Bob McKane, USEPA Bonnie Kwiatkowski & Ed Rastetter, Marine Biological Laboratory Marc Stieglitz & Feifei Pan, Georgia Institute of Technology ~ January 30, 2008, Presentation to NSF Riparian Zone Workshop, Indianapolis, IN ~ H 2 O NO 3 , NH 4 , DON Nassauer Application of an Eco-hydrology Model to Riparian Forest Buffers in Agricultural Landscapes
25
Embed
Bob McKane, USEPA Bonnie Kwiatkowski & Ed Rastetter, Marine Biological Laboratory Marc Stieglitz & Feifei Pan, Georgia Institute of Technology ~ January.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Bob McKane, USEPABonnie Kwiatkowski & Ed Rastetter, Marine Biological Laboratory
Marc Stieglitz & Feifei Pan, Georgia Institute of Technology
~ January 30, 2008, Presentation to NSF Riparian Zone Workshop, Indianapolis, IN ~
H 2O
NO 3, N
H 4, D
ON
Nassauer
Application of an Eco-hydrology Model to Riparian Forest Buffers in Agricultural Landscapes
Outline Describe GT-MEL, a spatially-distributed
eco-hydrology model
Demonstrate GT-MEL for a generalized agricultural-riparian system BMPs
Uncertainties & Challenges:
• Controls on C-N-H2O interactions
• Scaling up from reach to watershed
Georgia Tech (GT) Hydrology Model Spatially Distributed Hydrologic Processes
snobear.colorado.edu/IntroHydro/hydro.gif
GT is relatively simple3 “free” parameters vs. dozens for some hydrology models (e.g., HSPF)
434
3323
22212
s1111
QDdt
ds
QDDdt
ds
QETDDdt
ds
QQETDPdt
ds
PET1Qs
D1ET2
D2
Q1
Q2
Q3
s1
s2
s3
Bedrock
Q4
D3
s4
S = storage S = storage P = precipitation P = precipitation D = drainage (infiltration)D = drainage (infiltration)Q = runoffQ = runoffET = evapotranspiration ET = evapotranspiration
Pan, Stieglitz & McKane in prep
Logistic Curves For Drainage & RunoffLogistic Curves For Drainage & Runoff
Water Filled Pore Space (WFPS)0
f(x)
1
fc = soil field capacity, As = fraction of saturation area
Drainage
DDmax fD (s /sm )
0
Dmax
WFPSfc
D
0
Q
Subsurface runoff
QslopeQmax fQ (s /sm )
Qmax
0
1
Surface runoff
Qs P fAs (s /sm )
As
1.0WFPS
fc 1.0WFPSfc 1.0
(WFPS) (WFPS) (WFPS)
0 0 0
Climate Station
GT Applied to GT Applied to HJ Andrews Experimental ForestHJ Andrews Experimental Forest
Western Oregon Cascades
Photo: Al Levno
Multiple soil
types & layers
HJ Andrews Watershed #10, OregonHJ Andrews Watershed #10, Oregon 10-hectare forested catchment, clearcut in 197510-hectare forested catchment, clearcut in 1975
2x3x3 factorial: identify acceptable 2x3x3 factorial: identify acceptable tradeoffs in crop yield and water qualitytradeoffs in crop yield and water quality
NH4 fertilizer
Denitrification
H2O, DIN, DON H2O, DIN, DON
Simulating ten 100 X 100 m hillslope segments
Simulation Matrix:Simulation Matrix: 2 Forest Ages X 3 Buffer Widths X 3 Fertilizer Rates2 Forest Ages X 3 Buffer Widths X 3 Fertilizer Rates
Simulation #
158243
Simulation Matrix:Simulation Matrix: 2 Forest Ages X 3 Buffer Widths X 3 Fertilizer Rates2 Forest Ages X 3 Buffer Widths X 3 Fertilizer Rates
Simulation #
BMPWQ BMPYIELD
Tradeoff: Corn Yield vs. Water QualitySimulations with 100-m mature forest buffer
DIN
Exp
ort
to
Str
eam
(k
g N
ha-1
y-1)
NH4 Fertilizer (kg N ha-1 y-1)
0 50 100 150 200
Corn yield
DIN export
Co
rn Y
ield
(t
DM
ha-1
y-1)
BMP
EPA drinking water standard
Nit
rate
-N (
mg
/L)
Julian Day
Concentration of NO3 Exported to Stream
EPA Drinking Water Standard
No forest buffer, N fert = 200 kg N ha–1 y-1
100-m forest buffer, N fert = 100 kg N ha–1 y-1
NH4 Fertilizer
Does not consider in-stream attenuation of NO3 – Need to link GT-MEL to stream network model
• Where did all the fertilizer N go?Where did all the fertilizer N go?
• What processes were most important What processes were most important for protecting water quality?for protecting water quality?
• BMPs maximized denitrification, the main factor limiting N BMPs maximized denitrification, the main factor limiting N export to streams: export to streams:
• Forest buffer width most importantForest buffer width most important
• Mature forest denitrification highest: Mature forest denitrification highest: 20% more than young forest, 500% more than corn 20% more than young forest, 500% more than corn (more detritus = more denitrification) (more detritus = more denitrification)
• Sequestration of N in forest vegetation & soil unimportantSequestration of N in forest vegetation & soil unimportant
Crop ManagementCrop Management
• Fertilization rates, crop yields had to be reduced to meet Fertilization rates, crop yields had to be reduced to meet water quality standard, even with mature forest bufferwater quality standard, even with mature forest buffer
BMP SUMMARYBMP SUMMARY
Chesapeake
Peterjohn & Correll 1984
Willamette Valley
Next: real-world tests of GT-MEL
Chesapeake Willamette Valley
Regional applications of fine-scale process models are computationally expensive
GT-MEL
Simplified model of first-order watersheds
Statistically summarize model output describing fine-scale processes