Meeting minutes of the first COBECORE network workshop on the 6th of November 2017 (2017-11-06) at Het Pand, Ghent, Belgium.
- Emmanuel Kasongo (Follow-up committee, Unikis, DRC)
- Rob Allen (Follow-up committee, Met-Office, UK)
- Stefan Hauser (Follow-up committee, IITA, Nigeria)
- Jan Verbesselt (Follow-up committee, WUR, NL)
- Eric Tollens (Follow-up committee, KULeuven, BE)
- Inge Jonkheere (Follow-up committee, FAO, Italy)
- Tillo Behaeghe (ex-INEAC, BE)
- Pascale Boeckx (UGent, BE)
- Tom De Mil (RMCA, BE)
- Thales d’Hauville (Ulg - RMCA, BE)
- Quentin Groom (Plantentuin - Meise, BE)
- Kim Jacobsen (COBECORE - RMCA, BE)
- Steven Janssens (COBECORE - Plantentuin Meise, BE)
- Sophie Meeus (COBECORE - Plantentuin Meise, BE)
- Patricia Mergen (RMCA - Plantentuin Meise, BE)
- Innocent Banzy Ngulu-kulu (COBECORE - student UGent, BE)
- Piet Stoffelen (COBECORE - Plantentuin Meise, BE)
- Hans Verbeeck (COBECORE - UGent, BE)
- Filip VandeLook (COBECORE - Plantentuin Meise, BE)
- Jan Vandenbulcke (COBECORE - UGent, BE)
Introduction, round-table presentation (name, occupation past or present, research interests)
09:40 - 10:00:
Hans Verbeeck - The importance of COBECORE for long-term ecological and climatological research in the Congo Basin
10:00 - 10:20:
Kim Jacobsen - COBECORE activities at the RMCA and the State Archives
Q & A:
KAOWARSOM weather records (?) https://www.kaowarsom.be might be a good source of additional information
Koen Hufkens: Above all they have streamflow measurements if I’m not mistaken. (see question below). There is reference to this in the index as well.
Are there thermographs etc –> they can scan the graphs from pen recorder and digitize the data these contain
Kim Jacobsen: Yes, these have been observed in the archives but not digitized and not in vast quantities. We will keep an eye out for them. Problem is that we cannot take any documents out of the state archives. But sending photo’s to Rob is also a possibility.
Stresses that digitizing as much as possible is important – the particular use for specific data may only become apparent later on. Keep the photo’s. Rob asked if there are records from boats along the Congo river.
Koen Hufkens: I acknowledge this, the idea is also not to toss anything. We keep all pictures (raw). And extract what is necessary. However, in the interest of time, a selection has to be made. Meta-data might serve as pointers for new tasks. Innocent fills in some of this with doing high resolution work in combination with the daily data.
Kim Jacobsen: As to the boats, not to her knowledge.
Proba-V tools are available (to help with geo-referencing tasks?) eg. https://proba-v-mep.esa.int/ Her contact person is Dries Jeroen email@example.com
Koen Hufkens: QGIS will do this just fine for this particular task.
There are two initiatives which have also focused on teledetection / aerial photography from the time period after independence, 1970’ 1980’: (1)“SPIAF” (Canada (Laval); more information through the Ministère de l’environnement RDC; ask Inge Jonkheere or Emmanuel Kasongo for contact information; prof. Marius ‘Pinot’, Canadian mission in KIN); (2) A similar project was DIAF. Inge works with them and can contact them
Koen Hufkens: This is an interesting lead, and might give a publication with a more long term continued perspective.
Thales d’Hauville: Other flight plans available in the museum. Perhaps we should investigate which flight plans contain a complete coverage of photo’s
10:20 - 10:40
COBECORE activities at the Botanic Garden Meise
Mentioned one species that has oil ‘alabakia’, a project with UNILEVER
Tom De Mil:
Stomata/vessel measurements and coordination between the Botanic garden and RMCA is needed to achieve full exploitation of this dataset (see also HERBAXYLAREDD)
15N, C13 ISOFYS can contribute only 2 mg is needed, temporal and spatial variation is interesting to study
Koen Hufkens: I would take up Pascal on this offer. However, if the methodology uses nail varnish we might have issues. If you apply varnish you must know where because it can contaminate your C isotope signal if residue remains on the specimen. This is the reason I explored if this could be done without chemical interference back in 2013 together with Piet (just optical with a high res / or top illuminated microscope / binocular). I’m not sure what is currently implemented. Furthermore, some of the specimen might have been treated with compounds that might interfere with the analysis. We have to track the record of specimen.
General discussion (11:00 - 12:30)
There are drone images from CIFOR (contact Nils Bourland, RMCA); also images from a lidar flight (Saatchi ?). A suggestion to include a page with useful links on the COBECORE website
The outcome of this project must eventually be a practical application of the data, an ability to forecast 25-30 years ahead for forest based farming systems (answer: focus during second half of the project). What alternatives are there for the local people in order not to ‘eat’ the forests (answer: the FORETS project of CIFOR will focus on that); There is also a book is coming out ‘african cropping systems’ by eric, stefan and others… chapter on forest-based systems.
The larger time frame and implications of climate change on agriculture, deforestation and agro-forestry in the future are of paramount importance. How will precipitation and temperature impact these domains. But it is important to recognize that it is deforestation for charcoal/energy that is driving deforestation in the DRC and not agriculture, which much less exploitative by comparison, due to traditional practices and low intensity. Imports remain much more important compared to production in all Central African countries, except Cameroon.
We need dynamic planting calendars. Perhaps a larger scale review of all research done in Yangambi. We should also collaborate more with INERA to ensure data uptake by the DRC partners after COBECORE has ended. INERA is missing data from 1958 – 1980 (to complement emmanuael and innocent)
Koen Hufkens: This is a nice pitch indeed. How do planting calendars shift with respect to climate change (or have shifted). What are the effects of contraction of the season (with respect to various parameters).
Rob Allan & Quentin Groom:
Make sure to finish entirely digitization of a dataset – not too piecemeal, because that is then useless.
Koen Hufkens: This should not be a problem. There is also redundancy, all work packages are valid research wise in their own respect. Also, we would upload data site by site (with priority sites first). This will guarantee that data will be finished. This is what I did with Jungle Rhythms, you upload in batches.
Ispra-JRC has programs on deforestation in the DRC, including remote sensing data (teledetection); spot images (SPOT?)/satellite data/deforestation from various angles; conference in Kinshasa (?); Eric can give contacts, might even co-sponsor, Gregoire Dubois teledetection unit.
Koen Hufkens: I know Astrid Verheggen who works on this using Google Earth Engine. I can contact her as well.
Data should be published in international repositories, not just on the COBECORE website, putting data in repositories should start early in the project and be planned into the process early on.
Koen Hufkens: Ok, but the data should be well curated and of a consistent quality, hence should not be put out there before curation has been applied (which implicitly means a formal publication which describes the methodology).
Data sets should be large enough. This is particularly important for machine learning/AI; another BRAIN project by KIK: AI on old maps
The Google earth engine project (!) might be a useful platform to contribute the images of the aerial photography to; Other relevant initiatives include the planet archives (2m spatial resolution); INSIGHT (another BRAIN project);
Koen Hufkens: I develop code for GEE, I second that this is a very powerful tool. Planet archives are commercial data and expensive! I still have two scenes from the COBIMFO project (wordview 2 data at 2m).
Annual INEAC reports are very well documented – check them out, they might provide important context and protocols. They were very well documented.
Koen Hufkens: Correct, these are those annual reports as well as the bulletin. However, and Kim has stressed this, this STILL remains grey literature which is prone to be lost. A side project could be to digitize the bulletin formally (some of this has been done by other libraries, but not the whole series).
In the time of Mobutu – 1972 … “ERTS Zaire” was a project with reporting directly to the president (not under a ministry) – it was a teledetection/remote-sensing project. They hired the best students to conduct resource inventories (for mining, forestry). Check serdat library perhaps? – look around in Kinshasa? Ask Emmanuel/Inge? A paper was published in 1994 in Human Ecology
An interactive or geo-spatial page with the other aerial flight plans would be interesting; focus on gradients (?)
Koen Hufkens: This exists, I refer to Francois Kevryn for this. The flight plans have been mapped and are actually online (outlines not content – I think Cartesius covers this).
Together, Stefan Hauser and Emmanuel have meteo data for Yangambi and other stations. Hauser data is mostly for agronomy relevant stations (eg. M’vuazi) Emmanuel digitized data since 1980 to now. We could make a full chronology for Yangambi climate data. Stefan has 1 jan 1961 –> 2009, after 2010-2011, and also other INERA sites –> COBECORE could focus on those sites – to generate very long-term data sets (contact Stefan for exact stations)
14:00 - 14:30:
Eric Tollens, IFA/INERA after 1970
Tollens presentation discusses various crops: arabusta coffee, tenera oil palm and suggests digitising annual reports is important (see also presentation). Eric Tollens discusses with Tillo Behaeghe on how the last years before and first decades after independence were. There were enormous rates of return, very high investment rates by the Belgian government in agricultural research. The eqiuivalent of 2 billion USD today! In return, world-class agricultural developments (eg. Oil palm, cotton export, rubber – coffee …). Question remains: Why was so much invested? What was the idea? Given high rate of return – you could actually say that there was an underinvestment…
Recounts the last years around independence.
The problem now is that INERA does not diversify their investments – does not use the niche function of individual stations; does whatever where ever. Whereas INEAC had specific goals for specific stations (some where agroforestry, some were annual crops or livestock …)
Why doesn’t Belgium help? Or at least give a “plan de relancement”. Especially, given that there are still many projects in Yangambi, investments are ongoing …?
Eric Tollens: Congo financed itself, with Congo-generated money; in fact Belgium loaned several times from “Congolese” banks. There have been several “plans de relancement” at least 5 serious ones since independence – each has failed. Yangambi is not the ideal location for research into food crops; M’vuazi, Boketa, Gandadjika, other locations are better for food crops. The political context is no longer there for Belgium to invest large scale into Yangambi or INERA.
Stefan Hauser: Yangambi is too remote to be relevant today. Location Location Location
Eric Tollens (would like to look into publish rate of return – historical context – data – research context); will contact Kim in about a month.
15:00 - 15:30:
Rob Allan, ACRE, Copernicus C3S DRS & COBECORE (see also presentation)
ACRE has been active over 10 years. It receives little subsidy but is a very dynamic community. There is a German professor who will help with digitization if it’s surface pressure data; he has students that he can use for this. ISPD, ICOADS (?), 20CR (20th century re-analysis), 20CRv2c 1851-2012 (google and find links); INDARE (funding body that Rob has applied to often); similar data-rescue initiatives are Old weather; weather detective. It is easier to generate interest from the public for marine data crowd sourcing compared to terrestrial data, because of the historically interesting anecdotes of marine life. International journal of climatology DOI:10.1002/joc.4775 ICOADS release 3.0: a major update to the historical marine climate record (http://onlinelibrary.wiley.com/doi/10.1002/joc.4775/abstract)
Koen Hufkens: This is an important incentive for us to digitize folder 6119 which contains pictures of some of the stations (~120 if I’m correct). People like stories, and want to attach an image to the work. Might be nice for Innocent to do this if he has time left. It would be nice material to illustrate his thesis with.
15:30 - 16:00:
Quentin Groom, Citizen Science at the Botanic Garden (see presentation)
Closing remarks by follow-up committee:
At a particular point in time you need to go to the media and explain what you are doing. The “treasure” of yangambi, generated so much data, more than 50 years after independence and we must tell the general public about this. It is of strategic importance. Broad based public support is necessary. The fundamental research, about soils, climate, biodiversity, etc (200+ university trained scientists) … that it is being valorized now– that this is a public good. Should be stressed!
Koen Hufkesn: Although I couldn’t get stuff into the ENSIA publication Dan Grossman remains a good contact and can probably really help with this. If the results paint a consistent picture and he can probably generate some buzz.
We should increase the public reached by media ; emphasize also other projects of significance in the past/ongoing; outreach, vulgarization, PR/PA; not only climate but also socio-economy (farmers, foresters, impact of climate);
This data should go wider than only the “small” scientific world; GEO network platforms; to FAO via FAO platforms (aerial images), a larger collaboration with FAO/WUR (processing/LULCC) for processing and data hosting
Global Framework Climate Services (wider databases / science but also services and broader outreach)
Good opportunity to connect past-present-future