Sessions and Hypotheses

The goal of our Chapman Conference is for attendees to create new partnerships and then to use these collaborations to make new scientific discoveries in the Congo.  Conference attendees will review the current state of knowledge and identify where key information is lacking.  At the conference, they will make plans for collecting new measurements, for digitizing historic data, for making new models, and so on.  The key is that these plans will be geared toward making new discoveries.

The focal point of each session is a hypothesis.  Each hypothesis is a guideline for future research.  Ideally, speakers and poster presenters will provide their thoughts on the hypothesis by indicating whether they think it is true or false.  However, presenters are welcome to show their research on any issue related to the session topic.  We encourage presenters to suggest potential discoveries that might yet be made as well as the honest limitations of working in the Congo Basin.  We expect that some of the hypotheses will be answered during the Chapman conference; some might be further honed so that future research can be properly directed; and some might be waitlisted for a time when greater amounts of measurements are available.  In this manner, we will provide funding agencies with a roadmap they can use at their discretion.  We expect that funding agencies, including U.S. and especially foundations from around the world, will participate in our Chapman Conference.

The meeting will occur in Washington D.C. over three full days with both oral and poster sessions devoted to the hypotheses.  Key presenters will include Congolese researchers who have been studying their basin for decades as well as international researchers known for their remote sensing, modeling, and especially for their scientific intellect.


Basic Daily Schedule

Tuesday September 25

Climate & Precipitation

Wednesday Sept. 26

Rivers & Wetlands

Thursday Sept. 27

Water Resources

Morning Welcome

Rainfall & Runoff

Cuvette Centrale Deforestation & Hydrology
Lunch At AGU At AGU At AGU
Afternoon Tropical Rainbelt Carbon & Hydrology Water Diversions


Early Evening Open Open

Note: Each session is described in detail further below the brief descriptions.

The Program Committee is in the process of inviting hosts and speakers for each of these sessions.  We anticipate two hosts and three speakers per session.  To ensure discussions amongst participants, posters are the preferred scientific presentation for attendees.  We will soon open the abstract submission tool.  You are welcome to submit more than one abstract and hence present more than one poster.

Tuesday September 25: Daily Theme is Climate & Precipitation

Welcome and Opening of Meeting by Doug Alsdorf

Tuesday Morning: Session Topics is Rainfall & Runoff
Hosted by: Ed Beighley, Cyriaque-Ruffin Nguimalet

Invited Speakers: Alain Laraque, Christopher Ndehedehe, Jean-Marie Tshitenge

Hypothesis: Despite known variations in the discharges of the Congo and Oubangui Rivers, previous rainfall amounts have varied comparatively less across the basin.
We seek presentations that discuss any issue related to discharge and rainfall variations over time. 

Tuesday Afternoon: Session Topics is the Tropical Rainbelt
Hosted by:  Gil Mahe, George-Noel Longandio

Invited Speakers: Sharon Nicholson, Richard Washington, Wilfried Pokam

Hypothesis: The Congo Basin will experience significant changes both in rainfall amounts and in geographic locations from climate change, but these changes will not be uniform.

This session includes presentations that discuss issues regarding how rainfall happens in the Congo Basin.  Historically, the ITCZ has been the preferred vocabulary, but this is changing to focus on tropical rainbelts.  The rainbelts are more complex than the simple notion of the north-south moving ITCZ.

Wednesday September 26: Daily Theme is Rivers & Wetlands

Wednesday Morning: Session Topic is the Cuvette Centrale
Hosted by:  Fiachra O’Loughlin, Jose Wabakanghanzi

Invited Speakers: Paul Bates, Hyongki Lee, Stephane Calmant

Hypotheses: (1) The water in the Cuvette Centrale Is supplied mostly by rainfall.  (2) The water empties from the Cuvette Centrale mostly by ET.
This session welcomes any presentation that discusses the hydrodynamics or water balance of the Cuvette Centrale.

Wednesday Afternoon: Session Topic is Carbon & Hydrology
Hosted by: Bienvenue Dinga, Pascal Boeckx

Invited Speakers: Guy Moukandi, Travis Drake and Rob Spencer

Hypothesis: The annual-average amount of CO2 and CH4 evasion from all Congo Basin waters is more than 480 Tg C/yr, i.e., more than a value comparable to that of the Amazon per unit area.

This session welcomes presentations that measure or model the carbon cycle as it relates to hydrology in any of the Congo rivers or wetlands.

Thursday September 27: Daily Theme is Water Resources

Thursday Morning: Session Topic is Deforestation & Hydrology
Hosted by:  Guy Schumann, Molly Brown

Invited Speakers: Matt Hansen

Hypothesis: Deforestation of 30% of the headwater sub-basins will significantly increase headwater flows and hence increase downstream discharges.

This session is open to any presentation that discusses land cover and land use in the Congo Basin.  Ideally, these presentations would also suggest the connections of these changes as they relate to hydrology, rainfall, or carbon balance.

Thursday Afternoon: Session Topic is Water Diversions
Hosted by: Rapheal Tsimanga, Mark Trigg

Invited Speakers: TBA

Hypotheses: (1) Future hydroelectric power generation will not impact waters flowing in rivers.  (2) The suggested diversion of Oubangui River water to Lake Chad would be a significant impact.

This session is open to presentations that discuss any issue related to water resources and water usage in the Congo Basin.

Thursday Afternoon: Summary and Closing Remarks by Doug Alsdorf

Languages of the Conference

English is the preferred language of the conference.  Ideally, all oral and poster presenters will use English, however some presenters who only converse in French are welcome to provide their poster in French.  Professional French-English translation and interpretation services will be provided during the conference.  The goal is to have all talks presented in English but with interpretation into French available via headsets.  In addition to interpreters, a few bilingual scientists will also be available to assist with small group discussions.

Details of Each Half-Day Session

Each day has an overall theme.  Both sessions for that day have their own topics that fit within the daily theme.  Presenters can use the hypothesis for each topic to guide them regarding the content of their poster or talk.  However, any presentation related to the session topic is welcome.

Each half-day session will consist of three parts: (1) invited oral presentations, (2) poster presentations, and (3) break-out discussion meetings.  All oral and poster presenters are encouraged to focus on the hypothesis that guides their session.  Ideally, they would provide their own positive or negative results from having tested the hypothesis.  Furthermore, we welcome any research related to the general topic of the session as well as presentations that suggest a new hypothesis related to the session topic.

Each session will begin with three invited 20 minute talks.  The three speakers will be encouraged to provide their thoughts on the session’s hypothesis as well as any other research they have conducted related to the general topic of the session.  Attendees will be encouraged to ask questions at the end of each talk.  Thus, the oral portion of each session will last a little more than an hour.  Next, the poster portion will allow attendees plenty of opportunity to discuss one-on-one the specifics of the hypothesis as well as the overall session topic.  Ideally, they will create new research collaborations.  The poster portion of each session will last a little more than an hour.  In the forty-minute break-out meetings of each session, discussion leaders will guide the conversation so that decisions are made regarding each hypothesis.  Attendees will determine if the hypothesis is true or false.  If a decision cannot be made, then everyone will contribute to an understanding of what needs to be done in order to test the existing hypothesis or to create a new hypothesis.  The goal is to provide everyone, including funding agencies, with a robust set of research ideas.

Tuesday September 25, Morning: Session Topic is Rainfall & Runoff

Despite known variations in the discharges of the Congo and Oubangui rivers, previous rainfall amounts have varied comparatively less across the basin. 

If this hypothesis is true, then either plant physiology or subsurface lithologies account for the differences between rainfall and the expected river runoff.  If this hypothesis is false, then important rainfall variations have occurred in the past but have gone unmeasured in key regions of the basin.

For example, Matsuyama et al. [1994] used stream and precipitation gauge measurements combined with vapor flux convergence data from ECMWF and with normalized difference vegetation index (NDVI) mappings to find that seasonal variations in the basin water balance are largely related to characteristics of the deciduous forest located in the southern half of the basin and that NDVI and ET are in-phase with seasonal precipitation over the evergreen forests of the northern half of the basin. Regarding lithologies, Laraque et al. [1998] found that there is little seasonal variation in the discharges of rivers that drain the Batékés plateau despite monthly rainfall variations ranging from greater than 250mm to less than 50mm.  They indicate that a 200m to 400m thick sandstone aquifer underlies the plateau and that river flows are “highly diluted clear waters, relatively rich in dissolved silica” and thus suggest that the aquifer regulates the plateau river discharges leaving them independent of rainfalls. These examples imply that rainfall-runoff ratios may vary significantly throughout the Congo Basin. Nguimalet and Orange [2015] suggests that rainfall-runoff ratios over the Oubangui Basin do not vary in time. They studied rainfall data from 1935 to 2015 over the basin and found a 5% decrease in rainfall after 1970 and that Oubangui River discharge during this same period has also remained lower than its long-term average with a most recent decline of 22% since 1983.

To test this hypothesis, presentations during this session might discuss the 100 year long discharge record of the mainstem Congo and the 70 year long record of Oubangui discharge.  Others may present rain gauge data from the first half of the 1900s for comparison to more recent precipitation records. Presentations on basin-wide water and energy balance modeling parameterized with the most up to date vegetation and soils maps might demonstrate the bounds on the water volumes governed by ET, plants, and root zones.  Then, locations where the model results have significant errors resulting from inadequate measurements might be further discussed and perhaps targeted for field studies such as evaporative pans, flux tower measurements, water levels, stream discharges, etc.

Tuesday September 25, Afternoon: Session Topic is Tropical Rainbelt

The Congo Basin will experience significant changes both in rainfall amounts and in geographic locations from climate change, but these changes will not be uniform.

If this hypothesis is true, then agriculture, hydroelectric, transportation, and other water users will need systems designed to meet these multi-decadal fluctuations.  Not only are water users potentially impacted, but so too are ecosystems that are reliant on regular seasonal water pulses. If this hypothesis is false, then planning can be based on today’s water volumes and fluxes while ecosystem functioning remains little changed.

The tropical rainbelt annually moves back and forth across the equator with the seasons, producing a biannual rainfall maximum in the equatorial latitudes.  Most of the rainfall in the Congo Basin itself is associated with large-scale mesoscale convective systems.  These, in turn, are strongly influenced by orographic effects from the highlands surrounding the basin.  The factors governing the interannual variability of rainfall vary across the basin, but throughout the region there are strong influences from prevailing zonal circulation, the equatorial Walker cells, and sea-surface temperatures.  Because of these teleconnections with ocean temperatures, the question arises about the impacts that global warming will have.  Global warming could warm the Northern Hemisphere more than the Southern, or could impact the Atlantic Ocean more than the Indian Ocean.  These changes would potentially result in a geographic shift or a change in intensity of the tropical rainbelt.   Given that the left bank (southern) tributaries have annual flood pulses in March–April and the right bank (northern) tributaries have theirs in October–November, a geographic shift in the tropical rainbelt could modify this geographic distribution with perhaps wetter sub-basins and greater flood pulses in the north and dryer sub-basins and lower flood pulses in the south.

To test this hypothesis, presentations might show changes in river discharge using comparisons over time or comparisons from rivers located across the Congo Basin or even across sub-Saharan Africa.  Or, presenters might show temporal and geographic variations in rainfall amounts.  Models can be used to show the potential impacts from differential hemispheric warming and show if such rainfall shifts are possible.  These models might also show the precipitation amounts and patterns that we might expect in the future.

Note that we use the words “tropical rainbelt” rather than “ITCZ”.  Recent work is showing that rainfall is more complicated than the notion of an ITCZ migrating across the Congo Basin.  Presenters are welcome to address this issue.

Wednesday September 26, Morning: Session Topic is The Cuvette Centrale

(1) The water in the Cuvette Centrale is supplied mostly by rainfall. (2) The water empties from the Cuvette Centrale mostly by ET.

If hypothesis 1 is true, then carbon is produced within the wetland and sedimentation is minimal.  This would imply that the world’s second largest river is disconnected from the immediately adjacent and massive swamps of the Cuvette Centrale.  If false, then the supply is mostly from fluvial sources that bring their own carbon, nutrients, and sediment.  This would imply diffuse overbank flows and related deposits from rivers adjacent to the Cuvette or specific up-gradient pathways connecting upstream rivers to the Cuvette.  If hypothesis 2 is true, then there is little water from the wetland supplied to the adjacent rivers and thus the river sediments, nutrients, and carbon are not altered by the wetlands.  It remains possible for the wetland to receive water from an up-gradient river but to empty mostly by ET, and thus the wetland receives sediment, nutrients, and carbon but supplies none. If false, then the Cuvette is altering the biogeochemistry of down-gradient rivers.

During our Chapman Conference, we will explore the observations and models that test this hypothesis.  For example, conference attendees could present comparisons of water surface elevations in the Cuvette to those of immediately adjacent river water levels and thus show the local hydraulic gradient.  If the gradient throughout the entire water year is consistently downslope from the wetland to the river, then it’s possible that the wetland always empties via surface water paths and thus hypothesis 1 is true while 2 is false.  For such a false result, spatial mapping would need to demonstrate that the sizes of the surface flow pathways are large enough to accommodate the expected water volume emptied from the Cuvette wetlands.  Modelers might demonstrate that sufficient rainfall is available to supply all of the Cuvette waters.  Field studies could compare the timing and volume of rain events with water level changes in the swamps.  Water and energy balance modeling should be able to provide upper and lower bounds on the volume of water that is delivered by P or removed by ET.  For example, if modeling demonstrated that ET could remove 80% of one season of wetland storage change, then surface pathways should be evident to remove the remaining 20%.  This approach assumes a long-term wetland ΔS of zero and that groundwater is not sufficiently fast enough to remove the 20%.

Wednesday September 26, Afternoon:  Session Topic is Carbon and Hydrology

The annual-average amount of CO2 and CH4 evasion from all Congo Basin waters is more than 480 Tg C/yr, i.e., more than a value comparable to that of the Amazon per unit area.

If this hypothesis is true, then the Congo Basin plays a central role in global carbon dynamics, despite being less understood than the Amazon Basin. For example, Raymond et al. [2013] estimate the global carbon evasion rate from rivers, lakes, and reservoirs (but not from wetlands) at 2.1 Pg C/yr.  The Borges et al. [2015b] estimate of flux of CH4 and CO2 at 480 ± 80 Tg CO2 equivalents/yr from the Cuvette Centrale highlights the importance of the Congo Basin in the global carbon cycle (i.e., Congo wetlands alone could add significantly to the Raymond et al. value). If this hypothesis is false, then the Congo Basin can perhaps be understood by extrapolation of evasion studies from other tropical basins.

Like the previous sessions, presentations that demonstrate testing of this hypothesis might include field-based measurements or model-based results.  These presentations might test this hypothesis by showing field campaign measurements of the amounts of CO2 and CH4 evading from the various types of water surfaces in the Congo, including the mainstem River, major tributaries, headwaters, and the wetlands as well as direct measurements of gas transfer velocities within the Congo Basin.  Model results could predict the expected amounts of carbon for specific locations as based on the hydrodynamics of inundated areas, where temporal changes in water levels and flooded extents are known to impact carbon amounts.

Thursday September 27, Morning:  Session Topic is Deforestation and Hydrology

Deforestation of 30% of the headwater sub-basins will significantly increase headwater flows and hence increase downstream discharges.

If this hypothesis is true, then annual flood pulses in Congo Basin tributaries will increase in magnitude resulting in possible ecological changes such as modifications of fisheries habitats. If this hypothesis is false, then the loss of canopy and presumably its replacement by smaller vegetation does not impact the local water balance sufficiently to alter fluvial processes.

This hypothesis is based on the concept that deforestation decreases the amount of ET resulting in more of a given amount of rainfall becoming runoff, compared to the case of no deforestation. The 30% area value is chosen because currently around 30% of central Africa’s forests are under logging concessions [Laporte et al., 2007] but could instead be a sliding scale or tied to a set of economic and demographic drivers. For example, drivers of Congo deforestation include slash-and-burn clearing for agriculture, unregulated logging, and the collection of wood for fuel. These have become evident in the central and eastern portions of the basin since 1990 [Mayaux et al., 2013]. Presentations in this session that test this hypothesis could compare stream discharge data collected from the deforested regions before and after the advent of logging.  Assuming that local deforestation is the major driver of local hydrologic change during this time interval, gauge data might show recent increases in discharge and thus prove the hypothesis. This approach likely will require long-term gauge data from nearby, undisturbed basins in order to ensure that hydrological changes are specific to deforestation.  Isolating the deforestation signal might not be this straightforward.  In which case, presentations on hydrological modeling, first developed for deforestation studies in other tropical forests, will also be needed.

Thursday September 27, Afternoon:  Session Topic is Water Diversions

(1) Future hydroelectric power generation will not impact waters flowing in rivers.  (2) The suggested diversion of Oubangui River water to Lake Chad would be a significant impact.

If hypothesis 1 is true, then any environmental impacts from hydroelectric generation will not result from changes in river discharge. If false, then the societal improvements from having electrical power should be balanced against the challenges from having artificially changing river water flows and hence related changes in the channel depths.  If hypothesis 2 is true, then river discharge will be decreased in the largest right bank tributary of the Congo and thus potentially impact navigation by lowering water levels or altering sediment transport. If false, then water diversions to Lake Chad are sufficiently small.  The impact on the Cuvette Centrale will depend on its hypothesis testing (see Wednesday’s sessions).

Past and present hydroelectric facilities in the Congo Basin are usually located at cataracts, where run-of-the-river methods with no pondage are often used to turn turbines (e.g., RandGold’s efforts on the Nzoro River), rather than using methods that impound water and create an artificial pressure head with periodic storage releases. There are potentially two methods of testing these hypotheses; one uses measurements and the other relies on modeling. River discharge data collected before and after the installation of a hydroelectric facility might demonstrate that river flows have not been altered in terms of amplitude or timing; and thus, hypothesis 1 would be proven true (at least for that facility and perhaps for similar styles of installations). More generally, hydrologic models with reservoir and run-of-the river hydraulics could be used to demonstrate the potential for impacts.  These could also be used to show the impacts from siphoning water from the Oubangui River.  Additionally, presentations might also consider the sediment transport loads of the rivers.  In some cases the sediment loads might be altered by hydroelectric generating facilities or by the Oubangui-Chad diversion.  Sediment transport is known to alter the bathymetric profile of rivers and thus may have a sufficient volume to be an important consideration for downstream facilities and navigation.