Introduction

The African rainforest, is the second largest on Earth, covering ~630 million ha. The rainforest stores up to 66 Pg of carbon and is presently a persistent carbon sink (0.34Pg C yr^-1)¹. In addition, African rainforests support the forestry sector which contributes 3-6% of the gross domestic product (GDP) across the Congo Basin with most foreign export directed to Europe. As such, the African rainforest currently represent both a local and foreign (EU) economic driver. Yet, predictions regarding the future climate and the state of the forest remain uncertain in part due to a lack of legacy data (see Fig. 1) which provides the necessary climatological and ecological context for current research in the Congo Basin (e.g. the BELSPO Congo Basin integrated monitoring for forest carbon mitigation and biodiversity).

Fig 1. - an example climate record listing temperature, precipitation and relative humidity from the INEAC archives at the State Archives.

Long-term eco-climatological records

Our understanding of forest ecosystem responses to climatic change relies heavily on consistent long-term observations^2,3 to provide much needed baseline measurements. However, observing and measuring tropical plant species and the climatic conditions in which they reside is extremely demanding, particularly in the central Congo Basin. Consequently, established long-term observation programmes are rare. For example in terms of meteorological observations, the central Congo Basin is currently represented by only a few rain gauges, seriously limiting climate forecasts across the Congo Basin^4,5. Similarly, few long term forest inventory plots exist for the central Congo Basin⁶ limiting the accuracy of biomass estimates and our understanding of forest structure and function^7,8. This lack of long-term (historical) eco-climatological data leaves the central Congo Basin spatially and temporally under-represented. However, old climate records could provide valuable information about previous growing conditions of the forest.

Indeed, ecologically relevant data such as historical forest inventory plots which describe forest canopy structure, and leaf level traits derived from herbarium specimen, such as stomatal density and specific leaf area, can elucidate how the forest has changed over time in response to past climate change⁹. In particular, studies demonstrated a strong link between plant functional traits and prevailing climatic conditions⁹. As such, valuable insights can be obtained on how plants are affected by climate change from measuring a limited number of (leaf) functional traits. Strong clustering of leaf traits also implies that changes in environmental conditions can cause profound shifts in plant community structure and ecosystem functioning¹⁰. Therefore, functional traits could enable us to predict ecosystem functions, such as primary production, carbon sinks and nutrient pools and fluxes under a changing climate. Although traits are continuously being recorded, how functional traits are changing in time is a topic not often addressed.

Unexplored heritage, and retrospective analysis

There is an urgent need for more long-term eco-climatological (baseline) measurements. Currently, large amounts of ecological and climatological data, approximately five decades (~1910 – 1960), exists as unexplored heritage, stored in various federal archives and collections. Within this context the archives of the Institut National d’Etudes Agronomique du Congo Belge (INEAC) at the State Archives (SA, INEAC March 2013, nr 1546, National Archives), the Royal Museum for Central Africa (RMCA) and the herbarium collections of the Botanic Garden Meise (BGM), with its large collection of tropical African plant specimens and complementary legacy data, are key. The inventory of these archives includes historic forestry, climatological, ecological, biodiversity data and aerial photographs, with great potential and relevance for current and future basic and applied forestry research in the central Congo Basin, particularly within the context of climate change. As of yet, a comprehensive dataset of long-term eco-climatological legacy data for the central Congo Basin is lacking.

Fig. 2 - Map of the D.R. Congo, the Congo river and it's watershed. Major cities are marked in italics. Areas not included in climatological data processing are marked with red cross-hatches (i.e. province of Katanga). Locations of interest regarding herbarium and wood collections and archived data are marked (e.g. Yangambi and Luki). The white polygons at Yangambi / luki outline today's UNESCO Man-and-Biosphere reserves. The white dashed rectangle outlines the approximate coverage of historical aerial photographs of the forest cover and it's canopy structure.

Congo basin eco-climatological data recovery and valorisation

COBECORE brings together an interdisciplinary network of partners, including the main institutes curating eco-climatological legacy data of the central Congo Basin. The global objective of this 4-year project is to establish baseline measurements necessary in long-term ecological and climatological research, valorizing as of yet unexplored heritage. To this end, COBECORE will develop a multi-faceted database, by making the static analog INEAC archives digitally accessible and extracting eco-physiological relevant plant traits from historic BGM herbarium specimens. In order to speed up data processing, and provide highly visible public outreach, we will crowd-source transcription of the climatological data and stomatal density counts.

In particular, this database will include:

1. long-term historical (baseline) climatological measurements throughout the Congo Basin,
2. detailed forest inventories at the Yangambi and Luki research stations,
3. geo-referencing of historic aerial photographs and derived structural canopy properties around the Yangambi research station,
4. leaf and wood traits with eco-physiological importance derived from historical herbarium specimen (e.g. stomatal density, gwmax, anatomical wood features) from the Yangambi and Luki regions. The final database will allow for direct applications using contemporary scientific methods such as climate re-analysis, climate forecasting, forest ecosystem modelling and could guide sustainable forestry policies.

Compliance of the proposal with the scope of the call

There is a pressing need to establish baseline measurements in long-term ecological and climatological research within the Congo Basin (see project summary). The COBECORE project brings together an interdisciplinary network of partners, including the main institutes curating eco-climatological legacy data of the central Congo Basin (i.e. D.R. Congo), to establish these baseline measurements by valorizing as of yet unexplored heritage and realize the potential of eco-climatological legacy data available within the INEAC archives (State Archives), complementary historical archives (RMCA) and data derived of historical herbarium samples of the BGM herbarium.

The project will make information stored in analog archives digitally and easily accessible, which will result in a multi-faceted database of eco-climatological, forest inventory and aerial photography data for direct applications in contemporary scientific methods. Increasing the accessibility of eco-climatological legacy data through COBECORE will improve both global climate and ecosystem models, provide an increased understanding of temporal changes of forest composition and function and contribute to insights into tropical eco-physiology.

Subject of the proposal

State of the art and objectives

The project’s overall goal is to make the wealth of information locked in the static INEAC archives accessible to the international scientific community. In order to raise awareness of the global significance of the INEAC data, the COBECORE project will generate three main data streams, together with data integration grouped in four objectives.

Objective 1. A comprehensive historical climate record across the central Congo basin :

Currently there is no comprehensive database of spatially explicit historical climatological data for the central Congo Basin. This lack of data directly limits different ongoing research efforts. For example the retrospective analysis of wood cores and herbarium specimens relies greatly on co-occurring climatological records. COBECORE will therefore generate added value and aid in the interpretation of results from the HERBAXYLARREDD project (BELSPO BR/143/A3/HERBAXYLAREDD). HERBAXYLAREDD is aimed at linking plant data from the Herbarium of the BGM with wood samples of the Xylarium of the RMCA in Tervuren to foster a better understanding of patterns and processes in the Congo Basin rainforest. Accurate climatological data are also key in assessing the spatially explicit climate sensitivity of tropical tree species as explored in the Jungle Rhythms project (data collection supported by COBIMFO; BELSPO project SD/AR/01A). We acknowledge that a previous effort was made to partially digitize climatological data during the GNOSIS project (BELSPO project I2/2F/213). The GNOSIS database servers have since been taken offline and the data remains encrypted with the password missing (personal communications: S. Kauranne, Systems & Helpdesk KBIN). Regrettably, these data are considered lost. COBECORE will therefore duplicate part of previous research efforts when digitizing climate records stored at the State Archive.

More so, in Objective 4, we explicitly foresee continued access of the mobilized data long after COBECORE is finished. Data gathered during the digitization effort will contribute to digitization efforts in the State Archive such as outlined in the DIGIPAT initiative. Ingestion of the digitized data in State Archive databases will facilitate future digital consultations of the INEAC archives and support historical analysis such as Expo Congo.

Objective 2. Historic metrics of forest structure :

In order to establish baseline measurements of the forest structure within the central Congo Basin we propose to assess forest structure based upon various sources of historical data and methodologies. In particular we will use transcription of archived forest survey data in the INEAC archives, re-analysis of BGM herbarium meta-data, the analysis of geo-referenced historic aerial pictures using textural and land-use land-cover change analysis. Structural analysis of the canopy will provide information on the temporal stability of the canopy structure within the permanent sampling plots (e.g. COBIMFO) and support previous and ongoing research at Luki. Geo-referenced imagery (a pre-requisite before textural analysis) will be contributed to www.rdcmining.org and the Cartesius project (www.cartesius.be), an internal digitization initiative of the RMCA. We like to highlight that Objective 2 has a large overlap with thematic axis 2.2.4 “Management and valorisation of the federal audio, photographic and audio-visual heritage” and 2.2.5 “Geolocation for the valorisation of the historical, cultural and scientific heritage” in this BRAIN call. As such, we will seek to collaborate with projects in both axis where possible.

Objective 3. Historic leaf level traits from herbarium specimen :

In this objective we aim to provide specific leaf area, stomatal density parameters and leaf nitrogen and carbon content (traits) which are important parameters in ecosystem models and indicators of adaptation to climate change in their own right^11,12. Plant functional traits are used in comparative ecology to offer insights in plant ecological strategies¹³. In addition we will cross-link this data with leaf level data collected during the COMBIMFO field campaign, the phenology data extracted during in the Jungle Rhythms project, the work by Couralet et al. (2013)¹⁴ and previous digitization efforts of herbarium specimen at the BGM in the BIOSPHERETRAITS project.

Objective 4. Data integration and dissemination :

In the fourth objective we ensure proper data integration into a database structure and dissemination through a web portal and outreach towards the scientific community, the public and the private sector.

Methods

1. Data selection (Objective 1-3):

Historically, the INEAC had it’s main headquarters in Yangambi. Paper records of other INEAC field stations were centrally collected at the Yangambi research institute, before being duplicated to keep parallel records in the State Archives. The local significance of the research stations also means that a large fraction of the herbarium (and wood) specimens were collected in or around the greater Yangambi region. A similar singular focus exists around the Luki botanical gardens at the Atlantic side of the Congo Basin, and located in Mayombe forest. For practical reasons such as data volumes and the number of herbarium specimens to process, the valorization of the INEAC archives and the BGM herbarium will be carried out by focusing on specific locations and subsets of the respective archives (i.e. Yangambi en Luki, see Fig. 2). Although COBECORE makes a selection of data based upon the eco-climatological value of the data the project still ensures that large quantities of analog archives will become digitally and easily accessible and provide a framework for future more extensive digitization (see sections 3.1 & 3.2).

Currently the Yangambi research station is part of INERA (L’Institut national pour l’Etude et la Recherche Agronomique) and still holds a copy of the data stored in the State Archives. In order to fill in possible gaps of data (e.g. measurements, protocols) at the State Archive, due to incomplete copies, we will partner with the INERA at Yangambi through Resource and Synergies Development (RESYNDE) and their contracting branch in D.R. Congo. Ongoing negotiations concerning long-term collaborations between the INERA Yangambi (with Dr. Limba as acting director) and the RMCA are currently underway, advising full partnership for INERA. However, for practical reasons, notably time restrictions and the unquantifiable estimate of the data gap-filling duties involved before the start of the project, we have chosen to partner with the INERA Yangambi through contracting on an ad-hoc basis. Our partner in contracting will be Dr. Limba and his affiliated staff. During and after the project we will partner with Dr. Limba to disseminate the dataset generated through the COBECORE project. Within our project we will guarantee the transfer of knowledge of data management and digitization, in return for an intricate knowledge of the INERA-Yangambi library and it’s collection.

Climatological data :

As the COBECORE database will support tropical ecology research in the central Congo Basin we will limit transcription of climatological records to provinces of D.R. Congo with tropical climate regions (Köppen-Geiger classifications Af, Am and Aw respectively). These areas includes all provinces excluding Katanga which is considered sub-tropical (see Fig. 2). For the selected provinces we will focus on the longest and most complete records, both in terms of variables measured (e.g. temperature, precipitation, relative humidity, barometric pressure) and in terms of continuity of the time series. Where necessary missing data will be completed by digitizing complementary records in the INERA Yangambi archives.

Historical aerial photographs :

Assessments of canopy structure through historical aerial photographs will be limited to those areas currently located in and around present day permanent sampling plots, at Yangmbi and south of Kisangani, and data availability (covering a fraction of central the Congo Basin, see Fig. 2. dashed white rectangle). Furthermore, limited to no additional data field measurements on forest structure is available from other locations, reducing the potential of in depth analysis outside of the areas surrounding present permanent sampling plots. Scaling also requires expensive high resolution satellite imagery (~$500 / 100 km2 scene), discouraging such large scale research efforts.

Historical herbarium and wood specimens :

We will process 500 woody species with multiple replicates per species from both the Yangambi and Luki locations for leaf level traits (# determined through previous experience). Selection of the species is based on abundance in the BGM herbarium, occurrence in both Yangambi and Luki, accuracy of geo-reference data, species dominance in COBIMFO plots and presence of flowers/fruits, suggesting leaf exposure to sunlight. The list will also include the 68 species, or 95% of the current day basal area within recently established permanent sampling plots7. Complementary wood trait data and other plant functional trait data, e.g. plant height and seed size, will be retrieved from existing databases, the Flore D’Afrique Central and literature to augment the value of this dataset. The use of the two locations provides a space for time replacement, where co-occurring species at both locations (with a different climate) can provide insight in their eco-physiological plasticity under climate change.

2. Digitization (Objective 1):

Climatological data :

Meteorological data will be digitized with the help of the State Archive’s professional digitization equipment such as dedicated scanners. To alleviate any excessive demand on the State Archive’s time and man-power we will use project based digitization equipment in parallel. This digitization setup will include a high-resolution digital camera, cold lights and a reproduction stand. The parallel execution will ensure timely delivery of key data and added redundancy and cross institute access to a mobile setup. Images will be stored using a comprehensive file name structure, referencing the original State Archive INEAC index and the climate station’s location. The same data will be used to populate meta-data fields in the final database (objective 4). Data present in the State Archive adhere to a number of fixed formats (sizes, data structure; Fig. 1). We will correct the digital images for any colour imperfections, small deformations and increase readability using image processing software (Adobe Creative Cloud) for consistent and easy transcription. The BGM (partner 3) digitised all volumes of the “Flore d’Afrique Centrale”. This has fostered their expertise in text digitisation and mark-up. We will use the expertise of the BGM for our digitization efforts at the State Archive.

Digitization at the Yangambi INERA library will be carried out under RESYNDE management supported by Dr. Limba. Equipment for digitization at the Yangambi library will be provided by RESYNDE, ensuring timely delivery of data products. Only select data will be digitized, informed by previous work at the Sate Archive to avoid duplication of efforts.

Historical aerial photographs :

A significant selection of the historical aerial imagery of the Yangambi region exists in a digital form, yet without being geo-referenced. Missing digital data will be acquired through scanning of large prints on a flatbed scanner at the RMCA which holds this part of the INEAC archive.

3. Transcription (Objective 1):

Final valorization of the digital images of climatological data, i.e. converting image data to actual usable scientific data, will be performed by experts and through crowd-sourcing. Project researchers will transcribe 10% of the total dataset. Preference will be given in this effort to the most important meteorological records on file (longest and most complete time series) and/or very hard to read cases. This expert dataset will serve as a validation dataset for the crowd-sourcing campaign. The crowd-sourcing campaign will serve as a cost effective way of transcribing data while at the same time reaching out to the public to engage with current research within federal institutes. Both measures ensure a proper validation set to evaluate crowd-sourced transcriptions and a guaranteed quick return of usable data.

All data will be transcribed on the Zooniverse citizen science platform using their recently introduced project builder interface. The project will follow the structure of the successful crowd-sourcing project such as “Jungle Rhythms” and “Old Weather”. Citizen scientists will be asked to mark all locations of notations of text and numbers in an image (data sheet). Subsequently, the coordinates of these marks will be used to generate small subsets of the original digital image. These small subsets including a single word or number will then be shown 5 times to volunteers for further transcriptions. A majority rule will be used to determine the final accepted values. Both processes will be staggered in time where possible to ensure an optimal work-flow. Similarly, the number of herbarium specimens for processing of stomatal traits is limited due to the time intensive counting. Once more we will crowd-source this easy but time-consuming task, while retaining full control over 10% of the data for validation and rapid turnover of key species.

4. Geo-referencing and texture extraction (Objective 2):

Insight into the structure of the canopy is valuable to assess the representativeness of current day permanent sampling plots and the turnover time and processes governing the structure and function of tropical forests. Historical aerial imagery is currently only geo-located, providing an approximate location of all image data. To ensure proper alignment with current day satellite and aerial imagery these images need to be geo-referenced. For the larger Yangambi area we will geo-reference the historical images using 4 or more ground control points (recommended for aerial photography) determined by landscape features such as river banks, stream flow patterns and other important landmarks relative to current day high resolution satellite imagery. An example of one geo-referenced scene (Fig. 3) illustrates the potential of geo-referencing old aerial photographs in quantifying changes in forest structure and function. Figure 3 shows rapid conversion between past and present in terms of forest composition and structure. Here we see a fast transition from a mixed forest with a “rough” canopy (right dashed rectangle) to a mono-dominant Brachystegia sp. forest with a “smooth” canopy (left dashed rectangle). Geo-referenced images will support exploratory land-use land-cover change analysis and textural feature extraction using the fourier transform textural ordination (FOTO) method. The FOTO method uses a principal component analysis (PCA) on radially averaged 2D fourier spectra to characterize an image’s texture, linked to both indices of biodiversity and biomass^15,16. Both geo-referenced data and derived maps will be made publicly available to allow for more specialized and in depth analysis which are outside the scope of this call.

Fig. 3 - spatial and temporal changes in the canopy texture and species-composition as recorded by a diachronic comparison between (left) a recent (GeoEye-1, 11/11/2011 8:55) high resolution panchromatic 0.5 m remote sensing image and (right) a 1958 georeferenced aerial photograph with ~2m spatial resolution (Wild RC5a, Feb. 1958 10:30). Highlighted by a dashed rectangle is the fast transition between a mixed forest type in 1958 (right) and it's current mono-dominant Brachystegia sp. state (2012, left)

5. Leaf level traits (Objective 3):

Leaf traits of stomatal conductance, specific leaf area (the ratio of leaf area to dry mass) and leaf surface area, are important parameters in ecosystem models and indicators of adaptation to climate change in their own right^11,12. Stomatal traits (density and gwmax) of the selected herbarium specimens will be collected using a water based varnish method. Here a tin coat of varnish is applied to the leaf surface, dried, removed using clear tape and mounted on a microscope slide. At 3 locations (replicates) on the imprint images will be made at magnifications of 100x or 400x. High magnification limits the depth of field of the focal plane. Therefore, on each location, multiple images throughout the focal plane will be focus stacked to create a consistently sharp image across the field of view. Finally, stomatal parameters will be calculated for the focus stacked images. These parameters include, stomatal density (SD, stomates per m²) and gmax using the diffusion equation by Franks & Beerling (2009)¹²:

gwmax = ( dw * SD * pamax / y ) / (pd + π/2 (pamax / π )1/2),

where dw is the diffusivity of water vapour at 25°C and v the molar volume of air (constants), pamax is maximum stomatal pore area (m²) calculated as an ellipse using stomatal pore length (m) as the long axis and l/2 as the short axis; pd is stomatal pore depth (m) considered to be equivalent to the width of an inflated, fully turgid guard cell. The specific leaf area and leaf surface area parameters will be derived by scanning and weighing the leaves used in the analysis of stomatal traits. Similar analysis were executed by Dr. Stoffelen and Dr. Vandelook (partners 6 & 7) within the context of the Biospheretraits project. As such COBECORE will expand upon previous research efforts (see state-of-the-art).

6. Database management and dissemination (Objective 4):

We will use a “Not only” SQL (noSQL) database to store the variety of data formats generated. We prefer a noSQL solution over a relational database management system (RDMS) due to the complexity and varied content of the database as well as the future scalability of the database. The data will be hosted on a server at Ghent University where technical assistance is provided through the department ICT. The database, including data and meta-data, will be accessible through an application programming interface (API) to facilitate easy queries as well as flexible construction of front-end applications. A front-end for the database will be generated using the Shinyweb application framework with geographic information displayed using the Leaflet framework, while climate time series will be plotted using the plotly and D3.js libraries. An R toolbox will be written to easily query the data from the server using API commands as a stand-alone application. Furthermore, we will publish the whole database in a data journal (e.g. Scientific Data) to ensure the availability of the database, as a whole, long after the end of the project. In addition, relevant parts of the database will be submitted to specialized repositories to further ensure proper dissemination and accessibility of the data. These external data repositories would include, but not limited to, those of World Meteorological Organization (WMO), the TRY plant trait database¹⁷, the African tropical rainforest observation network (Afritron), biodiversity.be as well as Global Biodiversity Information Facility. Furthermore, we will reach out to the forestry industry and forestry research institutes for example the Center for International Forestry Research and Consultative Group for International Agricultural Research. In particular we will contact the World Agroforestry Center a partner organization of CGIAR, and the EU FLEGT initiative to help disseminate the dataset to their partners.

Data

The proposal deals with unexplored heritage as such all data sources will be available in either the State Archive, the RMCA or the Botanic Garden Meise (and by extension the Yangambi INERA library). No new specimens will be collected or measurements made, only existing historical data will be valorized through digitization of existing collections. Here, we merely convert analog archives into digitally and easily accessible data which will inform further research and/or policy, this in line with the scope of the call.

Costs associated with processing these existing datasets of unexplored heritage mainly involve the digitization and transcription of material stored in the archives or re-analysis of herbarium specimen (see methods). Costs include substantial manual labour indexing and digitizing data (archived paperwork or herbarium specimen), transcription, post-processing, data management, software and API development and public outreach efforts due to the crowd-sourcing component of the project. The timing and costs of these actions are reflected in the time-table and budget of this proposal.

Below we provide the list of data sources we will consult to create the three data streams as described in the Methods section.

Sections of the INEAC inventory (and page number) of particular interest include:

• p 421 – 559: Climatology and meteorology
• p. 63: Technical documents and protocols
• p. 229: Documents of the Forestry Division: this division managed exploitation of the forest surrounding Yangambi and was technically supported by the Forestry Station of Mayumbe in Luki and various forestry groups in other INEAC stations.

Additional information may also be found within the climatological data from specific research stations.

Sections of the RMCA archives pertaining to historical aerial photographs in the larger Yangambi region:

• Flight paths 1 – 20, pictures #58/04/78 – 58/09/06 (approximate coverage see Fig. 2)
• Exploratory research will be conducted on the:
  • The Bulletin d’INEAC
  • The Bulletin Agricole Du Congo Belge (published by the Ministry of Colonies)
  • Complete series of INEAC technical documents (protocols) matching wood traits will be extracted from existing databases to match processed historical herbarium specimens (see below).

Historical herbarium specimens:

500 species present in the BGM herbarium collection will be processed for leaf traits on stomatal conductance. Non-inventoried INEAC documents at the BGM may prove relevant within COBECORE. This will be discussed during working group meeting at the beginning of the project.

Supporting data:

Corresponding data, as outlined above, but missing in the State Archive will be sought in the Yangambi INERA library to complement the dataset. In particular efforts will be focussed on retrieving missing protocols and meteorological data. This work will be full-filled by RESYNDE, providing digitization expertise, in collaboration with INERA Yangambi.

Workplan and detailed description of the tasks

A detailed work plan describing individual tasks and work-packages is provided below. We also provide a flow-chart on the integration and interrelation of the work-packages and tasks (Fig. 4). Abbreviated Partners involved in the Work Package are listed below the sub-titles.

Fig 4. - COBECORE flow chart

Work Package 1

Identification and organization of complementary sources of ecological, climatic and land use change (meta)data

1.1 Preliminary identification of (meta-)data sources from the INEAC and related archives. (RMCA / UGent / SA / BGM)

This includes the collection of relevant existing printed and grey inventories; (limited) creation of new inventories to enable/improve access to archives. This work package will result in a comprehensive list of available data, and will be published as a brief report (D1 / see time table work plan).

1.2 Selective sampling of data from the INEAC archives at the State Archives, RMCA and in the federal collections of BGM following COBECORE. (RMCA / UGent / SA / BGM)

To limit the data streams to process we will selectively sample the best data available (providing some initial QA/QC). However, provisions will be made to ensure the future growth of the project outside the scope of COBECORE using flexible database structures and standard practices regarding inventory policies at the State Archives. The results and rational of the sampling will be noted for further reference and included in a brief report (D1 / see time table work plan).

1.3 Preliminary analysis of the state of Yangambi archives and their content. (INERA / RMCA / UGent)

In collaboration with RESYNDE an inititial analyses of the state of the Yangambi INEAC library will be executed. This to ensure proper project management, aligning expectations with the (state of the) available data.

Work Package 2

digitization, registration and validation of (meta)data

2.1 Registration of meta-data related to information extracted from the INEAC archives, such as corroborative research within the INEAC archives and in complementary archives and validation of meta-data types. (UGent / RMCA / BGM / SA)

Work Package 2.2 will include among others: a) geo-location (cross-checking photo locations), b) protocols (cross-checking methodology), c) taxonomic verification for current names, cross-referenced in botanical databases of all tree species , d) identification of possible sources of error in the data (when not otherwise specified), e) land use/landscape characteristics, including experimental layout and landmarks (on photographs and maps of forests). This work package will continue alongside a large part of the initial digitization process.

2.2 Digitization of legacy aerial photographs (RMCA / UGent)

Although some aerial photographs are digitized (see Fig. 2) remaining photos will be digitized on a flatbed scanner at the RMCA prior to processing. During this time we will also ensure proper file handling and post-processing prior to geo-referencing and further analysis.

2.3 Digitization from the INEAC archives at the State Archives, RMCA and in the federal collections of BGM (UGent / RMCA / SA)

Researchers from UGent and RMCA together with personnel from the State Archive / RMCA and BGM (depending on the location of the data) will digitize selected data streams from the INEAC archive. This work package involves the digitization of the climatological data, forest inventories, protocols as stored at the various federal institutes.

2.4 Digitization of herbarium specimens for stomatal density analysis + additional leaf trait parameters (BGM) The analysis focuses on stomatal density (SD), specific leaf area (SLA) and leaf area of the most dominant tree species in Luki and Yangambi. Measuring SD for 500 herbarium specimens each in Yangambi and Luki x 3 replicates per specimen would mean about 3000 preparations (including preparations, counts, entering data etc.), a truly unique data-set for comparative analyses. These measurements immediately provide leaf surface area data and scans for a comparative seed morphology (more details see WP3.3).

2.5 Digitization at the INERA Yangambi Archive, D.R. Congo. Gap filling of the missing parts of the climatological and ecologically relevant data streams. (INERA)

RESYNDE will be contracted to facilitate digitization of supporting data from the INEAC library in Yangambi. In collaboration with with Dr. Limba RESYNDE will digitize protocols pertaining to forest sampling and missing climatological data.

Work Package 3

Valorization of digitized data

3.1 Transcription of digitized meteorological records and ecologically relevant forest inventory data as well as stomatal density counts (through crowd-sourcing). (UGent / RMCA / BGM)

As described in the methods, 10% of the digitized data (both climatological as microscope slides of leaf stomata) will be processed by experts. Both dataset will be submitted to a crowdsourcing platform for further processing. Experts at UGent / RMCA and BGM will follow up on this process, and ensure the transfer of the final crowdsourced stomatal counts to the UGent partner for post-processing.

3.2 Geo-referencing of legacy aerial photography, as well as basic analysis of Land Use and Land Cover Change (LULCC) and textural metrics of canopy structure. (UGent)

We will use the free QGIS geographical information system in combination with contemporary high resolution imagery (Google Earth) to geo-reference the historical aerial photographs. Subsequently the geo-referenced data will be processed using the FOTO algorithm, extracting canopy texture features. In addition, we will manually outline forest features and areas of Land Use and Land Cover change (marking transitions between vegetation types).

In this work package an obvious link exists with both domain 2.2.4 (management and valorization of the federal audio, photographic and audiovisual heritage) and 2.2.5 (geo-location for the valorization of the historical, cultural and scientific heritage). We will actively seek out collaborations with funding allocated within these axis if applicable.

3.3 Extraction of supporting functional trait and geographical data from herbarium collections for relevant species, covered in forestry inventories. (BGM) For the species measured during the first year (WP 2.4), additional functional trait data will be extracted from the Flora of Central Africa and other works (reports, monographs etc.) not easily accessible to the scientific community but most of them kept in the Archives and Library of the Botanic Garden Meise. The families including the majority of dominant tree species have already been published, otherwise other works have to be consulted to fill gaps. Additional data of interest are functional trait data (plant height, dispersal type, seed size, leaf morphology etc.) and distribution data. All data have to be entered in the right format. During digitization data from herbarium labels will be retrieved and geo-referenced. These data will provide additional information on sampling time and geographical distribution.

3.4 Database design, valorization post-processing, data-warehousing, database population (UGent / RMCA)

This work package includes the technical aspects of the project regarding database design, management, and data ingestion. These tasks are spread out over the full duration of the project and included at appropriate times.

The database is layout is outlined early on in the project, although the choice for a noSQL approach leaves ample flexibility for later additions. Data processed by crowd-sourcing will need extensive post processing. After post-processing (selecting the final value for a particular crowd-sourced task) we will populate the final database.

Work Package 4

Enhancing the accessibility of the INEAC archives and valorizing their potential within a scientific and international context

4.1 Translation of the inventory of the INEAC archives from Dutch to French and English, to ensure easy access to the archived data stored at the State Archives and proper deposition of all digitized material in the State Archive database structure, for later online queries. (SA)

In this work package we will ensure the proper translation of the Dutch INEAC inventory at the State Archive. In addition, the State Archive will ingest all digital images (non-transcribed) of the INEAC archive. This supports ongoing digitization efforts and will make querying of the content of the INEAC archive easier as no access to the paper files would be required. The State Archive will incorporate this dat in their current database structure. This effort is therefore independent of the database development of COBECORE, but at the same time provides an additional backup of the digital data.

4.2 Creation of a geospatial web portal to access the data set based upon locality. (UGent)

In order to provide easy access to the data we will create a geospatial web portal which provides the possibility to browse and download the data on a site-by-site basis (or in bulk) and explore the data graphically. We refer to the Methods section for details on the implementation.

4.3 Extensive documentation of the project database and it’s functionality for future reference and extension of the work. (UGent / RMCA)

Although often overlooked documentation forms the backbone of many projects. It ensures reproducibility, transparency and easy transfer of data. As such we will provide extensive documentation in both English and French. This documentation will serve to provide as sources of information for public outreach as well as wealth of information for researchers after the completion of the project.

4.4 Final synthesis report and paper to be submitted to a data journal (e.g. Nature Scientific Data) to ensure proper citation (e.g. through a doi) of the open access data and longevity of the project output. (Ugent / RMCA / BGM / SA)

To increase the international visibility of the dataset we intend to publish the dataset in an open access data journal (e.g. Nature Scientific Data). Aside from increasing visibility it will also provide a safeguard against catastrophic data loss after the end of the project. A synthesis report based upon this publication will be submitted to BELSPO for internal review (see WP4.7).

4.5 Establish links to external data repositories such as those of the World Meteorological Organization (WMO), the TRY plant trait database and / or the African tropical rainforest observation network (Afritron). (Ugent / RMCA / BGM)

We further increase the visibility of the generated data by submitting parts of the database to specialized data repositories. Trait based data from the herbarium and wood samples will be submitted to the TRY plant trait database, the Afritron project, GBIF and biodiversity.be, climatological data will be committed to the WMO repository.

4.6 Internal review and international outreach (conferences) (Ugent / RMCA / BGM / SA)

This work package describes all tasks related to internal review, including workshops which provide project guidance, a closing symposium and international conferences through which project results are disseminated.

4.7 Public relations and Public outreach (Ugent / RMCA / BGM / SA)

Supporting the academic communication are public relations and public outreach. This component is key given the crowdsourcing component of the project. We will engage with the project through a dedicated website (specific to each crowdsourcing project) and overall project webpage. To sustain an active crowdsourcing community we will also provide feedback to the community through the Zooniverse Talk Forums. This effort requires low intensity but sustained efforts.

Valorization

Target groups and stakeholders

The data contained in the INEAC archives is of international significance, in particular for countries in the Congo Basin and D.R Congo in particluar. COBECORE will focus on the extraction of data from archives and data sources located in Belgium (gap filling with local data where necessary). As COBECORE progresses, it’s outputs will be made available to an international target audience and stakeholders.

The Network Workshops will promote the objectives, progress and achievements of COBECORE project both within the network and to international partners. An International Symposium at the end of the project will present the achievements of COBECORE to a wider audience of stakeholders, including information management specialists, software developers, researchers, students, policy makers, librarians and persons with an interest in the research of INEAC and the management and exchange of research information and knowledge.

Outputs of COBECORE:

• A COBECORE website will be developed and maintained by the UGent to advertise progress of the project
• The fully operational database will be accessible through the COBECORE website and it’s API
• COBECORE will be presented at (inter)national scientific meetings (mainly: the conference of the Association for Tropical Biology and Conservations (ATBC) and European Geosciences Union general assembly (EGU))
• Advertisement of the database will be done on affiliated organization web sites, mailing lists, flyers and posters will be printed for distribution at meetings etc;

In addition to the publication of research results in scientific journals, the following documents will be made available through the website:

• Translation of the Dutch inventory of the INEAC archives in French and English, with a link to the website of the National Archives
• Intermediary report “Working document: Location and state of eco-climatological data of the INEAC archives in Belgium”
• Intermediary report “Management and protocols used for the valorization of the INEAC archives” (R2 in the Work Plan)
• Final Report “Valorization of legacy data in multivariate databases: the COBECORE project” (R3 in the Work Plan) Proceedings of the International Symposium

Expected research results and their impact for science, society and decision making

Expected research results

The COBECORE project will result in a comprehensive database of three eco-climatological data. This database will include climatological data across the tropical climate regions (all D.R. Congo provinces, except Katanga, Fig. 2). In addition, the project will produce geo-referenced historical aerial photographs, covering the forests around Yangambi (among others the current day UNESCO man-and-biosphere reserve), and derived products such as texture metrics quantifying forest canopy structure, which is related to forest biomass and biodiversity, and land-use land-cover change maps. Furthermore, eco-physiological important wood and leaf traits (vessel diameter, stomatal density and gwmax) from 500 historical herbarium specimens.

The complete integrated dataset will be made available through an open access data journal publication and a dedicated website for interactive exploration of the data. As the three data streams have their own value, data will be submitted to dedicated field specific data repositories such as the plant trait database TRY, the world meteorological organization (WMO) and federal initiatives e.g. the in-house database of the State Archives and the cartesius.be project.

At present, scientific information is freely available and accessible in the INEAC archives. Yet, the cumbersome nature of retrieving and assimilating this information is unappealing for many, particularly foreign researchers, especially as there is no indication of the time and effort required. A better estimation of the time and effort required to retrieve, digitize, evaluate, corroborate and incorporate data from the INEAC archives to the online accessible database, will improve opportunities for more extensive validation of the INEAC archives, by making it a less hypothetical exercise. A more detailed inventory and estimates of the workload required for digitization and processing will be provided in intermediary reports “Working document: Location and state of eco-climatological data of the INEAC archives in Belgium (R1, see time table)”and “Management and protocols used for the valorization of the INEAC archives (R2, see time table)”. Finally, we will translate the INEAC archive inventory in French / English, to be deposited at the reading room of the State Archives and listed online for digital inquiries on the State Archive’s website. This will greatly enhance accessibility for Congolese partners and the international research community.

Expected impacts of the research and compliance of the research with the expected impacts

COBECORE is driven by national and international demand from the scientific community. As such, the project will have as significant impact on the scientific community. The project will expand current scattered local digitization efforts (mainly focused on the rift valley) into a basin wide analysis of historical eco-climatolgoical data, covering both a large spatial and temporal extent (~1920 - 1960).

As such, the project will provide the first comprehensive and consistently curated historical dataset of ecosystem and climatological parameters for the central Congo Basin, filling important data gaps and will support research into climate re-analysis, climate forecasting and forest ecosystem modelling. Furthermore, the translation of the inventory (currently a Dutch inventory for an essentially francophone archive) will also greatly increase accessibility for foreign (notably Central African) researchers. Additional translation to English and availability as an online pdf, will enable it to reach a much wider audience, stimulating future validation opportunities immensely.

Results of the COBECORE project will directly influence climate and forestry policy through a better understanding of forest structure and the spatial and temporal changes within the forest. The latter have direct consequences within the context of the United Nations reducing emissions from deforestation and forest degradation (UN-REDD+) program or the EU FLEGT action plan. Knowledge gleaned from the COBECORE dataset will inform carbon management but also support sustainable forestry practices under a changing climate and guarantee durable local and foreign forest product trade.

Although COBECORE is focussed heavily on providing much needed data streams for basic research the project does have a considerable societal impact. For example, an active part of the project will involve active awareness raising in the form of crowd-sourcing. A previous project, Jungle Rhythms, attracted over 6000 volunteers of with more than 1/3 officially registered in the Zooniverse community. Although there is a bias towards English spoken countries, participants are almost equally divided between Northern America (47%) and the EU (31%), illustrating the wide reach and societal impact of such a project. Currently, the project is ~70% complete 10 months after it’s start.

In addition, the project also supports capacity building in D.R. Congo due to local digitization efforts. The outcome of our research through policy such as sustainable forestry practices will also have a direct impact on the local population. Indeed, the information mobilized by the COBECORE project relates to issues that reach much wider than the scope of COBECORE. The scientific value of the data resides in the extent of the data collected in a region where the availability of reliable baseline information is practically absent. In this way, the INEAC archives are complementary to many current research topics and policy issues. As an extensive source of reliable information, data stored in the INEAC archives can help to elucidate or reinforce contemporary agronomic, ecological and climatic trends. In this way, the INEAC archives may guide or support policy efforts and priority setting in the realm of development cooperation, biodiversity conservation, poverty alleviation, agriculture, climate change mitigation, natural resource management and food-security in Africa, and globally.

TABLE

follow-up committee (provisional)

The follow-up committee will be involved in 2 Network Workshops and a final symposium. Follow-up committee members will serve as advisees throughout the project. Contact will be via at least one of the following: email, skype, or telephone (at their preference). They will also serve as representatives of larger networks of resource persons (e.g. CGIAR, FAO). Participation in intermediary and ad-hoc meetings will be on a voluntary basis.

Chris Jones is an atmospheric scientist at the Met Office where is main focus his the analysing of feedbacks between climate and the carbon cycle and looking at the long term commitments of ecosystem changes to climate change. His research was instrumental in enabling these first simulations which identified the potential large, positive feedback between climate and the carbon cycle. He has subsequently been involved in analysing the mechanisms behind this feedback by comparing models with both observations and other models in the C4MIP and CMIP5 model inter-comparison project.

Jan Verbesselt is associate professor in remote sensing at Wageningen University, Laboratory of Geo-information Science and Remote Sensing. He focusses at measuring and understanding ecosystem dynamics by developing novel spatio-temporal methods to detect, monitor and forecast changes using remotely sensed data from in-situ, terrestrial- and airborne LiDAR, and satellite sensors. The application of remotely sensed images for ecological modelling, and collaborative earth science for assessing vegetation, climate, and human impacts take a central place.

Stefan Hauser is a Systems Agronomist with 25 years of research experience in West and Central Africa. His background is in soil science and soil biology in the humid forest zones of Nigeria and Cameroon. Now based in International Institute for Tropical Agriculture, IITA-Nigeria, he recently returned from D.R. Congo after formally establishing IITA-D.R. Congo and conducting an academic training program for the Congolese national agricultural research system. He currently works on cassava and yam agronomy and plant nutrition in Nigeria, Ghana, Cameroon, and D.R. Congo. He is also the Focal Point at IITA for the CRP-Humid Tropics program.

Eric Tollens is professor emeritus of Agricultural Economics at the KULeuven. He worked in the Congo from 1970-1977, where he was a lecturer in the Institut Facultaire d’Agronomie at Yangambi (’73-’77). He is a member of the Governing Boards of the CTA in Wageningen and the World Agroforestry Centre in Nairobi. He has a long experience in the D. R. Congo. Presently, he is charged by IITA and International Food Policy Research Institute (IFPRI) to do a study on the restructuring of agricultural research in the D.R. Congo.

Inge Jonckheere is currently a Forestry Officer for the Food and Agriculture Organization (FAO) of the United Nations (UN). She provides critical guidance and advice in Monitoring, Reporting and Verification (MRV) within the context of the “reducing emissions from deforestation and forest degradation in developing countries” (REDD+) under the UN framework convention on climate change (UNFCCC). Furthermore, she is responsible for the remote sensing web portals for countries in the REDD+ context and coordinates capacity building for developing countries on tropical forest monitoring.

Sustainability, gender dimension, ethics

All partners of the COBECORE network will abide by the principles outlined in the Code of Ethics for Scientific Research in Belgium. This will be further encouraged by providing all partners (and newly recruited staff members) a copy of this document.

The COBECORE network represents 2 Federal Institutions (RMCA, State Archive), 1 Regional Institution (BGM) and 1 University. Gender is represented as follows: 1/7 partner representatives are female (a RMCA partner – see partner experience: 4.1), the follow-up committee includes 1 woman. Recruitment of staff by COBECORE will follow equal opportunity guidelines. When two candidates are equally ranked, preference will be given to the female candidate.

There is an interesting balance between experienced senior scientists and early-stage researchers, fostering guidance and transfer of knowledge. This transfer of knowledge will support career development for early-stage researchers, helping them grow within the Belgian research community.

All documents under review during valorization are of non-personal nature, as it pertains protocols and scientific measurements, and are outside of copyright law due to their age. We will adhere to an open access policy making all data available under a creative commons license ensuring transparency and unrestricted access to publicly funded research of international significance, with implications within climate and ecological research, mitigation efforts and sustainable forest management.

Sustainability of the COBECORE project will also be supported through by its linkage with international data repositories, such as (but not limited to) the World Meteorological Organization (WMO), the TRY plant trait database and the African tropical rainforest observation network (Afritron). Furthermore, we will publish the database in an open access format in a data journal (e.g. Scientific Data). This will enable free and unrestricted access to publicly funded relevant data of international significance, particularly for African researchers. Furthermore, It will also ensure proper citation of the data, it’s processing and funding agencies.

Finally, we will ensure complete and transparent accessibility to all data gathered. In addition to the above mentioned measures data will be burned to DVD(s), fully documented in French and mailed by post to INERA Yangambi to guarantee local dissemination were stable internet connections are limited.

Network

Partners’ expertise

Ghent University (Coordinator)

Dr. Hans Verbeeck holds a PhD from the Laboratory of Plant Ecology at Ghent University. He recently started the CAVElab on an ERC starting grant TREECLIMBERS (grant number 637543), modelling the response of lianas as key drivers of tropical forest responses to climate change. More generally, his research field is the ecology, carbon and water cycle of terrestrial ecosystems, mainly tropical and temperate forests. Ecosystem modelling is the core tool of his research, and the research questions that emerge from the modelling work are driving the design of field work activities. The CAVElab focusses on the role of lianas in the climate response of tropical forests as well as biogeochemical cycles of tropical forests in the context of climate and land-use change. Additionally attention is given to the interaction between functional diversity and biogeochemical cycles in (tropical) forests and methodological aspects of vegetation modelling. For example, previous research under his supervision of PhD Jackie Epila (supported by the Flemish University Development Cooperation) focussed on the assessment of the drought vulnerability of key tree species in Ugandan tropical forest.

Select publciations (h-index: 23):

1. Kearsley E, de Haulleville T, Hufkens K, Kidimbu A, Toirambe B, Baert G, Hyugens D, Kebede Y, Defourny P, Bogaert J, Beeckman H, Steppe K, Boeckx P, Verbeeck H. (2013) Conventional tree height-diameter relationships significantly overestimate aboveground carbon stocks in the Central Congo basin. Nature Communications, 4: 2269.

2. Verbeeck H, Peylin P, Bacour C, Bonal D, Steppe K, Ciais P (2011) fluxes in Amazon forests: Fusion of eddy covariance data and the ORCHIDEE model. Journal of Geophysical Research, 116, 1–19.

3. Verbeeck H, Betehndoh E, Maes W et al. (2014) Functional leaf trait diversity of 10 tree species in Congolese secondary tropical forest. Journal of Tropical Forest Science, 26.

4. Verbeeck H, E Kearsley (2015) The importance of including lianas in global vegetation models. Proceedings of the National Academy of Sciences, 201521343

5. Doetterl S., Kearsley E., Bauters M., Hufkens K., Lisingo J., Baert G., Verbeeck H., Boeckx P. (2015) Aboveground vs. Belowground Carbon Stocks in African Tropical Lowland Rainforest: Drivers and Implications. Plos One, 10: e0143209.

Dr. Koen Hufkens holds a PhD in biology from the University of Antwerp. During his PhD he bridged the fields of physics and biology. His interdisciplinary work has sharpened his communication and problem solving skills. In subsequent years this passion for interdisciplinary research has led him to participate in a variety of posts that he has navigated well, acquiring new knowledge in the analysis of large datasets whilst maintaining an integrative overview.

After a first stay in the US the candidate’s focus shifted from landscape ecology towards the study of phenology using satellite and near-surface remote sensing, and eco-physiological measurements. During these years he became involved in the PhenoCam project led by Andrew D. Richardson. The candidate then returned briefly to Belgium to pursue research within the field of tropical forest ecology. Thereafter, an opportunity to return to the Richardson Lab at Harvard University arose, to continue work on the PhenoCam project. Here he was involved in the synthesis and visualization of ~1000 site years of the continuously growing PhenoCam archive, representing approximately 8 TB of images (e.g. http://explore.phenocam.io). Through his research he has experience with big dataset management and increasing data accessibility through R toolboxes and interfaces (e.g. Amerifluxr R toolbox). Select publications show the reach of his previous work. More software tools can be found on his github repository.

Select publications (h-index: 14):

1. Hufkens, K., Keenan, T. F., Flanagan, L. B., Scott, R. L., Bernacchi, C. J., Joo, E., … Richardson, A. D. (2016). Productivity of North American grasslands is increased under future climate scenarios despite rising aridity. Nature Climate Change, 6: 710-714.

2. Doetterl S., Kearsley E., Bauters M., Hufkens K., et al. (2015) Aboveground vs. Belowground Carbon Stocks in African Tropical Lowland Rainforest: Drivers and Implications. Plos One 10: e0143209.

3. Kearsley E., de Haulleville T., Hufkens K., Kidimbu A., Toirambe B., Huygens D., Kebede Y., Defourny P., Bogaert J., Beeckman H., Baert G., Steppe K., Boeckx P., Verbeeck H. (2013). Conventional tree height–diameter relationships significantly overestimate aboveground carbon stocks in the Congo Basin. Nature Communications 4: 2269

4. Hufkens K., Keenan T., Richardson A. D., Sonnentag O., Melaas E., Bailey A., O’keefe J., Friedl M. (2012). Ecological Impacts of a Widespread Frost Event Following Early Spring Leaf-Out. Global Change Biology: 18: 2365 – 2377.

5. Hufkens K., Friedl M., Sonnentag O., Braswell B. H., Milliman T., Richardson A. D. (2012). Linking near-surface and satellite remote sensing measurements of deciduous broadleaf forest phenology. Remote Sensing of Environment: 117: 307-321.

Additional support will be provided by Dr. Jan Van den Bulcke. Jan Van den Bulcke obtained a MSc degree in Bioscience Engineering (Land and Forest management) and a MSc degree in Environmental Sanitation at the Ghent University in 2000 and 2001 respectively. He obtained his PhD in 2006 and is since 2007 postdoc at the Laboratory of Wood Technology (Woodlab-UGent). His main research interests are non-destructive testing, structural and chemical mapping / visualization of lignocellulosic materials, understanding climate – xylem growth using X-ray CT scanning at the inter- and intra-ring level and mathematical modelling of biological processes in collaboration with other groups, through joint PhD projects. As a postdoc, he is also involved in several research projects dealing with imaging and analysis of tree growth, with focus on technological expertise on scanning and imaging. The group at Woodlab-UGent has developed a fast and high-throughput X-ray based technique for analysis of increment cores, with special focus on the tropics. Through collaboration, a.o. in COST actions as (substitute) management committee member and as partner in international projects, he is acquainted with partnership management both at national and international level.

Select publications (h-index: 15):

1. Couralet, C., Van den Bulcke, J., Ngoma, L.M., Van Acker, J. & Beeckman, H. (2013). ‘Phenology in functional groups of central African rainforest trees’. Journal of Tropical Forest Science 25: 361-374.

2. De Ridder, M., Van den Bulcke, J., Van Acker, J. & Beeckman, H. (2013). ‘Tree-ring analysis of an African long-lived pioneer species as a tool for sustainable forest management’. Forest Ecology and Management 304: 417-426.

3. Van den Bulcke, J., Wernersson, E. L., Dierick, M., Van Loo, D., Masschaele, B., Brabant, L., Boone, M.N., Van Hoorebeke, L., Haneca, K., Brun, A. & Hendriks, C. L. L. (2014). ‘3D tree-ring analysis using helical X-ray tomography’. Dendrochronologia 32: 39-46.

4. De Mil, T., Vannoppen, A., Beeckman, H., Van Acker, J., & Van den Bulcke, J. (2016). A field-to-desktop toolchain for X-ray CT densitometry enables tree ring analysis. Annals of Botany mcw063.

5. De Mil, T., Ilondea, B. A., Maginet, S., Duvillier, J., Van Acker, J., Beeckman, H., & Van den Bulcke, J. (2016). ‘Cambial activity in the understory of the Mayombe forest, D.R. Congo. Trees, 1-13.

Royal Museum for Central Africa

Dr. Hans Beeckman graduated as forestry engineer and made his PhD on the numerical analysis of tree-ring data. He is senior scientist and curator of the Xylarium at the Royal Museum for Central Africa since 1995. In January 2014 he has been appointed as head of the new founded Wood Biology Service. The Service studies the material wood as part from living beings. This includes the analysis of cells and tissues (wood anatomy) and the growth of trees as a result of cambial activity. The actual focus is on the rainforests of Central Africa where structure and functions of tropical trees are being studied through wood analysis. Earlier focuses have been on woodlands, mangroves and montane vegetation. His current research inlcudes studies of the dynamics of cambial activity, the climate information in growth-ring patterns and the variability of wood traits. Furthermore, he addresses scientific questions on forest history (species diversity) and archaeobotany are addressed through analysis of charcoal containing layers in the soil (pedoanthracology).

Dr. Beeckman has daily contacts with the international communities of wood anatomists en dendrochronologists. He is also involved in the scientific authorities of CITES and other regulations aiming at conservation of tropical forests and fostering legal and sustainable timber trade. His research network includes Belgian and Congolese universities with which he has common PhD and postdoc projects. Through courses and identification services he has extensive contacts with the timber industries. His contact with the timber industry will assist in dissemination of the data within the context of sustainable forest management.

Select publications (h-index: 23):

1. Couralet, C., Van den Bulcke, J., Ngoma, L.M., Van Acker, J. & Beeckman, H. (2013). ‘Phenology in functional groups of central African rainforest trees’. Journal of Tropical Forest Science 25: 361-374.

2. Kearsley, E., de Haulleville, T., Hufkens, K., Kidumbu, A., Toirambe, B., Baert, G., Huygens, D., Kebede, Y., Defourny, P., Bogaert, J., Beeckman, H., Steppe, K., Boeckx, P. & Verbeeck, H. 2013. ‘Conventional tree height-diameter relationships significantly overestimate aboveground carbon stocks in the Central Congo Basin’. Nature Communications 4: 2269.

3. De Ridder, M., Van den Bulcke, J., Van Acker, J. & Beeckman, H. (2013). ‘Tree-ring analysis of an African long-lived pioneer species as a tool for sustainable forest management’. Forest Ecology and Management 304: 417-426. DOI: 10.1016/j.fotrco.2013.05.007.

4. Hubau, W., Van den Bulcke, J., Kitin, P., Mees, F., Van Acker, J. & Beeckman, H. (2012). ‘Charcoal identification in species-rich biomes: a protocol for Central Africa optimised for the Mayumbe forest’. Review of Palaeobotany and Palynology 171: 164-178.

5. Beeckman, H. 2016. ‘Wood anatomy and trait-based ecology’. IAWA Journal 37: 127-151.

Dr. Kim Jacobsen holds a PhD in Bio-engineering from the KULeuven and has a background in tropical agronomy, nematology, developmental biology and marine biology. Kim has served as an EU-project administrator for biodiversity information projects (incl. EDIT, SYNTHESYS and EU BON). In this function, among others, she organized training events, workshops, summer schools.

Furthermore, Kim holds key knowledge of the structure of the collections at RMCA related to the INEAC archives. She also as an extensive network with (inter-)national partners with a vested interest in valorization of legacy data pertaining to Central Africa (eg. CGIAR, FAO). These contacts will help to ensure proper dissemination of the data after the project’s completion to maximize the expected impact (see 3.2) of the project.

Select publications (h-index: 6):

1. Blomme, G., Jacobsen, K. et al. (2014). Fine-tuning banana Xanthomonas wilt control options over the past decade in East and Central Africa. European Journal of Plant Pathology 139:1-17.

2. Norgrove, L., Tueche, R., Jacobsen, K et al. (2012). Tackling black leaf streak disease and soil fertility constraints to enable the expansion of plantain production to grassland in the humid tropics. International Journal of Pest Management 58:175-181.

3. Jacobsen, K., Maes, L. et al. (2009). Host status of twelve commonly cultivated crops in the Cameroon Highlands for the nematode Pratylenchus goodeyi. International Journal of Pest Management 55: 293- 298.

4. Hauser, S., Mekoa, C. & Jacobsen, K. (2008). Bunch yield response of two cultivars of plantain (Musa spp. AAB, Subgroups French and False Horn) to hot-water treatment and fertilizer application planted after forest and bush/grass fallow. Archives of Agronomy and Soil Science 54:541-556.

5. Jacobsen, K., Fogain, R., Mouassom, H. & De Waele, D. (2004). Musa-based cropping systems of the Cameroon highlands: a case study of the West and North West province of Cameroon with emphasis on nematodes. Fruits 59:311-318

Botanic Garden Meise

Dr. Piet Stoffelen started his professional career with a study on the taxonomy and systematics of tropical African genera of the Rubiaceae. From 1999 till 2003, his main task at the National Botanic Garden was assisting the curator in everyday collection management. From 2004 onwards the Garden became a partner in several digitalisation projects (e.g. African Plants Initiative, European Network for Biodiversity Information and Belgian Biodiversity Platform). These projects required the introduction of international standards for digitalisation, a new database and the training of new staff. Between 2004 and 2006, to which Dr. Stoffelen made important contributions.

From 2007 onwards the rehabilitation of Central African Botanic Gardens and Herbaria became an important duty of the National Botanic Garden of Belgium. Dr. Stoffelen collaborated on these rehabilitation projects, funded by the European Union, the French Ministry of Foreign Affairs, the Belgian Cooperation for Development and the Belgian Ministry of Sciences. He was the Gardens representative in the inter-institutional committee of the ‘Congo-2010 project’. In 2011 Dr. Stoffelen became the acting curator of the vascular plants herbarium, with about 3 million herbarium specimens; The everyday management of the collections (loans, returned loans, digitalisation, scanning, etc.), the follow up of current projects, the implementation of new international standards and procedures and the efforts to link the herbarium to new emerging opportunities are a full time occupation. He was an official partner in several Belspo financed projects (e.g. XYLAREDD, BIOSPHERETRAIT) and collaborated in other related projects (e.g. COBIMFO, HEREBAXYLAREDD) and has expertise in digitalisation and collection managment. In 2016 he became head of the Collections Department at the BGM. During COBECORE Dr. Stoffelen, together with Dr. Vandelook (see below) will provide guidance and oversight of the processing of herbarium specimens at the Botanic Garden Meise.

Selected publications (h-index: 10):

1. Godefroid S., Van de Vyver A,Stoffelen P.,Robbrecht E. & Vanderborght Th. (2011). Testing the viability of seeds from old herbarium specimens for conservation purposes. Taxon 60: 565-569.

2. Massó i Alemán, S., Bourgeois, C., Appeltans, W., Vanhoorne, B.;De Hauwere, N.,Stoffelen, P., Heughebaert, A., Dahdouh-Guebas, F. (2010). The ‘Mangrove Reference Database and Herbarium’. Plant Ecology & Evolution 143: 225-232

3. Stoffelen P., Noirot M., Couturon E., Bontems E., De Block P. & Anthony F. (2009). Coffea anthonyi, a new self-compatible Central African coffee species, closely related to an ancestor of Coffea arabica. Taxon 58: 133-140.

4. Davis, A.P., Govaerts, R., Bridson, D.M. & Stoffelen P. (2006). An annotated taxonomic conspectus of the genus Coffea (Rubiaceae). Botanical Journal of the Linnean Society 152: 465-512.

5. Stoffelen, P., Vander Velde, A., Mergen, P. & F. Wautelet. (2005). Digital Imaging of Biological Type Specimens - Case Studies: Plants. In: Häuser C.L., Steiner A., Holstein J. & Scoble M.J. (eds.). Digital Imaging of Biological Type Specimens, A Manual of Best Practice, Results from a study of the European Network for Biodiversity Information. p. 291-292.

Dr. Filip Vandelook is a plant ecologist working in the science department of the Botanic Garden Meise since 2013. Dr. Vandelook is Honorary Research Associate of the Botanic Gardens Kew and obtained a Belspo Back to Belgium Grant in 2013. His research interests are situated in the fields of ecology and evolution of plant reproductive and plant functional traits. Dr. Vandelook has extensive experience in population biology, phylogenetic comparative methods and methods to study phylogenetic community structuring. These methods are applied to study the potential impact of climate change on individual plant species and community structure.

Dr. Vandelook has extensive international experience with research stays at the Philipps Universität Marburg, CIDE Valencia and RBG Kew, as well as with field work in Bolivia and the RDC. He developed a robust scientific network with both national and international partners with contacts in Europe (e.g. Universität Mainz, Philipps Universität Marburg, Universität Regensburg, Royal Botanic Gardens Kew, CNRS Mouli) and the rest of the world (e.g. UNIKIS, Lwiro, UMSS Cochabamba).

Selected publications (h-index: 10):

1. Janssens, S.B., Vandelook, F. et al. (2016). Evolutionary dynamics and biogeography of Musaceae reveal a correlation between the diversification of the banana family and the geological and climatic history of Southeast Asia. New Phytologist, In Press

2. Vandelook, F., Verdú, M. & Honnay, O. (2012). The role of seed traits in determining the phylogenetic structure of temperate plant communities. Annals of Botany 110: 629-636.

3. Vandelook, F., Janssens, S.B., Probert, R.J. (2012). Relative embryo length as an adaptation to habitat and life cycle in Apiaceae. New Phytologist, 195: 479-487.

4. De Frenne, P., De Schrijver, A., Graae, B.J., Gruwez, R., Tack, W., Vandelook, F. et al. (2010). The use of open-top chambers in forests for evaluating warming effects on herbaceous understorey plants; Ecological Research 25: 163-171.

5. Vandelook, F., Van de Moer, D., Van Assche, J.A. (2008). Environmental signals for seed germination reflect habitat adaptations in four temperate Caryophyllaceae. Functional Ecology 22: 470-478

State Archives of Belgium

As head of the Service “Contemporary Archives” at the National Archives of Belgium (NAB1) since 2011, Dr. Michaël Amara is well acquainted with the historical relevance and the conservation of archives of all types. One of his main tasks concerns the conservation and valorization of archives. More than 24 million images are available on the websites of the State Archives. He was involved in many projects of digitalization. He’s a specialist in First World War Studies and obtained his PhD from the University of Brussels [ULB] (Dep. of Contemporary History) (2007) with a thesis on Belgian refugees in France, Netherlands and UK during the First World War.

Selected publications:

1. H. Vanden Bosch, M. Amara and V. D’Hooghe (P.-A. Tallier dir.), Guide des sources de la Première Guerre Mondiale en Belgique, Brussels, NAB (2010), 1057 p.
2. M. Amara, 1914-1918: Des Belges à l’épreuve de l’exil. Les réfugiés de la Première Guerre mondiale (France, Grande-Bretagne, Pays-Bas), Bruxelles, Editions de l’Université de Bruxelles (2008).
3. M. Amara, “L’exfiltration des ouvriers belges, au confluent de la guerre clandestine et de la mobilisation» in R. VANDENBUSSCHE, Le résistance en France et en Belgique occupées (1914-1918), Villeneuve d’Ascq (2012), pp. 52-68.
4. M. Amara, “Les grands défis de la propagande belge durant la Première Guerre mondiale” in B. Rochet et A. TIXHON, La petite Belgique dans la Grande Guerre. Une icône, des images, Namur (2012), pp. 21-33.
5. M. Amara, “Belgian refugees 1914-1918” in P. Gatrell, Refugees in Total Wars, Manchester University Press (to be published in 2017).

Network’s composition

As an interdisciplinary project, COBECORE will embrace the expertise of the past for the benefit of future generations, using state-of-the-art technology, software and public outreach concepts. The know-how and experience required to bring this project to fruition is reflected in the choice of partners.

Ghent University (Ugent) : The UGent team unites colleagues from the CAVELab and the Wood Lab. The CAVElab has a strong research focus on the ecology, carbon and water cycle of terrestrial ecosystems, mainly tropical and temperate forests. For example the recent ERC starting grant funded TREECLIMBERS project focusses on modelling the response of lianas as key drivers of tropical forest responses to climate change. Previous projects such as FORMONCO (2015-2016) supported the development of a platform for long-term integrated forest ecosystem monitoring in D.R. Congo while the Congo basin integrated monitoring for forest carbon mitigation and biodiversity (COBIMFO) project, produced a first assessment of carbon stocks and biodiversity within the Congo Basin. The Wood Lab (represented by Dr. Van den Bulcke) has over 40 years of experience in scientific and technological research on wood and maintains close contacts with all segments of the wood-processing industry and research groups involved in the forestry-wood chain. The UGent together with the RMCA will supervise collection, sampling and evaluation of data as a prerequisite for the development of the online accessible database.

Royal Museum for Central Africa (RMCA) : As a centre of research and knowledge dissemination, the RMCA is an important interdisciplinary node for research on past and present societies and natural environments of Africa, and in particular Central Africa. The Museum has approximately 85 scientists. One of the particular roles of the museum is to make information sources and systems accessible to Africa, specifically in the context of development cooperation and often, in digital format. Examples include, FishBase (an international online fish encyclopedia); Central African Biodiversity Information Network (CABIN), which aims to develop a network of databases on biodiversity, to be installed and supported in collaboration with institutions in the D.R. Congo, Rwanda and Burundi; and the distribution of RMCA publications. Currently, the complete series of the Bulletin d’INEAC (published research results of the INEAC, in paper format and exclusively in French), the Bulletin Agricole Du Congo Belge (published by the Ministry of Colonies), an almost complete series of INEAC technical documents and a complete series of the microfilms are still stored at the RMCA.

State Archives of Belgium (SA) : The most extensive collection of INEAC archives are physically located at the State Archives in Brussels, where they are accessible in the reading room. Documents can be requested by citing a document number from the inventory. Considering the cumbersome nature of document retrieval and the need to frequently review data before entering it into a database, documents sampled from the INEAC archives will be scanned or photographed. This will allow (1) a maximum efficiency and reliability when gathering data for inclusion into the database, and (2) a digital record of the data will be made accessible through the online database. COBECORE will translate the inventory into French (and English), for optimum outreach and accessibility. The project will enlist an ICT technician to assist with document scanning and ingestion into the State Archive database.

Botanic Garden Meise (BGM) : The Botanic Garden Meise has considerable expertise in the digitization of biodiversity material and its presentation and distribution on the Internet. They are leaders in a work package for the pro-iBiosphere (FP7) project investigating the digitisation, connectivity, acquisition and dissemination of biodiversity information (http://www.pro-ibiosphere.eu/). Furthermore, the garden has developed its own Virtual Herbarium system, where high quality scans are available online from the herbarium catalogue. The garden has also digitised all volumes of the Flore d’Afrique Centrale and made them available on the internet. This has fostered their expertise in text digitisation and mark-up. By creating a database of paper literature it has opened up the possibility of new research, which would have been impossible from paper alone. The INEAC/SERDAT botanical library is in storage at the BGM and catalogued in their library. Additionally, there exists a considerable amount of non-inventoried INEAC documents at the BGM that may prove relevant within COBECORE. Informal examination of the documents suggests the contents relate mostly to the last two decades of INEAC (1942-1960).

Gender

Gender is represented as follows: 1/9 partner representatives are female (the RMCA-representative), the follow-up committee includes 1 woman. Recruitment of staff by COBECORE will follow equal opportunity guidelines. When two candidates are equally ranked, preference will be given to the female candidate.

International dimension of the project

COBECORE aims to define the proposed research in a context of global changes and sustainable development. All partners have a strong research focus on topics located in the D.R. Congo where they execute studies using field data and/or collection material. As such COBECORE extends on current expertise which is reflected in the international initiatives and opportunities we will seek out within the context of the proposal.

COBECORE will select INEAC-related data sets which have been digitized (or will be, cfr proposals submitted). Interoperability with these initiatives will be encouraged during the course of COBECORE:

• For example, select climate data from INEAC and INERA, from 1950 onwards (E. Tollens, personal communication; E. Tollens is in the follow-up committee)
• SOTER (SOil and TERrain) database by ISRIC-FAO-UGent: In order to make soil data more accessible primarily to Congolese scientists, soil maps were digitized (ArcView) at the Universities of Lubumbashi and Kinshasa and stored in a digital database, in collaboration with the Ghent University and the University College Ghent (VLIR-funding). Ninety published (INEAC) and non-published soil maps were stored in the database. The soil units were digitized as a polygon theme, representing the soil-mapping units on the original soil maps. Most digitized soil maps were geo-referenced (the UGent team responsible for this digitization effort will be contacted to share previous experiences).
• UNESCO – BIOSPHERETRAITS (Proposal submitted by RMCA/BGM, currently ongoing)
• BELSPO BRIAN HERBAXYLAREDD (currently ongoing)

The project will also reach out to relevant data access and interoperability initiatives:

• eI4Africa (http://ei4africa.eu): This project aims to boost the research, technological, development and innovation potential of African e-Infrastructures and to support policy dialogues and Euro-African cooperation in the framework of the joint Africa-EU Strategic Partnership
• The Dataverse Network (http://thedata.org) is an open source application to publish, share, reference, extract and analyze research data. It facilitates making data available to others, and allows you to replicate others work. Researchers, data authors, publishers, data distributors, and affiliated institutions all receive appropriate credit. A Dataverse Network hosts multiple dataverses. Each dataverse contains studies or collections of studies, and each study contains cataloging information that describes the data, plus the actual data and complementary files. Dataverse and the Dataverse Network were created by the Institute for Quantitative Social Science at Harvard University (USA).
• Research Data Alliance (http://rd-alliance.org): The alliance aims to accelerate international data-driven innovation and discovery by facilitating research data sharing and exchange, use and re-use, standards, harmonization and discoverability. Launched in March 2013, financed by the EC (iCordi), the Australian National Data Service and US NSF. EU Open Data Portal: (http://open-data.europa.eu)

Subsets of the complete data sets will be submitted to scientific field specific data repositories such as:

• the plant trait database TRY
• the world meteorological organization (WMO)
• the in-house database of the State Archives
• GBIF and biodiversity.be
• the RMCA cartesius.be project

Finally, we will reach out towards current policy making initiatives to make them aware of the available data as obtained within the project and the potential this has for predicting and assisting future sustainable forestry practices. As such COBECORE will support research into vulnerabilities of African tropical forests. The latter is key in supporting sustainable forest practises such as outlined in the EU FLEGT action plan and yet to be implemented Voluntary Partner Agreement (VPA) within the D.R. Congo.

References

1. Lewis, S. L. et al. Increasing carbon storage in intact African tropical forests. Nature 457, 1003–1006 (2009).
2. Cleland, E. E., Chuine, I., Menzel, A., Mooney, H. A. & Schwartz, M. D. Shifting plant phenology in response to global change. Trends Ecol. Evol. 22, 357–365 (2007).
3. Chuine, I. et al. Historical phenology: Grape ripening as a past climate indicator. Nature 432, 289–290 (2004).
4. James, R., Washington, R. & Rowell, D. P. Implications of global warming for the climate of African rainforests. Phil. Trans. R. Soc. Lond. B (2013).
5. Washington, R., James, R., Pearce, H., Pokam, W. M. & Moufouma-Okia, W. Congo Basin rainfall climatology: can we believe the climate models? Phil. Trans. R. Soc. Lond. B. Biol. Sci. 368, 20120296 (2013).
6. Verbeeck, H., Boeckx, P. & Steppe, K. Tropical forests: include Congo basin. Nature 479, 179–179 (2011).
7. Kearsley, E. et al. Conventional tree height-diameter relationships significantly overestimate aboveground carbon stocks in the Central Congo Basin. Nat. Commun. 4, 2269 (2013).
8. Doetterl, S. et al. Aboveground vs. Belowground Carbon Stocks in African Tropical Lowland Rainforest: Drivers and Implications. PLoS One 10, e0143209 (2015).
9. Ordoñez, J. C. et al. A global study of relationships between leaf traits, climate and soil measures of nutrient fertility. Glob. Ecol. Biogeogr. 18, 137–149 (2009).
10. Kunstler, G. et al. Plant functional traits have globally consistent effects on competition. Nature 529, 204–207 (2016).
11. Beerling, D. & Kelly, C. Stomatal density responses of temperate woodland plants over the past seven decades of CO2 increase: a comparison of Salisbury (1927) with contemporary data. Am. J. Bot. 84, 1572–1583 (1997).
12. Franks, P. J. & Beerling, D. J. Maximum leaf conductance driven by CO2 effects on stomatal size and density over geologic time. Proc. Natl. Acad. Sci. 106, 10343–7 (2009).
13. Wright, I. J. et al. The worldwide leaf economics spectrum. Nature 428, 821–7 (2004).
14. Couralet, C., Van Den Bulcke, J., Ngoma, L. M., Van Acker, J. & Beeckman, H. Phenology in functional groups of central african rainforest trees. J. Trop. For. Sci. 25, 361–374 (2013).
15. Ploton, P. et al. Assessing aboveground tropical forest biomass using Google Earth canopy images. Ecol. Appl. 22, 993–1003 (2012).
16. Barbier, N., Couteron, P., Gastelly-Etchegorry, J.-P. & Proisy, C. Linking canopy images to forest structural parameters: potential of a modeling framework. Ann. For. Sci. 69, 305–311 (2011).
17. Kattge, J. et al. TRY - a global database of plant traits. Glob. Chang. Biol. 17, 2905–2935 (2011).