A. Henderson-Sellers, M. Fischer, K. McGuffie and D. Noone

1
iPILPS Community Modelling of Stable Water
Isotopes
A. Henderson-Sellers, M. Fischer, K. McGuffie and
D. Noone Australian Nuclear Science Technology
Organisation, University of Technology, Sydney
and University of Colorado
Introduction to PILPS
iPILPS Phase 1 Experiment
Forcing and Locations for Phase 1
Phase 1 of iPILPS tests the hypothesis that
observation and analysis of the diurnal fluxes of
1H218O and 1H2H16O between the soil, plants and
atmosphere can accurately determine the
partitioning of precipitation into transpiration,
evaporation and total runoff. iPILPS will
contribute (i) to improving the accuracy with
which land-surface schemes partition net
available surface energy into latent and sensible
heat fluxes and thus (ii) to decreasing the
uncertainty in hydro-climate modelling and water
resource vulnerability predictions. Phase 1 of
iPILPS exploits novel stable water isotope (SWI)
data and analysis techniques in the development
and evaluation of numerical models of ILSSs. To
achieve the project aims, it is necessary
to 1. identify and validate hydro-climate
models ILSSs which already (or plan soon to)
incorporate SWIs 2. appraise SWI data applicable
to hydro-climatic and water resource aspects of
ILSSs 3. identify ILSS model and observational
data gaps and resolve them and 4. apply SWI data
to specific predictions of well understood
locations simulated by available ILSSs.
PILPS is a key component of the Global Land
Atmosphere System Study (GLASS) which aims to
foster an evaluation of the next generation of
land-surface schemes and to coordinate the
evaluation of land-surface schemes in their
different applications. GLASS is a GEWEX project
which serves as an interface between the
land-surface community and international
modelling. PILPS GEWEX research is also
affiliated with IGBP, particularly ILEAPS. It is
co-chaired by Professor Ann Henderson-Sellers of
ANSTO and Professor Andrew Pitman of Macquarie
University. Created in 1992, PILPS has offered
the land-surface simulation community a frame of
reference for intercomparison and high quality
data sets against which to evaluate their
modelling of land-atmosphere exchanges for almost
13 years. It is the World Climate Research
Programmes (WCRP) longest MIP Model
Intercomparison Project.

• For iPILPS Phase 1, it was determined that the
  only way of supplying adequately good forcing was
  to use an isotope-enabled atmospheric model. The
  REMO (REgionales MOdel, developed by the Max
  Planck Institute for Meteorology, Hamburg) had
  been shown to generate high quality simulations
  for two of the three selected locations (Sturm et
  al., 2004a,b). The spatial resolution of REMO is
  ½ degree ( 54km) with a model timestep of 5
  minutes. REMO is nested into ECHAM and the
  iPILPS Phase 1 forcings were derived from nesting
  into the climatological version of ECHAM which
  had a constant annual cycle in sea-surface
  temperatures (Fischer and Sturm, 2005).

• The three locations selected for iPILPS Phase 1
  lie in GEWEX CSE basins and offer a broad range
  of ecological and climatic conditions
• Mid-latitude (deciduous) grass/wood, Munich 48N
  11E
• Tropical (evergreen) rainforest, Manaus 3S 60W
• Mid-latitude eucalypt (evergreen) forest,
  Tumbarumba 35S 148E
• A field campaign has been launched at Tumbarumba
  in Australia as part of the IAEA Moisture
  Isotopes in the Biosphere and Atmosphere (MIBA)
  project which will also provide additional new
  diurnal isotope data.

d18O in Annual precipitation

Isotopes in PILPS iPILPS
The goals of iPILPS are to (i) offer a framework
for inter-comparison of isotope-enabled land
surface schemes (ILSS) and (ii) encourage
improvement of these schemes by evaluation
against high quality (isotope) observations. When
iPILPS was approved by GLASS in September 2004,
it was agreed that its first stage (Phase 1)
would focus on the stable water isotopes H218O
and 1H2H16O with carbon isotopic simulations to
be combined in from 2006.
iPILPS Simulations and Data Needs
All ILSS owner are welcome to join iPILPS and
anyone with quality observational data at diurnal
resolutions is encouraged to contact the iPILPS
convenors (below).
To date 5 ILSSs have submitted results to the
interactive web page and these results can be
viewed on the web. iPILPS ILSS evaluation is
being undertaken in two stages each with two
components (i) conservation check and spin-up
analysis and (ii) comparison and evaluation
against observations. However, there is a dearth
of diurnal scale observations of the stable water
isotopes H218O and 1H2H16O.

• Phase 1 began in late 2004 with distribution of
  its plan and call for participants. Simulations
  started in February 2005 with a First Workshop in
  Sydney in April. Preliminary analysis shows
  interesting dispersion of simulated 2H in
  diurnally transpired flux (above for Tumbarumba
  January hourly average over all days).

Contacts (ipilps_at_ansto.gov.au)
http//.ipilps.ansto.gov.au Matt Fischer
(mjf_at_ansto.gov.au) Ann Henderson-Sellers
(ahssec_at_ansto.gov.au)