Modeling land use effects in Hydrology: Nam Ken

1
Modeling land use effects in Hydrology Nam Ken
Mae Sa

• Hawaii, 8th October 2007

2
Past and new objective

• BEFORE
• Apply DHSVM to basins
• How land use change affects hydrology?
• NOW
• Study how land use change affects hydrology, but
  different approach
• Understanding vegetation dynamics
• Simplifying parameters, based with observations
• Semi-distributed hydrologic modeling
• Based on hillslope partitioning of the landscape
  as hydrologic units

3
Outline

• Recent study on Rubber root water uptake
  (Guardiola et al. 2008)
• Semi-distributed Modeling
• New approach based on Hillslopes discretization
• Bousinesq equation
• Hydrologic Units
• Future work

4
How land use change affects hydrology?
Distributed Model
Root-zone loss function
Fractional Coverage Hemi Fraction Coverage Truck
Space Aerodynamic Attenuation Clumping
Factor LAI Leaf Angle A B Scattering
Parameters Mass Release Drip Ratio Height Maximum
and Minimum Resistance Moisture Threshold Number
of Root Zones Root Zone Depth Overstory /
Understory Root fraction Overstory/Understory LAI
and Albedo
Soil Vegetation parameters
5
Setting Rubber extension in China
Native in the Amazon (10S 10N) T 24 28oC
P 1500 2000 mm Extended to Non-native
environments Marginal growth conditions
Rubber 8 ? 30 Forest 36 ? 24 Grassland 26
? 14
6
Experimental basin
INSTRUMENTS - STATIONS Tea Rubber Radiation,
Wind speed, Temp, RH, P,
Soil moisture Grassland Forest Soil moisture
and P
Monsoon climate (P 1100 to 1700 mm) - Rainy
season May to October - Dry season November to
April
Forest /2ary Forest 53 Paddy
6 Grassland/Bush 2 Swidden 23 Rubber 16
7
Experimental basin
Soil Characteristics Undisturbed soil samples
(SM desorption curves) In situ Ksat
measurements (Borehole permeameter)
8
Root-zone water balance
9
Results Discussion SM trends analysis
Evaporation Fraction -
?(?)
2005 Negligible precipitation
Emax Priestley-Taylor
10
Results Discussion SM trends analysis
E ? at the surface E from deeper layers
?
E ? at the surface E constant deeper layers
Emax Priestley-Taylor
11
(No Transcript)
12
Why Root-water uptake during shedding?
NOV DEC JAN FEB
MAR APR MAY
13
Spring flushing paradox
What triggers shedding/flushing?

• Leaf flushing during the dry season the paradox
  of Asian monsoon forest
• New leaves during hottest driest season
• Trees rely on subsurface water
• Climate is not the primary control of the
  phenology? Increase in DAY LENGTH

Leaf flushing during the dry season the paradox
of Asian monsoon forest

• Leaf flushing during the dry season the paradox
  of Asian monsoon forest
• New leaves during hottest driest season
• Trees rely on subsurface water
• Climate is not the primary control of the
  phenology? Increase in DAY LENGTH
  or INSOLATION

Is rubber a spring flushing tree?
Elliott et al. (2006) Global Ecology and
Biogeography 15, 248-257
14
Results Discussion
15

Results Discussion SM trends analysis
16
Conclusions

• Data analysis approach
• RZ loss function applies to native vegetation.
• Non-native, loss function depends also on
  day-length
• Reduction of water storage observed
• Increase in water demand during water deficit
  periods
• Now the intention is to translate this into a
  model

17
Semi-Distributed Modeling

• Using hillslopes as basic building blocks for
  catchment modeling.
• Subsurface flow controlled by their geometry and
  characteristics such as depth, porosity, Ksat
• Each hillslope has a homogeneous characteristics,
  i.e. vegetation type, soil type, and energy and
  water inputs. (Semi-Distributed model)

18
Semi-Distributed Modeling
Divided in 887 hillslopes
Divided in 815 hillslopes
19
Semi-Distributed Modeling
Boussinesq equation
20
Geomorphometric analysis
Slope 12o
lt 12o is the gradient required for saturation
to occur
Trucker Bras (1998) Water Resources Research
34(10)2751-2764
21
(No Transcript)
22
Semi-Distributed Modeling Namken
23
Semi-Distributed Modeling Mae Sa
24
Semi-Distributed Modeling SHU

• Different Root zone models on top of hsB
• Pervious
• Rubber
• Forest
• Grassland / crops
• Rice Paddies
• Impervious
• Roads
• Green Houses

ROUTING TO THE GAUGE
25
Semi-Distributed Modeling SHU
Pervious Land Uses forest, crops, grassland,
rubber
Transpiration f(?, day-length, temp)
Canopy storage
Soil evap.
Runoff
Infiltration saturation excess (return flow)
L
K(?)
Percolation (N) INPUT FOR hsB
? gt ?fc ? Percolation (N) and/or Runoff
L and Transpiration depends on land use
26
Simple Distributed Modeling SHU
27
Semi-Distributed Modeling SHU

• Rice Paddies (6 in NK)
• Different vegetative phases
• Seedling flooding fields (15 days)
• Vegetative (different length variety)
• Reproductive (25 35 days) complete growth
• Ripening (25 - 35 days) (irrigation stop)
• Each phase
• Ponding depth
• Crop factor (depends of vegetation growth)
• There are 1 to 2 harvest/year (upper and lower
  basin)
• January and April/May, or only April/May

28
Semi-Distributed Modeling SHU

• Flooded P(t) No flooded P0

• When flooded? Saturated soils
• Starting day random
• P(t) Rainfall ? determines irrigation
• Darcy law to calculate percolation
• When dry ? Unsaturated soils
• K(?) using Van Genuchten
• Darcy law to calculate percolation

Irrigation
Precipitation
P(t)
K(?)1
L1
L2
K(?)2
K(?)1 ltlt K(?)2
Percolation (N)
10 days average Kc 1 1/0.87 1.02 1.08 1.06
1.17 1.24 1.28 1.35 1.32 1.29 1.15 0.92 Ponding
200 150 125 100 100 100 100 100 100 100 75 50
0 (mm)
29
Semi-Distributed Modeling SHU

• Impervious
• Very low Ksat, for roads and GH roofs
• Mainly Infiltration excess
• Green Houses ? deeper study about water uses

OL evaporation
L
K(?)
30
Future work

• Finish Rubber root water uptake paper
• Couple the root zone models with hsB
• Route the outflow to compute streamflow
• Paper about Semi-distributed hydrological
  modeling in SEA using hsB and HSU. Impacts on
  land use change

31
Rice Paddy
Modified Penman Monteith
Priestler a 1.26
32
Agriculture
33
Simple Distributed Modeling Hillslopes
34
(No Transcript)
35
(No Transcript)
36
Soil moisture trends analysis
2006 80 mm precipitation
37
Discussion
38

• Rubber drier with depth
• dampening with depth
• delay in response

39
The effects of introducing non-native vegetation
on hydrological partitioning in a tropical
catchment

• Hawaii, 8th October 2007