Transcript WHY SET UP RIVER FLUXES AT GLOBAL SCALES

WHY SET UP RIVER FLUXES AT
GLOBAL SCALES ?
QUESTIONS
 Global river weathering rates
PRECURSORS
Clarke 1924, Alekin 1950s,
Livingstone 1963
Origins of Sedimentary rocks
Garrels and Mackenzie, 1971
Biogeochemical cycles: carbon,
Garrels, Mackenzie, Hunt, 1973
Nitrogen, Sulfur, silica
 Global denudation
Fourier 1960, Janssen and Painter,
1974
Coastal geomorphology/Sedimentology
Milliman 70’s
Pollutants Inputs to oceans
Goldberg 70’s
 Earth System and Global change
IGPB 80’s
CONTINENTAL AQUATIC SYSTEMS AT THE ANTHROPOCENE
WATER CH 4 CO2
CH4 CO2 NO 2
1
ATMOSPHERE
2
6
ANTHROPOSPHERE
18
3, 5
4
WATER DUST
CH4 CO2 N 2O
22
7, 12
TERRESTRIAL VEGETATION, SOIL
CO2
8
11
CH4 N2O
5
22
R
I
V
WETLANDS
9
E
EROSION
R
19 WEATHERING S
7
GROUNDWATERS
10
CH4 CO2 WATER
WATER
COASTAL
ZONE
O
P
E
N
SEDIME NTS
21
CARBON
NUTRIENT S
O
C
E
A
N
20
RESERVOIRS
13
15
16
LAKES
14
SILTING
CONTINENTAL SEDIMENTS
17
ENERGY
COASTAL
SEDIMENT
UPLIFT
SURFICIAL LITHOSPHERE
additional fluxes or reservoirs due to human activitie
s
OCEAN
SEDIMENT
Combining lithology and relief typology
Classification of 15 relief patterns at global scale combining a
relief roughness indicator and mean altitude at 30’ resolution,
re-aggregated into 7 relief super-classes
Relief Typology (Meybeck et al., 2001)
GLOBAL MAPPING OF RIVER FLUXES
OCCURRENCE OF CONTINENTAL AREA EXPOSED TO
WATER WEATHERING (RHEIC REALM) AND TO RIVER
FLUXES TO THE OCEANS (EXORHEIC REALM)
Total land
149 Mkm2
A
% land
area
B
% weath.
flux
C
% flux to
ocean
10.7
0.1 ?
0.1 ?
33.7
0
0
Endorheic
3.4
1
0
Exorheic
21
7
7
Endorheic
1.1
2
0
Exorheic
27.3
53
55
Exorheic
2.75
37
38
Glaciated
16 Mkm2
Endorheic
Arheic
50,2 Mkm2
Non
glaciated
133 Mkm2
Exorheic
Oligorheic
Rheic
82,8 Mkm2
Mesorheic
Hyperheic
• 10.7% of the present land is glaciated
• 33.7% is arheic (less than 3 mm/y runoff)
• 90% of weathering fluxes is related to 30% of the continents area
Global figures
Organisation of the continental surfaces by water
into major units
Total area 133 M km2
Exo(%)
25,7
Endo(%) Σ
9,0
34,8
Arheic
60,1
5,2
65,2
85,8
14,2
100%
Rheic
Σ
River network
River network : Vörösmarty et al. 2000 a & b, modified and adapted
- The arheic areas are below 3 mm/yr annual runoff
- Due to uncertainty on the water balance ‘arheic’ areas may occur in non-desertic
regions, as NE Siberia, Mackenzie basin, Missouri basin, Patagonia etc. ...
EXORHEIC
ENDORHEIC
940 km3 y-1
37 200 km3 y-1
MOUNTAINS
PLATEAUX
PLAINS
PLATEAUX
HILLS
HILLS PLAINS
Area
[M km2]
62.7
11.1
14.3
27.6
5.75
2.5
0.4
8.7
Runoff
[mm y-1]
293
445
153
424
86
37.5
102
35
Population
density
[p km2]
46.5
67
26.5
46
35
11
36
16
(after Meybeck et al. 2000).
Global average river runoff and population density for major relief classes
HOLOCENE/ANTHROPOCENE EVOLUTION
FLUVIAL NETWORK (POTENTIAL) AT THE LATE GLACIAL MAXIMUM
(18 000 BP)
H. Dürr,Sisyphe
• Most of North America and of Europe was glaciated
• Due to lower sea level (-120m) an extended area of continental platform was
exposed to river transport (e.g. West Siberia and Bering sea, South China and
Aragura seas, Patagonia)
CHEMICAL FLUXES CONTROLS
BASALT OUTCROPS FOR SIX CLASSES OF RIVER RUNOFF
Runoff : UNH
Litho : H. Dürr, Sisyphe
• Basalts of various ages are found at various positions on all continents
• They are exposed to temperature ranging from -10°C to +30°C
• Hower the main weathering control is probably the river runoff ranging over 2
orders of magnitude, even at the regional scale (e.g. african rift valley and Deccan)
PRISTINE RIVER CHEMISTRY
VARIABILITY OF NATURAL RIVER CHEMISTRY
AND LITHOLOGY
Sum of cations (µeq/L)
Dominant ions
Example
50
Ca2+, Cl-
Rio Negro * (Amazonia) quartz sands
70
Na+, HCO3-
Rio Tefe * (Amazonia) quartz sands
500
Mg2+, Ca2+, HCO3-
Basaltic basins
600
Mg2+, HCO3-
Peridotite basins
4 000
Ca2+, HCO3-
Carbonated basins
5 000
Mg2+, SO42-
Coal schists
9 000
Na+, SO42-
Semliki R. Rift Valley
20 000
Na+, SO42-
Bituminous Shale (Montana)
50 000
Na+, Cl-
Urubamba tributary (Amazonia)
* Rain and vegetation control
There is no mean river water that can be used as a global or
even regional reference
PRISTINE RIVER CHEMISTRY
GLOBAL OCCURENCE (% of area) OF WATER TYPES AND
THEIR ORIGINS (Pristine rivers model)
Origin
Type
% Total
Rock dominated
Rain
Evaporated
dominated Sili cate Carbonate Pyrite Evaporites

Na2SO4
3,2
NaCl
6,8
Na2CO3
3,6

MgCO3
2,4

MgSO4
2,0
MgCl2
0,1
CaSO4
5,2
CaCO3
76,9
Total
100











2,6


 
35,4
45,1



5,2
3,6
8,2
• Ionic types in pristine rivers are more diverse than originally thought by
Gibbs (1972)
• CaCo3 is dominating in 77% of rivers (area weighted)
• Rain and vegetation recycling is dominating over 2.6% of the continents
area and the evaporation over 8.2% (rheic realm only, runoff > 3 mm)
• Rock weathering control extends over 89% of the continents area
• Evaporated waters may result in many chemical types
PRISTINE RIVER CHEMISTRY
PRISRI : GLOBAL DISTRIBUTION OF DIC
MEDIUM-SIZED BASINS
3 500 - 200 000 km2, rheic basins (n = 480)
%
DIC CONCENTRATION
% HCO3- / -
DIC EXPORT
RARE
99,5
99
UNCOMMON
90
COMMON
75
VERY COMMON
50
25
COMMON
10
UNCOMMON
1
0,5
0,1
RARE
1
10
DIC mg/L
10
50
100
In 50% of basins
HCO3- exceed 80% of
anions
0,5
1
10
DIC concentration ranges
over 2 orders of
magnitude
50
g C.m-2.y-1
100
DIC export ranges over
3 orders of magnitude
Natural Space Variability of River Suspended Fluxes
A.
TSS Weighted means mg/L
Very low
Low
Medium
High
Very high
Extremely
high
5 - 20
20 - 100
100 - 500
500 2 000
2 000 10 000
> 10 000
Sacramento
Red
Deer (Alb)
Annapolis
(NS)
St.
Lawrence
(CA)
Stikine (BC)
Colorado
Eel (CA)
Matanuska
(AL)
Mississippi
Rio
Grande (TX)
(AZ)
Little
Colorado
(AZ)
Control factors : relief, lakes, lithology, runoff.
Ref : Meybeck, Laroche, Dürr, Syvitski, 2003. Glob.Planet.Change
Natural Space Variability of River Suspended Fluxes
B.
Daily TSS Yields kg.km-2.d-1
Very low
Low
Medium
High
Very high
Extremely
high
< 10
10 - 50
50 - 200
200 1 000
1 000 5 000
> 5 000
Chaudière
Lacustrine
Rhone (CH)
St.
Lawrence
(PQ)
Sacramento
(CA)
Colorado
Mississippi
(AZ)
Eel (CA)
Alpine
Rhine (CH)
Control factors : relief, lakes, runoff, river regime, lithology.
Ref : Meybeck, Laroche, Dürr, Syvitski, 2003. Glob.Planet.Change
Natural Space Variability of River Suspended Fluxes
C.
% of time needed to carry 50% of TSS flux
Very long
Long
Medium
Short
Very short
Extremely
short
> 16%
16 - 8%
8 - 3,4%
3,4 - 1,4%
1,4 - 0,4%
< 0,4%
Sacramento
Red
Deer (Alb)
Eel (CA)
WallaWalla (OR)
Mississippi
St.
Lawrence
Fraser (BC)
(CA)
Stikine (BC)
Control factors : basin size, river regime.
Ref : Meybeck, Laroche, Dürr, Syvitski, 2003. Glob.Planet.Change
GLOBAL BUDGETS ARE REACHING THEIR LIMITS
EXAMPLE : DISSOLVED SILICA
Global average (mg/L)
Approach
Clarke, 1924
8,3
Few, big temperate rivers
Livingstone, 1963
13,1
d.o.
Meybeck, 1979
10,4
Biomes typology, 60
rivers, Amazon included
Probst, 1992
8,9
Multiregression
(Meybeck’s data)
Meybeck and Ragu, 1996
7,7
250 rivers no typology
Meybeck, 1999 (unpubl.)
9,2
d.o. + 9 morphotectonic
tpes (lytho. Control)
Treguet et al., 1995
9,0
(Meybeck + Ragu data)
Meybeck (unpubl.)
8,75
43 pristine rivers and tribs
(exorheic + endorheic)
GLOBAL MAPPING OF RIVER FLUXES
GLOBAL SEDIMENT YIELD MAP
Sediment yields reflect land erosion : they are maximum in South East Asia
where heavy rainfall, active tectonics and erodible rocks are found
Ludwig et al, 1998
PRISTINE RIVER CHEMISTRY
GLOBAL DISTRIBUTION OF RIVERS RANKED PER
INCREASING TOTAL DISSOLVED SOLIDS (∑+ cation sum)
(pristine rivers model, n = 1 329 basins)
30
25
% AREA
20
• The most commonly found ionic
content (∑+) at the continents
surface is between 375 and
6000 µeq/L
• The most commonly found ionic
content (∑+) in one liter of water is
between 375 and 1500 µeq/L
• The ionic fluxes are essentially
related to medium-mineralized
waters, between 750 and
6000 µeq/L
15
10
5
0
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
30
25
% WATER
FLUXES
20
15
10
5
0
30
25
% I ONIC
FLUXES
20
15
10
5
0
+
( µeq/ L)
GLOBAL SYNDROMES OF RIVERINE CHANGES
GLOBAL SYNDROMES OF RIVERINE CHANGES
• Flow regulation
• River course fragmentation
• River bed silting
• Neoarheism
• Salinization
• Chemical contamination
asphixiation, inorganic contamination, xenobiotics occurence
• Acidification
• Eutrophication
• (Microbial contamination)
• (Aquatic species introduction & invasion)
SOME GLOBAL CHANGES AFFECTING RIVER FLUXES
 2,54 Mkm2 of irrigated land (in dry and semi arid and arid
regions)
 More than 5 % of global river runoff decrease (> 2000 km3/y)
Hundred of thousands of small to giant reservoirs
Total reservoir area >0,5 M km2 (Great Lakes + Caspian).
NEOARHEISM
RIVER FLUXES TRENDS AFTER DAMMING THE
COLORADO EXAMPLE (1910-1960)
A : annual water flow
B : annual sediment flux
• Colorado changes are some of the
most dramatic change documented
in a river system
• This evolution was triggered by the
construction of the Hoover Dam in
1936
TE17
GLOBAL MAPPING
GLOBAL IMPACT OF LARGE RESERVOIRS :
SEDIMENT TRAPPING EFFICIENCY
• Coastal zone now gets 30% less sediment
• 700% increase in water held in rivers
• Tripling of river runoff travel times
Sediment starving is a growing issue in some coastal zone
UNH
Vörösmarty et al. 2003
Global nitrogen fluxes through rivers : preindustrial vs contemporary
• The global N fluxes (tot N) have increased more than 3 times
• Regionally the fluxes have increased more than 10 times
• Agriculture and urbanization are the two major N sources
Green et al. 2003
UNH
NUTRIENTS FLUXES HETEROGENEITY
(From GEMS-GLORI analysis)
AREA CLUSTERS
The impacted temperate zone (N. America, Europe, China...)
corresponds to 27,5 % of lobal area but to 52 % of P-PO43- fluxes and
to 6 % of DIN fluxes to oceans
The dry and non- impacted wet tropics plus subarctic regions
corresponds to 50,7% of global area and only to 30% of P-PO43- and
21,3 % of DIN fluxes
FLUXES RANKING
The most polluted rivers that represent only 5 % of global water
discharge would contribute to
32 % of NO348 % of NH4+
54 % of PO43- fluxes
Conclusions
- Coastal basins morphology is highly variable from narrow strips
(Peru-Chile) to very deep basins (Mississippi-Amazon)
- Mean runoff in coastal basins range over 3 orders of magnitude as
for other river fluxes (sediments, carbon, nutrients)
- Population pressure within coastal basins varies over more than 2
orders of magnitude from 0.3 inhab/km2 for the Laptev Sea or
the Gulf of Carpentaria to more than 300 inhab/km2 in
South and East Asia
- Our segmentation allows for determination of first order inputs to
Oceans (e.g. North Atlantic) as well as second order inputs
for about 30 regional seas
- Direct inputs to open oceans are actually limited by extended
mediterranean type seas (Mediterranean proper, Black Sea,
Gulf of Mexico / Caribbean, …) and by other regional seas
(Persian Gulf, Adaman)