Summer Synthesis Institute

1
Summer Synthesis Institute
Overview of Synthesis Project Synthesis Project
Descriptions Summer Institute Logistics

• Vancouver, British Columbia
• June 22 August 5

2
Water Cycle Dynamics in a Changing Environment
Advancing Hydrologic Science through Synthesis
Murugesu SivapalanPraveen Kumar, Bruce Rhoads,
Don Wuebbles University of Illinois Urbana,
Illinois
3

• Objective 1. Conduct synthesis activities that
  will produce transformational outcomes in
  hydrologic science towards improved
  predictability of water cycle dynamics in a
  changing environment

• Objective 2. Use the synthesis activities as test
  cases to evaluate the effectiveness of different
  modes of synthesis for advancing the field of
  hydrologic science.

4
Limits to predictability

• Prediction means making probabilistic
• statements about future system states
• given the current and past observed states
• and our understanding of how nature works.
• The four classical limits of predictability are
• System identification (correct boundary
  conditions, driving forces)
• Characterization of initial states based on all
  available information
• Translation of our understanding of how nature
  works into a perceptual model of the system
  (identification of relevant /dominant processes,
  how they are coupled
• Appropriate mathematical representation (i.e.,
  numerical or predictive model) (parameters, model
  structure) to produce probabilistic statements

5
Behavior Structure Response

• Need a new predictive system which combines the
  traditional mechanistic perspective with an
  evolutionary perspective that includes (explicit
  or implicit) treatment of structure forming
  processes.

6
Predicting the Co-evolution of Coupled Physical
(Hydrological)-Biological-Human Systems
Praveen Kumar
7
Raupach, M.R., Barrett, D.J., Briggs, P.R. and
Kirby, J.M. (2005). Simplicity, complexity and
scale in terrestrial biosphere modelling. In
Predictions in Ungauged Basins International
Perspectives on the State-of-the-Art and Pathways
Forward (Eds. S. Franks, M. Sivapalan, K.
Takeuchi, Y. Tachikawa). IAHS Publication No.
301. (IAHS Press, Wallingford, UK). p. 239-274.
Figure 1.
8
LANDSAT image over coastal Sarawak, Malaysia, 1996
9
Dietrich et al., 1995
10
Typical banded vegetation pattern in Niger
(Valentin et al., 1999)
Spotted pattern (creosotebush) in the Chihuahuan
Desert of New Mexico (Wainwright et al., 2002)
Northern Territory Acacia woodlands
www.nt.gov.au/ipe/pwcnt
Saco et al., 2006
11

• Its a tangled WEB!
• WEBWater, Earth, Biota (Gupta, 2003)
• Water, as the life blood of the planet, links
  many earth components
• Many examples of interacting behavior across
  coupled systems
• Some connections are obvious, others subtle and
  devious
• C, N and P cycles are closely tied to the water
  cycle
• Biological activity, habitat structure are all
  dependent on the spatial fragmentation and
  temporal stability of water
• Space-time structure is often interesting as
  scale increases
• Many of the key problems in science ? scaling,
  nonlinearity, predictability, space-time
  oscillations emerge
• Methods of deconvolution of effects limited
  ?statistical or mechanistic

Upmanu Lall
12
Working Hypotheses

• Goal improved predictions of water cycle
  dynamics .. through increased understanding
• Water cycle dynamics is very complex, too
  difficult to predict using traditional (purely
  statistical or purely mechanistic) methods
• Patterns help us to reduce the complexity through
  reduced dimensionality, and thus help to improve
  predictions
• Patterns (both observed and so far unobserved)
  are emergent properties arising out of complex
  interactions and feedbacks between a multitude of
  processes.
• Study of patterns (how to describe them, why they
  emerge, their impact on the overall response)
  yields new insights and lead to increased
  understanding.
• Study of observed patterns (why they emerge) may
  give insights into unobservable or as yet
  unobserved patterns, and help to make improved
  predictions

13
Investigation of Emergent Patterns

• Top-down questions pattern description,
  measurement and identification. What can we learn
  from existing datasets?
• Theoretical questions deep, why type
  questions Why does this pattern emerge? Under
  what circumstances do we expect it to occur? What
  are the underlying rules?
• Bottom-up questions what are the consequences of
  these patterns (what are their effects on
  processes of interest)? How do they scale up? How
  does the understanding (e.g., their ecological
  function, organizing principles etc.) improve our
  capacity to make predictions?
• Human interactions how do human activities
  interact with these patterns in time and space?
  How are the patterns affected by human
  activities?
• Study of patterns needs a multitude of
  perspectives (concepts, data, methods etc. from
  different disciplines)
• Synthesis means people with different backgrounds
  and experiences coming together to study a common
  question or pattern or prediction problem and to
  help each other to generate increased
  understanding

14
1. How do spatial patterns emerge?

• developing a theoretical framework for
  exploration of interacting hydrological,
  pedological, geomorphological, ecological and
  meteorological processes that contribute to soil,
  topographic and vegetation structure formation
  and their evolution.

15
2. How do patterns of heterogeneity manifest in
hydrological response and functioning?

• developing a prediction framework for accounting
  for (explicitly resolving and parameterizing) the
  interactions of topographic, soil, vegetation,
  geomorphologic, ecologic heterogeneity and their
  manifestation in hydrological, biogeochemical and
  ecological responses and functioning.

16
Themes Projects

• Theme Interactions between hydrosphere and
  biosphere processes
• Session 1 Quantifying Vegetation Adaptation and
  Response to Variability in the Environment
• Session 4 Comparing catchment-based estimates of
  vegetation water use (Horton Index) with remote
  sensing measures of vegetation structure, water
  use, productivity
• Theme Interactions of landscape processes
  (within intensively managed landscapes)
• Session 2 Contaminant Dynamics across Scales
  Temporal and Spatial Patterns
• Session 3 Temporal and spatial patterns of
  basin scale sediment yield

17
Schedule

• June 21 Arrival (June 20 by request)
• June 22-30 Session 1 (led by Siva and Ben)
• July 1-10 Session 2 (led by Suresh and Nandita)
• July 11-19 Session 3 (led by Marwan)
• July 20-30 Session 4 (led by Peter and Paul)
• July 31-August 2 Wrap up work
• August 3-5 Capstone event
• Depart August 5 or 6

18
Session 1
Quantifying Vegetation Adaptation and Response
to Variability in the Environment
Ben RuddellArizona State University
Siva SivapalanUniversity of Illinois
Ciaran Harman University of Illinois
Gavan McGrath University of Western Australia
19
Goal

• Can we quantify the relationship between
  vegetation (NPP) and the precipitation water
  balance (Horton Index)?
• Data-based analysis is happening at the
  University of Arizona.
• Can we produce a simple process-based model to
  test hypotheses on how the adaptation and
  activity of vegetation controls the water balance
  (and vice versa)?

20
Precipitation Water Balance
P Precipitation S Surface/Fast Runoff W
Soil Wetting E Plant Evaporation U
Lateral/Subsurface Runoff Q Total Runoff H
Horton Index (Troch) Budyko Lvovich argued
for competition between WS or EU. What is the
strategy of the plant ecosystem, and how does is
affect the competition?
21
Vegetation Adaptation to Variability in Limited
Resources

• LAI as a measurable proxy for NPP in terrestrial
  ecosystems
• R is the a limiting resource
• Water
• Energy (solar)
• Carbon Dioxide
• Nutrients (N)
• BUT R varies in time, and the limiting resource
  switches.
• Use dimensionless numbers to quantify adaptation
  of ecosystem to resource variability

Try Later
22
Session 2
Contaminant Dynamics across Scales Temporal and
Spatial Patterns
Nandita Basu University of Iowa
Suresh Rao Purdue University
Aaron Packman Northwestern University
23
Single Tile (1 km2)
Cedar Creek(700 km2)
Mississippi Basin (3 million square km)
Tile Network(10 km2)

• Contaminants
• Nutrients (C, N, P)
• Pesticides
• Hormones

23
24
Conceptual Model
Landscape (non-linear filter)
Climate (rainfall, ET)
Streamflow
Biogeochemistry (non-linear filter)
Contaminant Loads
Aquatic Habitat and Biodiversity

• Non linear filters create emergent
  patterns/signatures across scales
• Signatures integrate ecosystem structure and
  function
• Relationship of water flow and water quality to
  stream ecosystems
• Examining signatures using data analysis and
  models

25
Questions Spatial Patterns

• How does the mean annual contaminant load scale
  with basin area? Is it a mono-fractal or a
  multi-fractal?
• How do climate and land-use changes impact these
  patterns?
• What is the effect of reaction non-linearity
  along the network on these patterns?
• What is the coefficient of variation of the
  contaminant load and how does that scale with
  area?

25
26
Network Models Spatial Patterns
Nitrogen loads across Mississippi Basin
(Alexander et al. 2000)
Nitrogen Loads across a 700 km2 watershed in
Indiana
26
27
Questions Temporal Patterns

• What are the correlation timescales for
  contaminant loads compared to water?
• Memory effects/old water vs. new water?
• Others.

27
28
Network Models Temporal Patterns
Note different correlation time scales for
nitrate compared to precipitation (P) and
streamflow (SF)
28
Zhang and Schilling 2005
29
Conceptual Approach The 4-Ps
Processes
Patterns
Purpose
30
Session 3
Temporal and spatial patterns of basin scale
sediment yield
Marwan Hassan University of British Columbia
Aaron Packman Northwestern University
31
Introduction

• Studies of the sediment yield are important for
• identifying sources of sediment in river basins
• measuring soil erosion
• determining rates of landform change
• testing land use/management practices
• Sediment yield provides a simple lumped
  representation of the linkages between the
  erosion processes which operate within the
  drainage basin and the downstream sediment
  delivery
• Controlling Factors
• Climate and vegetation
• Basin Size
• Elevation and Relief
• Rock Type
• Land-use and human activity

32
Comments

• A number of studies have attempted to model and
  explain regional or global sediment yield trends
  in terms of climate, topography, and landscape
  history.
• However, several problems have arisen, including
  data quality and the lack of accepted
  extrapolation procedures and issues of scale
  dependency.

33
Objectives/questions

• To examine the spatial and temporal variation in
  sediment yield using the regional sediment yield
  relations for the landscape.
• To examine the magnitude and frequency of
  sediment mobilizing events (hydrologic VS
  Geomorphic Magnitude and frequency, spatial
  patterns).
• To explore the geomorphic connectivity of the
  landscape (efficiency of sediment transfer) ---
  we plan to link sediment sources (slopes, banks
  and bed) to sediment yield using reach-scale
  budgets.
• To explore the response of the landscape to land
  use and soil conservation practices (has been
  studied for small basins but not at the landscape
  scale).
• To identify possible landscape responses to
  climate change and other disturbances.
• Is it possible to transfer findings from one
  scale of investigation to another?
• Modelling?

34
Potential data sets

• Mississippi
• Yellow River
• Yangtze
• Global

35
Session 4
Comparing catchment-based estimates of vegetation
water use (Horton Index) with remote sensing
measures of vegetation structure, water use,
productivity
Peter Troch University of Arizona
Paul Brooks University of Arizona
36
Background
Hydrological research has demonstrated the strong
control that ecosystems have on the partitioning
of precipitation into runoff and ET.
Ecological research has focused on the strong
control that water availability has on
productivity.
37
Background

• Recently, Troch et al. revisited Hortons 1933
  paper and subsequently demonstrated that the
  Horton Index (a runoff-based estimate of the
  fraction of potentially available water used by
  vegetation)
• Remains relatively constant across years in a
  single catchment
• Converges to a common, high value (0.95) as
  conditions become more arid

The natural vegetation of a region tends to
develop to such an extent that it can utilize the
largest possible proportion of the available soil
moisture supplied by infiltration (Horton, 1933,
p.455)
38
Project Activities

• This project will
• expand on the number and climatological
  characteristics of catchments used in Troch et
  al.s initial Horton Index work, and
• add remotely-sensed analyses of vegetation
  structure and activity (LAI, NDVI, ET, PSN, GPP).
• Our overarching hypothesis is that variability in
  vegetation structure and activity will be more
  closely related to plant available water
  (estimated using the Horton Index) than to
  precipitation directly

Remote sensing provides catchment wide estimates
of vegetation structure and function in response
to available water
39

• Questions and Locations
• How are catchment-based estimates of plant
  available water related to remotely-sensed
  information on vegetation across a range of
  climate and vegetation?
• subset of 92 MOPEX sites all 432 MOPEX sites

• How does the storage and release of seasonal snow
  influence the Horton Index and vegetation
  response?
• Subset of MOPEX sites and experimental catchments

• Is the magnitude or variability in the Horton
  Index related to catchment nutrient export?
• experimental research catchments

• Are remotely sensed data on vegetation response
  biased by climate or species?
• FLUXNET sites

40
Locations MOPEX catchments
41
Locations Research Catchments and FluxNet
Andrews
Hubbard Brook
Bartlet Ex. Forest
Blodgett Forest
Shortgrass Steppe
Cedar Bridge
Baltimore
Konza
Niwot Ridge
Jasper Ridge
Duke Forest
Valles Caldera
Phoenix
Walker Branch
Coweeta
Sevilleta
S. CA climate grad.
Walnut Gulch
Santa Rita Mesquite
Freedman Ranch
42
Data USGS Streamflows

• Data Availability
• from 10/1999 to 12/2008
• 363/88 catchments
• Computed
• Baseflow and Direct runoff
• Methods
• Local minima
• One parameter filter (Lyne Hollick, 1979)
• Recursive filter (Eckhardt 2005)

43
Data MODIS Products

• Vegetation Indices (MOD13A2)
• NDVI - Normalized Difference Vegetation Index
• EVI - Enhanced Vegetation Index
• 1 km 16 days
• LAI/Fpar (MOD15A2)
• LAI - Leaf Area Index
• FPAR - Fraction of Photosynthetically Active
  Radiation
• 1 km 8 days
• Gross Primary Productivity (MOD17A2)
• GPP - Gross Primary Productivity
• PsnNet - Net Photosynthesis
• 1 km 8 days
• Data spatially aggregated for 431 MOPEX
  catchments
• AVG STD CNT
• Data Source

44
Data Climate

• Data Availability
• from 10/1999 to 12/2008
• Sources
• NCDC
• Precipitation
• Temperature (min, max)
• 8700 sites
• SnoTel
• Precipitation
• Temperature (min, max)
• Snow water equivalent
• 703 sites
• USGS
• Precipitation (57)

45
Capstone Event

• 2.5 days (August 3-5)
• 6-9 CUAHSI-funded early career participants
• Presentations from students and CUAHSI
  participants
• Dialogue about synthesis among students,
  returning mentors, CUAHSI participants

46
Vancouver UBC

• Lows - 55-50 F (14-15 C) and Highs - 70-75 F
  (22-23 C)

47
Living _at_ UBC

• Marine Drive Residence
• Brand new building
• Private bedroom with single bed, telephone and
  clock radio, in four-bedroom shared apartment.
• Guests share two washrooms, lounge area and
  kitchenette.

• Food
• Meal vouchers _at_ food court and cafeteria
• Some group meals and partial per diem for meals
  off campus
• Internet
• Wired internet _at_ Marine Drive
• Wireless on campus/Geography
• Transportation
• Cab/shuttle to/from airport
• Bike rentals
• Possible group vehicle

Log on to cybercollaboratory for important
information about travel documents and air
transport
48
What will my day be like?

• Working individually
• Working in pairs or small groups
• Meeting with faculty mentors
• Meeting as a full group to discuss progress, next
  steps
• Seminars with mentors or visitors
• Weekly brown bag webinar with full synthesis team

49
Project Coordinator

• Jennifer Wilson
• Work 217-244-6193
• Cell 217-778-4054
• Email jswilson_at_illinois.edu
• Attending summer institute at beginning and end
• Keep in touch via email, cybercollaboratory
  throughout summer
• http//hydro.ncsa.uiuc.edu/cybercollab