Modeling Environmental Systems at the Landscape Scale

1
Modeling Environmental Systems at the Landscape
Scale

• Todd Lookingbill
• Landscape Analysis and Quantitative Ecology
• March 2, 2005

2
ObjectiveVegetation Pattern

• Background - Gradient analysis
• Sample vegetation (and environment)

3
ObjectiveVegetation Pattern

• Background - Gradient analysis
• Sample vegetation (and environment)
• Array samples according to species compositional
  trends

(Whittaker 1956)
4
ObjectiveVegetation Pattern

• Background - Gradient analysis
• Sample vegetation (and environment)
• Array samples according to species compositional
  trends
• Relate these trends to environment

(Whittaker 1960)
5
ObjectiveVegetation Pattern

• Background - Gradient analysis
• Virtually always find
• temperature trend (usually as elevation)
• moisture trend (as local terrain)
• (sometimes) soils or parent material
• Environmental proxies to leverage sparse data
  on actual abiotic distributions

6
Objective modeling the patterns (spatial) of
resource and direct environmental gradients
vegetation pattern
temp. moisture fertility
7
Terminology Review

• A model is a description of a system
• System a collection of interrelated objects
  (Haefner 1996)
• Model types
• ___________________
• ___________________
• ___________________
• ___________________

8
vegetation pattern
temp. moisture fertility
9
3 Models

• Conceptual model of nutrient systems
• Empirical statistical model of atmospheric
  systems
• Process simulation model of hydrospheric systems
• Leading towards Modeling environmental-plant
  relationships

10
Soil Development Conceptual Model
(Jenny 1941)
11
Hillslope Scale Soils Model
Catena
coarser less OM less water
shallower on steeper slope (deeper on flat tops
and bottoms)
water table
more clay more OM more water
12
Watershed Scale Soils Model
(Likens Bormann 1995)
13
Small Watershed Approach
Outputs
Inputs
14
The Black Box Model
From where does the additional 11.7 kg/ha/yr Ca2
come? Are there any internal reservoirs that
increase on an annual basis?

• Output consistentlygreater than input
• 1963-1974
• Input Output Net Loss
• 2.2 13.9 -11.7
• kg/ha/yr

(Bormann Likens 1977)
15
Opening up the Black Box
16

• How have the things changed over the past 30
  years?
• Rain is becoming less acidic. Resulting in
  decrease in hydrologic losses of Ca.
• Input Output Net Loss
• 108 684 -576 1963-1974
• 44 351 -307 1992-1993
• eq/ha/yr
• Forest is maturing and acquiring less Ca on an
  annual basis.

17

• Important attributes of HBEF watersheds
• Long-Term Record

18
Temperature

• ___________________
• ___________________
• ___________________
• ___________________
• ___________________

19
Landscape-scale temperature
20
Data are sparse and biased
21
H.J. Andrews Experimental Forest
22
Sample Design

HJA Stratified Approach within Watersheds
Portable Microloggers
23
Temperature ModelRegression Approach

• Begin with regressed lapse rates
• Fine-tune as data allow
• correct for radiation load Taspect
• (Ho maximum)
• correct for distance to stream/water
• (Ho minimum)

24
Radiation ProxyTransformed Aspect
0/360

• Aspect ranges 0-360, with 0 360 degrees
• TAspect -cos(45 aspect)
• SW 1 NE -1 NW SE 0

90
270
180
(Beers et al. 1966)
25
Temperature Regression Models
26
Spatially ImplicitRegression Models
Tmax
Tave ?0 - ?1 Elevation ?2 ln(dstrm) ?3
Radiation ?
Tmin
(Lookingbill Urban 2003)
27
Moisture
28
Catchment Hydrology Conceptual Model
Inputs/outputs
Watershed Hillslopes Patches Gridcells
?Storage Inputs Outputs
Regulators?
29
Soil Water Regulators
(Landscape Scale)

• ___________________
• ___________________
• ___________________
• ___________________

30
Potential Soil Water Regulators
31
Drainage Indices

• Local curvature indices (convex, concave)
• Convexity rate of change in slope
• profile curvature (down slope)
• plan curvature (across slope)
• inverse concavity
• units degrees per unit distance

32
Profile curvature
convex
concave
Plan curvature (contour line)
33
Drainage Indices

• TCI ln (A / tanß)
• A upslope contributing area calculated from
  flow accumulation ( of cells x area of cells)
• ß slope in degrees
• takes on high values for convergence zones, low
  values for divergent ridges
• basis for Topmodel

(Beven Kirkby 1979)
34
TCI
A
z
B
A
A high slope, low A, low W B low slope, high
A, high W
B
35
Drainage Indices

• Topographic Relative Moisture Index (TRMI)
• Summed scalar index
• Aspect
• Slope
• Plan curvature
• Profile curvature
• Relative slope position

(Parker 1982)
36
Synoptic Sampling of Soil Water Content
Gravimetric Sampling
Time Domain Reflectometry
37
Regression Models

• Response Variables
• Surface soil moisture (0-20 cm)
• Deep soil moisture (80-100 cm)

• Explanatory Variables
• Elevation
• TAspect
• Slope
• Dstrm
• TCI
• Radiation

38
Regression Models

• Shallow Soil Moisture (0-20 cm)
• y ?0 - ?1 Dstrm ?2 Elev - ?3 PRRsummer
• R2 0.54 (p ltlt 0.001)

• Deep Soil Moisture (80-100 cm)
• y ?0 - ?1 TCI ?2 Elev
• R2 0.35 (p0.03)

(Lookingbill Urban 2004)
39
DynamicProcesses
Summer
Winter
40
Dynamic Process Simulation Models
(Band et al. 1991, 1993)
41
Dynamic Formulation

• Climate parameters
• maxT, minT, and ppt from HQ met station 1999-July
  4, 2001
• scale base station T and ppt to appropriate
  elevation using climate submodel
• modeled VPD, tRad, and Rad attenuation

• Soil parameters
• Hydraulic conductivity 90.0 m/d
• Theta (sat) 0.45
• Theta (dry) 0.05
• Psi (min) -150 m
• Soil structure parameter 1.01
• Capillary length scale 0.05
• Most change permitted per permutation 0.5
• 11 nodes from 0 - 2.0 m
• Initial soil moisture 10

• Vegetation parameters
• albedo canopy 0.15
• albedo soil 0.29
• canopy interception coefficient 0.0002
• canopy light extinction coefficient -0.48
• max canopy conductance 0.0016
• slope of vpd-cc curve 0.01
• max plant avail soil water potential -140 m
• max plant rooting depth 2.0 m
• aerodynamic resistence plant canopy 30
• aerodynamic resistence soil surface 70
• LAI 6.0

42
Summary
vegetation pattern
nutrients temp. moisture
43
CART Model ofVegetation Pattern
44
NMS Model of Vegetation Pattern
45
Primary References

• Gradient Analysis
• Whittaker, R.H. 1956. Vegetation of the Great
  Smoky Mountains. Ecological Monographs 261-80.
• Whittaker, R.H. 1960. Vegetation of the Siskiyou
  Mountains, Oregon and California. Ecological
  Monographs 30279-338.
• Nutrient Conceptual Models
• Likens, G.E. and F.H. Bormann. 1995.
  Biogeochemistry of a Forested Ecosystem.
  Springer-Verlag, New York.
• Temperature Regression Models
• Lookingbill, T. and D. Urban. 2003. Spatial
  estimation of air temperature differences for
  landscape-scale studies in montane environments.
  Agricultural and Forest Meteorology 114141-151.
• Beers, T.W., P.E. Dress and L.C. Wensel. 1966.
  Aspect transformation in site productivity
  research. Journal of Forestry 64691-692.

46
Soil Moisture

• Drainage Indices
• Beven, K.J. and M.J. Kirkby. 1979. A physically
  based, variable contributing area model of basin
  hydrology. Hydrologic Science Bulletin 2443-69.
• Parker, A.J. 1982. The topographic relative
  moisture index an approach to soil-moisture
  assessment in mountain terrain. Physical
  Geography 3160-168.
• Regression Models
• Lookingbill, T. and D. Urban. 2004. An empirical
  approach towards improved spatial estimates of
  soil moisture for vegetation analysis. Landscape
  Ecology 19417-433.
• Process Simulation Models
• Band, L.E., D.L. Peterson, S.W. Running, J.
  Coughlan, R. Lammers, J. Dungan and R. Nemani.
  1991. Forest ecosystem processes at the watershed
  scale basis for distributed simulation.
  Ecological Modelling 56171-196.
• Band, L.E., P. Patterson, R. Nemani and S.W.
  Running. 1993. Forest ecosystem processes at the
  watershed scale incorporating hillslope
  hydrology. Agricultural and Forest Meteorology
  6393-126.