOC highlights that considering feedbacks between geochemistry and topography to understand the development of soil fertility in the African Great Lakes regions can significantly improve our insights into the role of tropical soils to reach several key sustainable development goals, such as climate mitigation, zero hunger, and to help raise awareness of the need to maintain sufficient soil resources for future generations. Future work realized in project TropSOC based on the database will provide further insights into biomass and plant trait responses to soil geochemistry in forests, as well as cassava yield responses and SOC dynamics in cropland along the investigated geomorphic and geochemical gradients across the region.
Datasets are given as tab-delimited .csv files. For each .csv file the metadata describing data structure and assessment methods are given in a .pdf file of the same name. Moreover, additional .pdf files for each main section of the database (basic information, forest, cropland, and microscale meteorology) are given, providing an overview of the structure within each section. Note that the “basic information” section of the database provides the linkages between individual data points, e.g., from soil analysis and the location and/or soil depths where these samples were acquired (for linkages, see also Fig. 6).
Overview of linkages between datasets in the TropSOC database v1.0. Note that for each data .csv file a .pdf file is given detailing the metadata of the respective data sheet.
The database comprises basic information of all plots and single point sampling positions where data were collected during project TropSOC. An overview of the structure of the database is presented in Appendix Table A2. The basic information of the database is structured in the following way.
The key element to link all data tables for which data were collected and
analyzed is the plot ID and its derivative, the sample ID. This
identifier allows us to link the results from sample analysis with the
locations given in
TropSOC's forest data consist of seven parts (Table A2 for overview)
structured as paired .csv–.pdf files containing the data (.csv) and
accompanying metadata (.pdf) describing parameters and methods.
Additionally, an overview of all collected forest data is given in file
TropSOC's cropland data consist of the following seven parts (Table A2 for
overview), structured as paired .csv–.pdf files containing the data (.csv)
and accompanying metadata (.pdf) describing parameters and methods.
Additionally, an overview of all collected cropland data is given in file
The meteorological data comprises four parts (Table A2 for overview), structured as paired .csv–.pdf files containing the data (.csv) and accompanying metadata (.pdf) describing parameters and methods.
The current version, v1.0, of TropSOC includes several thousand individual plant and soil samples collected across 136 sites, spanning cropland and forests in the East African Rift Valley system and a large variety of parameters. A total of 36 .csv data sheets are available that give all analyses completed for specific samples. Data sheets are structured according to the descriptions given in Sect. 3 and described and elaborated on in the accompanying metadata files. The current distribution of data points across the various levels of the database hierarchy is shown in Table 2. All individual data entries present in the database have passed quality control completed by experts who were involved in the creation of the data. When applicable, reports on the quality assessment of each parameter can be found in the metadata .pdf files accompanying the .csv files.
Overview of the current number of data points in TropSOC
v1.0 on plant, soil, and meteorology and their affiliation to the
hierarchical levels of forest and cropland. Numbers in tables refer to the
number of data entries at the lowest available aggregation level (
Users may access the TropSOC database v1.0 and its supporting information
through the Supplement provided as part of this submission.
Version 1.0 of the database is also available through the data download
section of the Congo Biogeochemistry Observatory (CBO) (
Users are encouraged to provide feedback and corrections to existing data if problems are discovered by contacting CBO (contact@congo-biogeochem.com) or the corresponding author of this paper (sdoetterl@usys.ethz.ch). Corrections will be implemented in consecutive versions of the database that can be downloaded via the CBO site.
Updated versions of the database will be periodically released following
either substantial changes or new peer-reviewed publications using the
dataset. The versions of these official releases are tracked using an
associated version number, e.g., TropSOC v1.0, and so on. These official
releases will be archived at ETH Zurich's Research Collection via ETH's Soil
Resources Group (
TropSOC is a community effort with multiple contributors operating at different levels (Fig. 7). Governance of TropSOC is required in order to ensure continuity of services and to plan for the future evolution of this data repository. Studying the rapid environmental changes to the African Tropics is a central research objective for the scientists of the Congo Biogeochemistry Observatory (CBO) making it the ideal body to govern future versions of TropSOC. The governance structure of TropSOC is briefly described in Fig. 7. While the TropSOC core team is responsible for the original version of the database, its maintenance, management, and archiving, scientists involved in the CBO oversee the establishment of cooperative agreements in the long term and act as a steering committee for modifications to TropSOC suggested by the research community. The main role of the steering committee is to determine the feasibility of major changes to TropSOC proposed by the community and to coordinate activities that would build upon TropSOC or continue similar research work within the framework of CBO. Although the structure of TropSOC is oriented around individual and research projects, the nature of scientific research is often more group-focused. For example, teams of researchers generally work together to seek out funding and to conduct research. Thus, in some cases a group or team of individuals may seek to utilize or modify TropSOC for their purposes. Such groups can petition the scientific steering committee (SCS) to be formally designated a CBO member group. Approved organizations should nominate a member to serve on the steering committee.
A simplified depiction of the TropSOC governance. The scientific steering committee (SCS) is responsible for approving major management decisions. The TropSOC core team as data maintainers are responsible for implementing broader changes together with new data contributors. All interested scientists are welcome to contribute data to future versions of the database or access the data for their own research.
Interested researchers are also invited to contribute data to future versions of TropSOC in order to grow the database. Anyone can be a data contributor provided they agree to the terms of use and follow the proper steps for contributing data to TropSOC. If such suggestions arise, the CBO steering committee together with the TropSOC core team are responsible for approving the suggested changes and additions to the database. Upon approval, the TropSOC core team will coordinate with the new data contributors to implement the suggested data additions. In the case of organizations or individuals making larger changes or additions to TropSOC, a designated data maintainer from new contributor groups is required to coordinate the technical aspects of the implementation of changes together with the TropSOC core team. Within the pool of data contributors, individuals with significant experience working with TropSOC may be designated by either the steering committee or database maintainers as expert reviewers. These individuals are tasked to assist maintainers and oversee peer review and quality assessment of contributed new entries.
All data presented in this study is part of the publication and added as a
Supplement consisting of data tables (.csv) and accompanying metadata
descriptions (.pdf files). In addition, the database and its metadata are
archived and published in the open-access environmental and geoscience data
repository GFZ Data Services, accessible at
The TropSOC database is an attempt to gather the data used in individual studies in one place and in the same format to facilitate comparisons and synthesis activities. TropSOC is unique in that it includes measurements and monitoring data of bulk soil and vegetation responses in the African tropical context on carefully selected and comparable land use, geomorphic, and geochemical gradients on the landscape scale. Building on the data gathered along these gradients during several years of field activities and carrying out numerous lab experiments to investigate the impact of soil geochemistry and land degradation on biogeochemical cycles in tropical plant–soil systems, TropSOC is, to the best of our knowledge, the largest integrative project database on plant–microbial–soil systems in the Congo Basin to date. TropSOC's open-access database structure and participatory approach makes it a suitable tool for scientists to study experimentally defined soil disturbance and plant responses, as well as to test some of the assumptions behind modeling biogeochemical cycles in land surface models. Furthermore, we hope to encourage the community to increase the effectiveness of that investment and to use the TropSOC database as a repository to increase the impact of their own research results. As such, TropSOC is an interactive database that is open for contributions. In addition, TropSOC now manages one of the largest topically structured soil and plant sample archives for tropical eastern Africa with several thousand samples and more than three tons of plant and soil material stored at ETH Zurich. Subsamples of all the above are available upon request to interested researchers.
Finally, we hope that work based on the TropSOC database can help to provide answers on the role and magnitude of geochemistry, as well as soil mobilization, in controlling biological processes and fluxes of carbon and nutrients in the tropics in order to better constrain soil processes in models ranging from profile to global scales (Todd-Brown et al., 2013). Reducing the uncertainties associated with our understanding of tropical (agro-)ecosystems in diverse but rapidly changing landscapes is one of the most pressing issues for securing the future well-being of hundreds of millions of people and to constrain land loss in an area that is home to some of the last and most fragile populations of great apes in the wild. Further, elucidating the gravity of the consequences for soil functioning that can be observed in TropSOC's study area can contribute to reducing the large uncertainty associated with terrestrial biogeochemical processes in models. Finally, our hope is that studies based on TropSOC's data will also help to raise further awareness for the necessity of creating the socioeconomic fundamentals for sustainable land management in tropical Africa.
Basic chemical and physical soil parameters
aggregated at land use and geochemical regions. Displayed are average values
and standard deviations taken over 10 soil increments at 10 cm from
0–100 cm soil depth derived from NIR-MIR spectral data, calibrated on
samples from three depth increments (0–10, 30–40, and 60–70 cm).
See metadata files 223_soil_spec.pdf and
323_soil_spec.pdf for details. Abbreviations:
CEC
Structure of the TropSOC database. For each topic a .pdf file is given that contains an overview of the available data on soil, vegetation, and weather collected for the investigated forest and cropland plots. Each dataset then comprises a data-containing .csv file and an additional metadata-containing .pdf file of the same name.
Remaining soil and plant samples are logged and barcoded at the Department of Environmental Science at ETH Zurich, Switzerland. As long as idle sample mass remains available, samples for independent research and to stimulate collaboration with the CBO network and the TropSOC project group are available upon request. The authors cannot guarantee that they can revisit the study sites in order to provide additional sample material in future campaigns. Samples will be given to researchers free of charge. Sample preparation and transport are subject to a handling fee.
The supplement related to this article is available online at:
SD functioned as the project leader. SD and PF were lead coordinators for compiling the database, were responsible for data analysis, and designed the metadata. BB, MC, LK, DM, MR, LS, and FW collected and created datasets and also analyzed these data before their inclusion into the database. RKA, FB, MC, CB, AM, MM, JM, SMM, LN, AS, RU, and CV were technical contributors and participated via data collection. GB, MB, PB, GC, LNC, AH, KK, BBM, BR, JS, BV, KVO, and KV were conceptual contributors and participated in the design of the study, as well as by giving advice and feedback during the campaign. SD and PF wrote the paper. All authors supported data analysis and gave feedback during the writing process.
The authors declare that they have no conflict of interest.
Publisher’s note: Copernicus Publications remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This work is part of the DFG funded Emmy Noether Junior Research Group
“Tropical soil organic carbon dynamics along erosional disturbance
gradients in relation to soil geochemistry and land use” (TropSOC). Micrometeorological data from three of our weather
stations (Bukavu, Lukananda, Bugulumiza) were made available and are
administered by the Trans-African Hydro-Meteorological Observatory (TAHMO).
The authors would like to thank in particular the following collaborating
institutions for the support given to our scientists and this project:
International Institute of Tropical Agriculture (CGIAR-IITA), Catholic
University of Bukavu (UCB), Mountain of the Moon University (MMU), Kyaninga
Forest Foundation (KFF), ETH Zurich, and the Max Planck Institute for
Biogeochemistry in Jena. Special thanks goes to the many student assistants
for their important work in the laboratory and all guards, sentinels, and
field work helpers making the sampling campaign possible under difficult
conditions. Lastly, we thank the editors of ESSD as well as the reviewers
Yao Zhang and Rose Abramoff for their valuable insights and comments
during the review process.
This research has been supported by the Deutsche Forschungsgemeinschaft (grant no. 387472333).
This paper was edited by Jocelyn Lavallee and reviewed by Yao Zhang and Rose Abramoff.