FluxLetter

The Newsletter of FLUXNET Vol. 2 No. 1, March, 2009

Highlight FLUXNET sites Hinda, Kissoko and Tchizalamou flux tower sites Hinda, Kissoko and Tchizalamou by Yann Nouvellon

Three main types of UR2PI in these eucalypt planta- and the -covariance system In This Issue: are found in the South-Western tions as part of a larger network and meteorological station were Republic of Congo (Central of flux tower sites in tropical then moved to a nearby savan- Africa): secondary natural for- plantations (rubber-tree, coco- nah at Tchizalamou (4°17'21.0''S, ests, savannahs, and eucalypt nut-tree, coffee and eucalypt 11°39'23.1''E), where flux meas- FLUXNET site: plantations (about 40000 ha) that plantations) managed by the urements have been carried out “Hinda, Kissoko and Tchi- have been established over sa- CIRAD “Tree-based Tropical since July 2006. The three sites zalamou ” vannahs since the 1970s for Plantation Ecosystems” research are part of the CARBOAFRICA Yann Nouvellon ……...….Pages 1-3 team (http://www.cirad.fr/ur/ flux tower site network (http:// pulpwood and charcoal produc- tion. The Hinda (4°40'52''S; 12° ecosystemes_plantations (in www.carboafrica.net; see the Editorial: 00'13''E) and Kissoko (4°47'29''S; French only). The Hinda and “Carbon cycling in Sub-Saharan Why is it so hard to sustain a Kissoko sites were operated Africa” special issue in Biogeo- 11°58'56''E) flux tower sites flux network in Africa? were established by CIRAD and successively for two years each, sciences for some early pub- Bob Scholes……….…...... Pages 4-5

Celebrating the first year of FluxLetter Rodrigo Vargas and Dennis Baldoc- chi……….…...….…...... Page 6

FLUXNET graduate student : Sally Archibald ...... Page 7

FLUXNET young scientist: Agnes de Grandcourt ...... Pages 8-9 Christopher Williams ….Pages 10-11

Research: Flux-measurements in a Miombo woodland in Western Zambia in relation to defores- tation and degradation Werner L. Kutsch, Lutz Merbold, Mukufute Mukelabai, and Maurice Muchinda…………....Pages 12-15

Figure 1: Closed-path (Li-6262+ sonic Young 81000V) and open-path(Li-7500 + 81000V) systems for flux measurements at the Hinda site (two years of measurements with the closed-path system, and one month with the two systems, for comparison)

FLUXNET SITE cont. on page 2 Page 2

Hinda, Kissoko and Tchizalamou flux tower sites FLUXNET SITE cont. from page 1

lished CARBOAFRICA results) and the French network F-ORE- T (http://www.gip-ecofor.org/f- ore-t/). Measurements made at these sites aim at: i) quantifying and understanding the exchanges of water vapour (Fisher et al.,

2009), CO2 and energy between the and the atmos- phere, ii) identifying the main regulators of gross and net pri- mary production (GPP, NPP) and net ecosystem production (NEP), and iii) parameterizing/ validating ecosystem process- based models for these two types of ecosystems. The sea- sonality of net primary produc-

tion (NPP), CO2 effluxes (Epron et al., 2004), fine root growth (Thongo et al., 2008), LAI and specific leaf area (Nouvellon et al., 2009) have Figure 2: Coarse root CO2 efflux measurement at the Kissoko site been studied at these sites. At Kissoko, NEP has been assessed At the Tchizalamou savannah soko, and Tchizalamou flux Africa surface. Another chal-

in two ways: 1) from eddy- site, soil CH4 and N2O effluxes tower sites, nutrient cycling at lenge in these studies is to covariance measurements of are currently being monitored, adjacent eucalypt and savannah assess the environmental im-

CO2 net ecosystem exchange as well as the seasonal dynamics ecosystems has also been stud- pact of eucalypt plantations, and (NEE), and 2) from biometric of LAI and NDVI. These data are ied: for 10 years soil water more specifically the effect of NPP measurements and hetero- being merged with data obtained content and soil solution chem- savannah on car- trophic soil respiration (Rh) at other savannah sites of the istry have been measured con- bon sequestration, water re- measurements (NEP=NPP-Rh). CARBOAFRICA flux tower site tinuously down to a depth of 5 sources, and soil fertility. At the Hinda and Kissoko sites, network (in South-Africa, Bot- m, as well as nutrient fluxes Chronosequence studies have canopy structure parameters, swana, Sudan, Mali, Niger, Benin, between the soil and the plants been carried out to assess the gap fractions, leaf gas exchange, etc.; Merbold et al., 2008) in (e.g. Laclau et al, 2005). This changes in carbon and nutrient

and root CO2 efflux (Marsden et order to develop/validate provides a unique opportunity stocks in the soil (D’Annunzio al., 2008) have been measured in “process-based” models simulat- to study the interactions be- et al., 2008), and in the biomass

order to parameterize canopy or ing water vapour and CO2 ex- tween the carbon, water, and (Saint-André et al., 2005) after

individual tree-based models changes between savannas and nutrient cycles. This is impor- savannah afforestation. Soil CO2 (CANOAK, G’DAY and the , that will further tant since the response of Afri- effluxes after savannah affore- MAESTRA). Leaf gas exchanges be applied at the regional and can savannas to station (Nouvellon et al., 2008) measurements, and eddy- continental scales, using remote is likely to be determined by and changes in the isotopic covariance measurements have sensing data, soil and vegetation the complex interactions be- composition of soil C and of

both shown marked differences cover maps, and meteorological tween these three major cycles, soil CO2 effluxes have also been in the water-use efficiencies of data. and since savannas cover about measured in order to investi- the two studied eucalypt clones. In the vicinity of the Hinda, Kis- two thirds of the sub-Saharan gate the dynamics of savannah-

FLUXNET SITE cont. on page 3 Page 3

Hinda, Kissoko and Tchizalamou flux tower sites FLUXNET SITE cont. from page 2

PI contact information: trenched-plot technique. New Phy- Yann Nouvellon : tologist, 177:676-687. Merbold, L., Ardo, J., Arneth, A., [email protected] Scholes, R.J., Nouvellon, Y., de Grand- Agnès de Grandcourt : court, A., Archibald, S., Bonnefond, [email protected] J.M., Boulain, N., Bruemmer, C., Brueggemann, N., Cappelaere, B., Ceschia, E., El-Khidir, H.A.M., El-Tahir, B.A., Falk, U., Lloyd, J., Kergoat, L., Le Further Reading Dantec, V., Mougin, E., Muchinda, M., D’Annunzio R., Conche S., Landais D., Mukelabai, M.M., Ramier, D., Roup- Saint-André L., Joffre R., Barthes B., sard, O., Timouk, F., Veenendaal, E.M., 2008. Pairwise comparison of soil and Kutsch, W. L., 2008. Precipitation organic particle-size distributions in as driver of carbon fluxes in 11 African native savannas and Eucalyptus planta- ecosystems. Biogeosciences Discuss., tions in Congo. Forest and 5, 4071–4105. Management 255:1050-1056. Nouvellon, Y., Epron D., Laclau, J-P., Epron, D., Nouvellon, Y., Roupsard, Kinana, A., Mabiala, A., D’Annunzio, R., O., Mouvondy, W., Mabiala, A., Saint- Deleporte, P., Saint-André, L., Mars- André, L., Joffre, R., Jourdan, C., den, C., Roupsard, O., Bouillet, J-P., Bonnefond, J.M., Berbigier, P., Hamel, and Hamel, 0., 2008. Soil CO2 efflux O., 2004. Spatial and temporal varia- and soil carbon balance following tions of soil respiration in a Eucalyptus savannah afforestation in Congo: plantation in Congo. Forest Ecology comparison of two site preparation and Management, 202 :149-160. treatments. Forest Ecology and Man- Epron, D., Marsden, C., Thongo agement, 255: 1926-1936. m’bou, A., Saint-andré, L., D’Annunzio, Nouvellon, Y., Laclau, J-P., Epron, D., R., Nouvellon, Y., 2009. Soil carbon Kinana, A., Mabiala, A., Roupsard, O., dynamics following afforestation of a Bonnefond, J-M., Le Maire, G., Mars- tropical savannah with Eucalyptus in den, C., Saint-André, L., 2009. Within- Congo. Plant and Soil (in press). stand and seasonal variations of spe- Fisher, J.B., Malhi,Y., de Araújo, A.C., cific leaf area in a clonal Eucalyptus Bonal, D., Gamo, M., Goulden, M.L., plantation in the Republic of Congo. Hirano, T., Huete, A.R., Kondo, H., Figure 3: Soil CO2 effluxes measurements at the Tchizalamou savannah site Forest Ecology and Management Kumagai, T., Loescher, H.W., Miller, (submitted) S., Nobre, A.D., Nouvellon, Y., Ober- derived, and newly eucalypt- CSIR, B. Scholes; DISAFRI, R. Saint-André L., Thongo M’Bou A., bauer, S.F., Panuthai, S., von Randow, Mabiala A., Mouvondy W., Jourdan C., derived soil C after savannah Valentini, etc.). C., Rocha, H.R., Roupsard, O., Saleska, Roupsard O., Deleporte Ph., Hamel afforestation (Epron et al., 2009). S., Tanaka, K., Tanaka, N., and Tu, O., Nouvellon Y. 2005. Age-Related New research (planned but not Research team: K.P., 2009. The land-atmosphere equations for above- and below- water flux in the tropics. Global yet funded) will focus on carbon, CIRAD: L. Saint-André, Y. Nou- ground biomass of a Eucalyptus hybrid Change (in Press). water, and nutrient cycles in the vellon, A. de Grandcourt, O. in Congo. Forest Ecology and Manage- Laclau J-P., Ranger J., Deleporte Ph., ment 205:199-214. secondary natural . Hamel, O. Roupsard, JP Laclau, Nouvellon Y., Saint-André L., Marle S., Thongo M’Bou, A., Jourdan, C., Dele- Our research is carried out in C. Jourdan, JP. Bouillet, C. Mars- Bouillet J-P. 2005. Nutrient cycling in a porte, P., Nouvellon, Y., Saint-André, collaboration with scientists den, P. Deleporte, L. Mareschal, clonal stand of Eucalyptus and an A., Bouillet, J-P., Mialoundama, F., adjacent savanna ecosystem in Congo. from INRA, University of Nancy, G. Le Maire Mabiala, A., Epron, D., 2008. Root 3. Input-output budgets and conse- elongation in tropical Eucalyptus planta- CNRS (ESE and CEFE), Second UR2PI: A. Thongo M'Bou, A. quences for the of the tions: effect of soil water content. University of Naples (S. Castaldi Kinana, A. Mabiala, Y. plantations. Forest Ecology and Mana- Annals of Forest Science, 65 (609): 1- gement 210:375-391. and A. Rasile; N2O and CH4 flux Nouvellet (Head of UR2PI) 7. measurements), and scientists INRA: JM Bonnefond, M. Irvine, Marsden, C., Nouvellon, Y., Thongo, Saint-André, L., A., Jourdan, C., Ki- from other research Institutes J. Ranger, B. Zeller ( see http://www.cirad.fr/ur/ nana, A., Epron, D., 2008. Two inde- ecosystemes_plantations/publications involved in CARBOAFRICA University of Nancy: D. Epron pendent estimations of stand-level for other references related to these (Max Plank Institute, W. Kustch; root respiration on clonal Eucalyptus sites) stands in Congo: up scaling of direct measurements on roots versus the Page 4

Why is it so hard to sustain a flux network in Africa? Bob Scholes

The processes which eddy co- why there should be many more where in Africa? variance equipment measures – flux measurements in Africa. But The most obvious - but not moving from this agreement to necessarily the greatest - obsta- “...The biggest single bar- transpiration and carbon assimi- rier, in my experience, is lation – are fundamental to many an implemented, operational, self cle is funding. Very few African of the practical ecosystem man- -sustaining network has proven countries have national competi- the shortage of skilled agement issues in Africa. Fur- to be much harder. This article tive research grant programmes technicians. .” discusses some of the obstacles, – and when they do, the awards thermore, the rising profile of the UN Framework Convention in order that we can realistically are far too small to establish and on Climate Change and the work out how to surmount maintain a flux site. African re- “...A related challenge is potential for the trading of car- them. searchers can access various finding African students bon credits means that the car- I have been the co-investigator international funds, but the entry with the necessary bon cycle itself is of interest to at the Skukuza flux site since barriers are daunting. The first is mathematical skills to researchers and policymakers in 2000. This is probably the long- simply the knowledge, experi- deal with complex flux est flux record – patchy as it is – ence and confidence to write a Africa. But the present global data, and the interest in distribution of flux measure- anywhere in Africa. I work for a winning proposal. The second is ments is massively skewed to- large scientific organisation in that these awards are often very doing so in a ‘marginal’ wards the northern hemisphere, one of the most technically- large - and the home institutions field such as ecology” advanced countries in sub- of the African researchers are developed, extra-tropical coun- tries. Any global synthesis of flux Saharan Africa. If it is so hard for not equipped to satisfy the ac- -derived data runs the risk of my team to keep the Skukuza counting rules for disbursing large biases as a result. There- site functioning, how much them effectively. The third is that require international partners – fore it is easy to make a case for harder is it for researchers else- the international funds inevitably not a bad thing in itself, but hard for an African researcher to set up unless they are already part of the club. The biggest single barrier, in my experience, is the shortage of skilled technicians. Running a flux site successfully requires a good working knowledge of electron- ics, instrument maintenance, calibration and trouble-shooting, data handling and telecommuni- cations, as well as some basic micrometeorology and ecologi- cal knowledge. Finding this pack- age of skills in a single individual, and then being able to retain this relatively highly-skilled person under difficult field conditions on an uncompetitive salary is a great challenge. A related challenge is finding African students with the neces- Figure 1. The CarboAfrica training workshop on flux site characterisation, held at Phalaborwa, South Africa in March 2008. The fifteen re- searchers, from all parts of Africa, were learning how to collect information to satisfy the basic FluxNet package of information on vegetation, sary mathematical skills to deal and climate. Bob Scholes is explaining how to record tree height; the workshop organiser Rudzani Makhado looks on from the left. with complex flux data, and the Behind him is an armed game guard, because the field practicals took place in the Kruger National Park. Werner Kutsch and Lutz Merbold (third from the left) of the Max Plank Institute at Jena were part of the training team. cont. on page 5 Page 5

Why is it so hard to sustain a flux network in Africa? cont. from page 4

interest in doing so in a populations. an under-represented ecosystem ‘marginal’ field such as ecology. Poor infrastructure is probably or region is not enough). There South Africa, with its recent “..Filling the data gaps in the single greatest developmen- are no African journals that are apartheid history of poor maths Africa will require genu- tal challenge in Africa. Getting a suitable, and which might be and science education for black ine, long-term partner- bottle of calibration gas to a more supportive. But without Africans, is particularly crippled ships with developed- remote site often doubles or the endorsement of peer- in this regard, but the problem is country researchers, a trebles its cost. When an instru- reviewed publications, the Afri- widespread in Africa – the substantial amount of ment fails, it usually takes can researcher cannot continue brightest students are siphoned money (largely from out- months to get it to a service to compete for the international off into banking, commerce and side Africa), well-targeted agent, have it repaired and then funding needed to sustain the government. The majority of reinstalled. Communications work. A classic catch-22. African PhDs end up working and sustained capacity bandwidth limitations prevent The above litany of woes is outside their fields, and outside building (including in the large raw flux data files from neither a complaint nor an ex- of Africa. such non-traditional ar- being regularly transmitted and cuse. But it is an appeal for un- To purchase and service flux site eas as proposal writing, analysed, meaning that problems derstanding, and a request for instruments in Africa is at least paper publication and are often only detected long thoughtful assistance. Filling the 50% more expensive than in technician training), and after they have occurred. data gaps in Africa will require Europe or North America. The Having overcome these hurdles, genuine, long-term partnerships equipment suppliers work above all, patience. ..” the proud operator of a new with developed-country re- through a system of local agents, African flux site is still not home searchers, a substantial amount who add a mark-up, but do not and dry. For a whole range of of money (largely from outside have the capacity to supply much general, the natural hazards reasons, it is hard for African Africa), well-targeted and sus- technical support. The import faced by flux sites in Africa are researchers to get their work tained capacity building (including process often involves punitive quite extreme. The tower is a published in the international in such non-traditional areas as taxes and substantial delays. magnet for lightning, which regu- literature. When flux measure- proposal writing, paper publica- Currency fluctuations make larly fries our delicate electron- ments in developed countries tion and technician training), and budgeting a gamble. ics. Instruments fail when the were in their start-up phase two above all, patience. Have your cables been chewed temperatures climb into the high decades ago, it was possible to by a hyena, or your solar panels forties (Celcius). Vegetation publish a short or imperfect flux trashed by a grumpy rhinoceros? fires, an ecological feature of record in a good journal. That is contact: Admittedly, the Skukuza site has savannas, can be expected every no longer the case. Acceptance Dr. Bob Scholes been deliberately located in the year or two. Vandalism or theft now requires a long record, lots CSIR, Pretoria, South Africa Kruger National Park, and those is a continual problem where the of supplemental information, and [email protected] are the risks you take. But in tower is within reach of human some new angle (simply being in Page 6

Celebrating the first year of FluxLetter

Rodrigo Vargas and Dennis Baldocchi

The FLUXNET office has been and six young scientists from all you very much for your contri- If you want to contribute to any editing FluxLetter as a quarterly over the to show the butions and support! section of FluxLetter please diversity in research and person- contact Rodrigo Vargas newsletter to present news, FluxLetter is starting its second ality among the community. In ([email protected]) sites, results, and opinions of year with this issue that high- addition, we have highlighted or Dennis Baldocchi interest to the FLUXNET com- lights opinions, sites, students, FLUXNET sites in Asia, Africa, ([email protected]) munity. With this issue we cele- young scientists and research America, Europe, and Oceania brate our first year of this effort done in African ecosystems. We to show specific examples of THANK YOU!! and we continue working to invite scientists around the research done in this growing make FluxLetter a powerful world to propose new sections, network. information, networking, and send opinions and contribute During this past year it has been communication resource for the with highlight research and a pleasure to interact with all the community. To date we have FLUXNET site sections. Our collaborators that have sup- published four issues of FluxLet- goal for the following issues is to ported and contributed for the ter which are available on-line at create a space to show the di- published issues of FluxLetter. two websites: (http:// versity of research, study sites The success of this newsletter is www.fluxnet.ornl.gov/fluxnet/ and opinions along the only possible because of the newsletters.cfm network. or http://bwc.berkeley.edu/ invaluable support that we have FluxLetter/). We have high- received from FLUXNET scien- lighted five graduate students tists around the world. Thank Page 7

Highlight Graduate Student Sally Archibald

My interest in environmental the extent to -term flux measurements for a science started with an interest which a land- savannah system. Data from this in botany – specifically the evolu- scape burns tower provide interesting in- tion of life-history traits, but I depends to a sights into land-atmosphere soon became fascinated in large extent on fluxes in systems that are water- broader issues. Punctuated by a the amount of controlled – where most of the brief period as a news reporter I fuel and the biological activity occurs in short went from a Masters Degree at connectivity of bursts associated with rainfall the University of Cape Town - the fuel bed. events. The coexistence of two where I investigated how fire and Currently different life forms – trees and grazing interact to control vege- people in Af- grass also results in interesting tation - to a job at the CSIR in rica only seem dynamics. We have recently Pretoria. At the CSIR I have the to influence published a paper looking at the opportunity to work on my PhD fire by decreas- drivers of inter-annual variability Figure 1 - Sally Archibald which is looking at the drivers of ing its extent – not by increasing in this system. fire regimes in Africa. Specifically something that is an integral part it as has been assumed in several I love the fact that my career I am interested in identifying the of a larger vegetation-fire-climate global analyses. Whether this means I can live and work in my limits of climate and human con- system has been an exciting was always the case, when the home country, South Africa, but trol of fire in Africa, and the intellectual journey. Especially human impacts on fuels and still interact with scientists from extent to which fire regimes can when one adds humans into the landscape connectivity were less, all over the world. I am excited be manipulated. picture. is not certain. I am involved in a to expand my research to a Switching from a purely ecologi- I am registered for my PhD at theoretical modelling exercise to global scale when I finish my cal to a more biogeochemical the University of the Witwaters- explore this question. I want to PhD. and systems approach has been rand in Johannesburg. One inter- determine the impact humans illuminating for me; and changing esting result from my thesis have had on fire regimes over contact: Sally Archibald my view of fire as something that work is that ignitions never seem paleo-ecological time. [email protected] affects plants and animals, to to be limiting fire in Africa – and Some interesting thresholds appear in the relationships be- tween fire, vegetation and cli- Further Reading mate. Very little fire is apparent Archibald, S., Roy, D. P., Van Wilgen, B. at tree covers of 40% or greater W., and Scholes, R. J.: What limits fire?: An examination of the drivers of burnt – and it is also obvious that area in southern Africa, Global Change there are large parts of Africa Biology, 15, 613-630, DOI: 10.1111/ which are being maintained at j.1365-2486.2008.01754.x, 2009. tree covers of < 40%. Currently Archibald, S. A., Kirton, A., van der I am visiting the Department of Merwe, M. R., Scholes, R. J., Williams, Ecology and Evolutionary Biology C. A., and Hanan, N.: Drivers of inter- at Princeton University where I annual variability in Net Ecosystem am working with other graduate Exchange in a semi-arud savanna eco- students to model fire’s role in system, South Africa, Biogeosciences, 6, 1-16, 2009. these vegetation boundaries.

While at the CSIR I have the Archibald, S., and Scholes, R. J.: Leaf opportunity to be involved in green-up in a semi-arid African savanna several other research projects. - separating tree and grass responses to The Skukuza flux tower in the environmental cues, Journal of Vegeta- tion Science, 18, 583-594, 2007. Kruger National Park is one of a handful of towers which has long Figure 2 - The relationship between rainfall, tree cover and fire in southern Africa. Fires hardly occur above about 40% tree cover, but most of the region is also maintained at tree covers less than 40% - presumably largely due to the effect of fire. Page 8

Highlight Young Scientist Agnes de Grandcourt

I was born in Paris, France. As grasslands, forests and lakes discover the world of the re- eucalyptus (in Brasil), hevea (in granddaughter and daughter of inside the campus. So we could search with trainings. I started Thailand) and coffee (in Costa- foresters I spent soon a lot of learn to recognize birds’ songs with a 6-months study in eco- Rica). I arrived in Congo in June time outside, during the week- or plants during the course, physiology with Alexandre Bosc, 2006. I am associated with ends and the holidays. I started walking outside the classroom. It who was a PhD student at INRA UR2PI, a research center fo- to help my father during forest may obvious for a US Bordeaux. I applied for the first cused on plantation productivity. inventories at 15, during the student, but for a French and time what I learned during my At UR2PI, we are around 30 holidays. more a Parisian student, it was courses. It was exciting to have a members, 8 researchers (3 Con- At 16, I decided to have a job exceptional. At the end of these question and to find how to golese and 5 French) and 7 PhD related to forests. After my 4 years, the French students answer to this question. I made students (6 Congolese and 1 baccalauréat, I made my first 4 have to choose between re- my first respiration chambers… French). I am in charge of an years studying biology and ecol- search and engineering. To be looking a little bit strange but eddy-covariance station, measur- ogy at the University Paris 11. It able to choose, I decide to stop airtight proof. After that, I suc- ing the exchanges above a grass- was a great place to study, with the university for one year to ceeded in finding a job in French land at the Tchizalamou site at Guiana, in the tropical rainforest, 75 km of Pointe Noire, the town which I wanted to see since where I live (more or less 2 years as a forest lover! I worked hours of car, on road and sand in the molecular biology lab of track). I also conduct research INRA for 2 months. on vegetation dynamics (above- After this year, I definitely ground and below-ground). This choose to do research and grassland is very poor and no big spend my fifth year of my univer- herbivores can live here. Unlike sity studies at Nancy (east of some other sites in Africa, we France) studying Forest biology. have no problem of vandalism During 10 months, I studied the caused by the elephants! There growth and respiration of roots is nothing bigger than jackals. But of tree seedlings at different light the jackals sometimes eat the levels. I did my doctorate in connectors of the soil French Guiana (at 4°N) on the probes!! We also have to man- acquisition and the utilization of age the fire, as villagers burn the phosphorus by forest seedlings grass every year. The guardians associated with endomycorrhi- at the site make a great job to zas, in relation with soil phos- avoid the eddy-covariance sta- phorus and light. tion to be roasted! After my PhD, my vision of After more than 2 years of sur- ecophysiology changed of scale vey of the grassland, we hope to and gained height. I was chosen follow the impact of the change by Dr Jean-Pierre Bouillet and of land use, as the grassland is CIRAD (a French agricultural supposed to be planted with research centre working for eucalyptus. The plantation was international development) to supposed to be done last Octo- follow the CarboAfrica project ber, but it has to be postponed in Congo (still at 4°, but 4°S this by the eucalyptus plantation time). This CIRAD team is owner, because of the world working on the plantation eco- financial crisis. There are 40.000 systems, and three other eddy- ha of eucalyptus around Pointe covariance sites are running on Noire and this eddy-covariance Figure 1: Agnes de Grandcourt in South Africa for the first CarboAfrica annual meeting. cont. on page 9 Page 9

Highlight Young Scientist cont. from page 8 station was previously above an where I met a lot of very inter- eucalyptus plantation. esting researchers. It was a great The fluxes measured at the chance for me, as a beginner at Tchizalamou grassland site are the eddy-covariance research, to sent to the CarboAfrica data- meet all these persons and to base. With studies in 11 sub- understand better all the chal- Saharan countries, CarboAfrica lenges around this topic. aims to quantify and predict the The CarboAfrica project allows GHG budget of Sub-Saharan me to start cooperation with Africa and its associated spatial other laboratories. Especially, and temporal variability. More the Department of Environmen- locally, I want to analyze the tal Sciences (DSA) of the Second differences of functioning be- University of Naples (SUN) tween the grassland and the (Simona Castaldi and Americo eucalyptus plantation. Rasile) and I decided to study My data has been sent to the the impact of the termites on Figure 2: Agnes de Grandcourt harvesting aboveground biomass. The living and dead biomass is separated by species and organs. FLUXNET database. I was the GHG emissions (). I pleased to participate in the really love to study the interac- biotic factors such as fungi dur- workshop at LaThuile in 2007, tion of the plants with other ing my doctorate, or termites Further Reading during my post-doc. Bosc A, de Grandcourt A, Loustau D, Work in Congo is a great ex- 2003, Variability of stem and branch maintenance respiration in a Pinus perience, even if it is quite hard pinaster tree. Tree physiology 23: 227- to find spares quickly (no big 236 drugstore around) and the inter- net connection can be really de Grandcourt A, Epron D, Montpied P, Louisanna E, Béreau M, Garbaye J, slow! I will be sad to leave the Guehl JM, 2004, Contrasting responses Congo at the end of the year, as to mycorrhizal inoculation and phos- my post-doc will finish in Octo- phorus availability in seedlings of two ber. I hope to see all the Car- tropical rainforest tree species. New boAfrica participants at Pointe Phytologist 161: 865-875

Noire for the final conference Merbold, L., Ardö, J., Arneth, A., Scho- before I leave. All the Fluxnet- les, R. J., Nouvellon, Y., ters will be welcome to our final de Grandcourt, A., Archibald, S., Bon- conference. nefond, J. M., Boulain, N., Bruem- mer, C., Brueggemann, N., Cappelae- re, B., Ceschia, E., El-Khidir, H. A. M., El contact: Agnes de Grandcourt -Tahir, B. A., Falk, U., Lloyd, J., Ker- [email protected] goat, L., Le Dantec, V., Mougin, E., Muchinda, M., Mukelabai, M. M., Ra- mier, D., Roupsard, O., Timouk, F., Veenendaal, E. M., and Kutsch, W. L.,

2008: Precipitation as driver of carbon fluxes in 11 African ecosystems, Bio- geosciences Discuss., 5, 4071-4105.

Figure 3: Map of Central Africa Page 10

Highlight Young Scientist Christopher Williams

My name is Christopher A. process-based modeling. Flux was intiated by doctoral re- Continuing with the modeling of Williams, and I’m pleased to tower data continue to provide search under John Albertson, dryland vegetation dynamics, I have been invited to introduce one of the central data con- CEE at Duke Univ. We took examined how the interannual you to some of my work in Afri- straints, and I have had the good two flux towers to sites in the correlation of rainfall (long- can ecosystems. In general, my fortune of working with out- Central Kalahari (Ghanzi, Bot- period modes) and variability in research focuses on assessing standing colleagues at multiple swana), and set them up in con- annual rainfall may degrade grass environmental controls on eco- flux tower sites (e.g. South Af- system structure, function, and rica, Botswana, Blandy and UVA distribution; inferring what this Experimental Farm and Forest; means for water, carbon, and Duke Hardwood and Pine Sites; energy exchanges; and exploring Sardinia, Italy; and SMACEX of the likely effects on ecosystems Ames, Iowa), as well as in Bot- and the resources they provide. swana and South Africa. Over the years my work has become a loose combination of Highlights from Africa field data, remote sensing, and My focus on African ecosystems

Figure 2: Skukuza, South Africa view

trasting land cover types for a forage resour short intensive field campaign. ces for a range of climate and Not surprisingly we found that grazing settings (Williams et al. soil water is a dominant control GCB 2006), by advancing a clas- on evaporation, plant productiv- sical model of equilibrium dry- ity, and respiration (Williams & land dynamics. …In other Albertson WRR 2004). words, we took about a month Based largely on these field ob- of flux tower data from a rela- servations, I coded up a simple tively remote location in Bot- mechanistically-based numerical swana and used it as a basis to model for the study of multi- comment on interannual vegeta- decadal interactions between tion dynamics of savannas. soil, vegetation, and climate Needless to say, we could use a (Williams & Albertson WRR bit more observational data… 2005). Somewhat unconvention- This brings me to postdoctoral ally, we found that the annual work with Niall Hanan, Colo. surface water balance is largely State’s NREL. Though tasked insensitive to vegetation type with unraveling the roles of and abundance for a wide range climate, vegetation dynamics, of dryland conditions. Instead, fire, and land use in driving diur- vegetation largely tracked annual nal to interannual variability in rainfall, which emerged as the Africa's carbon cycle, I continued dominant control on annual to explore site-based results by plant production. analyzing canopy-scale observa- Figure 1: Chris with daughter Chloe cont. on page 11 Page 11

Highlight Young Scientist cont. from page 9 tions from the Skukuza, South diagnosed physiological and Africa tower. We’ve found structural limitations on photo- some interesting lags in produc- synthesis across Africa (Williams tivity responses to rainfall pulses et al. 2008). Most recently, we and uncovered hysteresis in the have explored how climate oscil- temporal evolution of evapotran- lations induce interannual fluc- spiration, productivity and respi- tuations in rainfall and productiv- ration (in review). ity (in review). Keeping to the continental scale With all of this modeling work, I of the funded project (African can’t help but long to get back in Carbon Exchange, NASA + the field, and so perhaps soon NOAA), Niall and I synthesized you’ll see another dot on the estimates of the major fluxes in global flux tower map. That said, the African and global carbon and to state the obvious, a long- Figure 3: Ghanzi, Botswana field site cycles (Williams et al. CBM term record of high quality, semi 2007), shedding light on the vast -continuous measurements at a welcome continued dialogue fire but slight fossil emissions. site in Africa would require along those lines. Then, using Simple , I international collaborations and I contact: Christopher Williams [email protected]

Further Reading Williams, C.A., N.P. Hanan, I. Baker, A.S. Denning, G.J. Collatz, J.A. Berry, R.J. Scholes, (2008) Interannual variabil- ity of photosynthesis across Africa and its attribution, Journal of Geophysical Research – Biogeosciences, 113, G04015, Williams, C.A., N.P. Hanan, R.J. Scholes, A.S. Denning, J.A. Berry, J. Neff (2007) Africa and the global carbon cycle, Carbon Balance and Management, 2:3. Williams, C.A., and J.D. Albertson (2006) Dynamical effects of the statisti- cal structure of annual rainfall on dry- land vegetation, Global Change Biology, 12, 1-16. Williams, C.A., and J.D. Albertson (2005) Contrasting short- and long- timescale effects of vegetation dynamics on water and carbon fluxes in water- limited ecosystems, Water Resources Research 41, W06005, Williams, C.A., J.D. Albertson (2004) Soil moisture controls on canopy-scale water and fluxes in an African savanna, Water Resources Re- search, 40:1-14,

Figure 4: site during intensive campaign in Ghanzi, Botswana Page 12

Flux-measurements in a Miombo woodland in Western Zambia in relation to deforestation and forest degradation Werner L. Kutsch, Lutz Merbold, Mukufute Mukelabai, Maurice Muchinda

Miombo woodlands cover the transition zone between savan- nas and rainforests in Southern Africa. Since they are located in semi-arid areas with a distinct dry season, they belong climati- cally to savannas; however with about 800 – 1000 mm of annual precipitation the tree coverage can be more than 60 % which is commonly defined as threshold between savannas and wood- lands. Miombo woodlands cover about 3 million km2 in southern Africa. The fact that almost nothing is known about Miombo wood- lands in terms of fluxes (the only data that are available are three weeks of eddy-covariance data measured during the SAFARI- 2000 campaign by Scanlon & Albertsen 2004) motivated us to choose the Kataba Forest near Figure 1: Kataba Forest Reserve during the wet season as seen from the fluxtower. (Picture by W. Kutsch) Mongu, the capital of the West- ern Province of Zambia as one woodlands are affected by high as the resulting impact severely: price per tree we were reported of the core sites of CarboAfrica. forest degradation and defores- charcoal production has become was about 20 Eurocent. Since Figure 1 shows the Kataba For- tation mostly for charcoal pro- a full-time job of migrating work- the buyers are not bound to the est Reserve during the wet sea- duction. Domestic burning has ers that ‘buy’ the trees from land and the ecosystem services son. switched from fuelwood to char- local communities, produce that the Miombo woodlands The forest is located East of the coal during the past decade along charcoal and leave (Fig. 3). The provide, they do not care for Zambezi floodplain on an area with urbanization and population that is part of the Kalahari Sands. increase. The higher charcoal The satellite picture in Figure 2 demands in urban areas have shows that the reserve is em- also changed the supply site. bedded into a landscape highly Whereas in rural areas firewood affected by land use change. The has been mostly produced in light areas are former woodlands small amounts by the demanding that are converted into agricul- households themselves, charcoal tural land or clear-cuts that have is produced in much higher not regenerated yet. amounts and traded over much In Zambia, as in many of its bigger distances. This has neighbouring countries, Miombo changed the production as well Figure 2: Satellite image of the area with the Zambesi floodplain on the left. (Courtesy of Google Earth) cont. on page 13 Page 13

Flux-measurements in a Miombo woodland cont. from page 12

Figure 3: Some impressions of charcoal production and transport in Zambia. (Pictures by W. Kutsch and L. Merbold) sustainable use of the wood- bian per capita emissions be- mental design, particularly, in urement. I addition, we will de- -1 lands. tween 1.6 and 3 tons CO2 y . designing measurements of ancil- scribe our concept to derive According to the NGO Mon- The regeneration potential is lary data concerning soil carbon data that will improve the data gabay which provides detailed almost unknown. Only one study dynamics and regeneration po- base to answer the other two information about deforestation by Chidumayo (2002) has re- tential of vegetation affected by questions. and other parameters of Zambia ported about the carbon that is logging for charcoal production. on its website (http:// accumulated in the aboveground Three questions are crucial: Experimental design at rainforests.mongabay.com/ biomass within the first ten years 1.) How much carbon from the Kataba Forest deforestation/2000/Zambia.htm), of regrowth after a clearcut. On standing biomass and from soil is In September 2007 the tower about 445,000 ha y-1 of wood- average it was 28 % of the bio- lost? which was erected during the land were deforested or de- mass of undisturbed areas. 2.) How big is the sink that is SAFARI 2000 campaign was graded between 1990 and 2000, The high uncertainty in carbon lost? equipped with an eddy- thereafter the rate has even fluxes related to deforestation, 3.) How big is the regeneration covariance system (open as well increased. Each hectare of defor- forest degradation and regenera- potential? as closed path for comparisons), estation means a loss between tion in Miombo woodlands has In this contribution we will give a a CO2 profile system and micro- 21700 (FAO) and >40000 kg been another motivation for our preliminary answer on the sec- climate that are pow- (Chidumayo 2002) of woody research in Zambia. It played an ond question based on a first ered by a photovoltaic system. biomass. This increases the Zam- important role in our experi- year of eddy covariance meas- The construction work (Figure 4

cont. on page 14 Page 14

Flux-measurements in a Miombo woodland cont. from page 13

sinks. We can conclude from our pre- liminary results that deforesta- tion and forest degradation in the Miombo woodlands do not result in the loss of a carbon sink. However, the carbon pools that are lost from the vegetation and the soil are enormous when the actual rates of deforestation are taken into account. Accord- ing to a recent publication in the Biogeosciences Special Issue on carbon fluxes in Sub-Saharian Africa Canadell et al. (2009) showed that deforestation ac- counts for about 95% of Zam- bian carbon emissions. From our point of view the crucial ques- tion is: How is the regeneration potential of the woodlands Figure 4: Field work during tower instrumentation in September 2007. Left: Waldemar Ziegler as field blacksmith, centre top: Lutz Merbold working at the tower, centre bottom: Mukufute Mukelabai and Werner Kutsch at the control unit of the system, right: Olaf Kolle working at the affected? If full re-growth is pos- solar power station. (Fotos by Lutz Merbold and Werner Kutsch) sible (Chidumayo 2004 observed shows some impressions) lasted the next years. Forest in an example for an high resilience) and charcoal can about 4 weeks. The tower is undisturbed old-growth forest be produced in a sustainable located in the North-West cor- First results that is carbon neutral which way, it might be a way to avoid ner of the forest reserve. The Since the first rains usually occur might contradict recent opinion fossil fuel burning. If re-growth is main footprint area lies in the in October, we decided to set based on mostly Northern hemi- impossible due to the loss of soil almost undisturbed area within the beginning of the sphere old-growth forests that fertility and seed bank decline the reserve, because East and ‘meteorological year’ at Mongu were calculated to be carbon Miombo woodlands should be- South-East are the prevailing on September 1st. The finger- directions (Figure 5). print (Figure 6) shows how In addition, we designed four strongly the system is driven by inventory plots that are located water availability: The first rains along a disturbance gradient in October resulted in an in- Figure 2: LIDAR-derived image of vegetation cover at the TCEF. from outside to the centre of crease in nighttime respiration, the reserve. In these plots we followed by a strong uptake conduct inventory studies to get rates from November on. Dur- ancillary data on above and ing the peak growing season belowground carbon pools as maximum net uptake rates were -2 -1 well as soil carbon dynamics and around -20 µmol CO2 m s . tree growth. Since the area west They slowly decreased after the of the tower was completely rains stopped in April. clear-felled during 2008 we de- Annual NEE for the first year of cided to place another inventory our measurements between plot in that area, too. Thus, September 2007 and August future projects will be able to 2008 resulted in an uptake of follow the regeneration during about 5 g C m-2 y-1. Thus, Kataba Figure 5: Wind distribution during the first year and location of the inventory plots. The white triangle symbolizes the 50 % fetch area for the sector 90 – 120 °, which is the main wind direction. cont. on page 15 Page 15

Flux-measurements in a Miombo woodland cont. from page 14

FluxLetter The Newsletter of FLUXNET

Vol.2 No.1 March, 2009

FluxLetter is produced quarterly at the FLUXNET Office with support from the National Science Foundation.

Figure 6: Fingerprint of the gap-filled EC fluxes between 1.9.2007 and 31.8.2008 and the relative plant available water for the upper 100 cm of the soil. come part of actions like REDD and studies about ecosystem Scanlon, T. M. and Albertson, J. D.: Canopy scale measurements of CO (Reduction of Emissions from services that are provided by 2 and water vapour exchange along a Deforestation and forest Degra- Miombo woodlands. Unfortu- precipitation gradient in southern dation). The REDD mechanism is nately, CarboAfrica is coming to Africa, Glob. Change Biol., 10, 329–341, 2004. This issue of FluxLetter was under negotiation for the post- an end in September 2009 and edited, designed and produced by:

Kyoto commitment periods. The we will have to stop our work at basic idea is to pay compensa- Mongu. contact: Werner L. Kutsch Rodrigo Vargas tion for not using the timber and Dennis Baldocchi [email protected] protecting intact forest ecosys- Further Reading tems. The crucial parameter will Chidumayo, E. N.: Changes in miombo Lutz Merbold FLUXNET Office, 137 Mulford woodland structure under different land Max-Planck-Institute for Biogeo- Hall, University of California, be whether the compensation tenure and use systems in central Zam- Berkeley, CA 94720 will be high enough to generate bia, J. Biogeogr., 29, 1619–1626, 2002. , Jena, Germany ph: 1-(510)-642-2421 Fax: 1-(510)-643-5098 alternative labour for people that produce charcoal and alternative Chidumayo, E. N.: Development of Mukufute Mukelabai Brachystegia-Julbernardia woodland after Maurice Muchinda We plan to make the FLUXNET energy for people that use char- clear-felling in central Zambia: Evidence newsletter a powerful information, coal. Thus, further studies are for high resilience. Applied Vegetation Zambia Meteorological Depart- networking, and communication Science. 7:237-242, 2004 necessary and they have to in- ment, Lusaka, Zambia resource for the community. If you want to contribute to any section or clude socio-economic aspects of propose a new one please contact the Canadell, J. G. , Raupach, the charcoal production, techni- M. R., Houghton R. A.: Anthropogenic FLUXNET Office. THANKS!! CO emissions in Africa Biogeo- cal as well as income alternatives 2 sciences, 6, 463-468, 2009