water forum2 yang
TRANSCRIPT
-
7/23/2019 Water Forum2 Yang
1/21
Developing a High-Resolution
Texas Water and ClimatePrediction Model
1Water Forum II on Texas Drought and Beyond, Austin, Texas, 22-23 October, 2012
Zong-Liang Yang
(512) 471-3824
-
7/23/2019 Water Forum2 Yang
2/21
Hyperresolution global land surface
modeling: Meeting a grand challenge
for monitoring Earths terrestrial water
2
-
7/23/2019 Water Forum2 Yang
3/21
Hyperresolution global land surface
modeling: Meeting a grand challenge
for monitoring Earths terrestrial water
An opinion paper by Wood, E. F., et al., 2011, WRR, 47, W05301.
A grand challenge to the community
Hyperresolution: O(1 km) globally; O(100 m) continental scales
Need for hyperresolution: global food production; waterresources sustainability; flood, drought, and climate changeprediction
Six major challenges: surfacesubsurface interactions due to finescale topography and vegetation;
landatmospheric interactions; soil moisture & evapotranspiration;
inclusion of water quality as part of the biogeochemical cycle; representation of human impacts from water management;
utilizing massively parallel computer systems in solving 109unknowns; and
developing the required in situ and remote sensing global data sets.
3
-
7/23/2019 Water Forum2 Yang
4/21
Hyperresolution regional land surface
modeling for Texas
An opinion presentation proposed today at Water Forum II.
A grand challenge to the community
Hyperresolution: O(100 m) CONUS; O(10 m) Texas
Need for hyperresolution: Texas food production; waterresources sustainability; flood, drought, and climate changeprediction
Six major challenges: surfacesubsurface interactions due to finescale topography and vegetation;
landatmospheric interactions; soil moisture & evapotranspiration;
inclusion of water quality as part of the biogeochemical cycle;
representation of human impacts from water management; utilizing massively parallel computer systems in solving 109unknowns; and
developing the required in situ and remote sensing data sets.
4
-
7/23/2019 Water Forum2 Yang
5/21
How to obtain hyperresolution weather
data for Texas?
O(10 m) is not feasible; but
O(1 km) is possible.
5
-
7/23/2019 Water Forum2 Yang
6/21
The grid resolution of regional and global models hasreduced hugely, and will continue to do so
e.g. 4 km resolution Weather Research and Forecasting (WRF)Model
4 km Topography
e.g. increasing spatial resolution in Global Models
EarthSimulator
www.nec.co.jp
www.nssl.noaa.gov/wrf/
3.5 km cloud resolution
N Atlantic ocean eddies
Revolution in ModelingShuttleworth (2011)
-
7/23/2019 Water Forum2 Yang
7/21
7
High Performance Computing Petascale [O(1015)] Computing Architectures
Massively parallel supercomputers (104105multi-core processors)
Worlds Fastest Supercomputer in 2011
579.4trillion math operations per second (TFlops)
3936 nodes, 62976 core processors
Texas Advanced Computing Center, UT-Austin
Ranger
5/19/2011
Worlds Fastest Supercomputer in 2013
10peta math operations per second (PFlops)
500,000 processors
Texas Advanced Computing Center, UT-Austin
Stampede
January 2013
-
7/23/2019 Water Forum2 Yang
8/21
8Yuan (2011); Shukla (2009)
-
7/23/2019 Water Forum2 Yang
9/21
While climate projections at O(100
year) timescales are useful, water
resource planning also needs climate
predictions at O(10 day) to O(1 year)time scales!
9
-
7/23/2019 Water Forum2 Yang
10/21
-
7/23/2019 Water Forum2 Yang
11/21
11
Step1: Seasonal Climate
Forecasts
Yuan (2011); Shukla (2009)
-
7/23/2019 Water Forum2 Yang
12/21
12
-
7/23/2019 Water Forum2 Yang
13/21
13
Step 2: Seasonal Climate
Dynamic Downscaling
Yuan and Liang (2011)
GRL
-
7/23/2019 Water Forum2 Yang
14/21
Dynamic Downscaling with Bias
Correction Improves the PDF of Daily
Maximum Temperature in Summer
Daily maximum temperature
Temperature bins (oC)
4-
6
6-
8
8-
10
10
-12
12
-14
14
-16
16
-18
18
-20
20
-22
22
-24
24
-26
26
-28
28
-30
30
-32
32
-34
34
-36
36
-38
38
-40
40
-42
42
-44
44
-46
46
-48
48
-50
Probability(%)
0
2
4
6
8
10
12
14
16
WRF_NNRP
WRF_CAM
WRF_CAMbc_ave
WRF_CAMbc_std
Black:WRF simulation drivenby NNRP
Green:
WRF simulation drivenby GCM output withoutbias correction
Blue:With mean value biascorrection
Red:With mean value andvariance biascorrections
The PDF is computed over the central US region(4050N, 10085W) at 60-kmresolution Xu and Yang (2012)
-
7/23/2019 Water Forum2 Yang
15/21
Xing Yuan & Eric Wood, Princeton University
-
7/23/2019 Water Forum2 Yang
16/21
UT has world-class expertise
in Understanding and modeling terrestrial
hydrological processes & global water cycle
Land Model for Climate Prediction
Land Model for Weather Forecasts Mapping geospatial datasets
Observing the global water cycle
High Performance Computing Lonestar
Ranger
Stampede
-
7/23/2019 Water Forum2 Yang
17/21
NCAR Community Land Model (CLM4)for Climate Models in 2010
Co-Chairs: David Lawrence (NCAR), Zong-Liang Yang (Univ of Texas at Austin)
-
7/23/2019 Water Forum2 Yang
18/21
Collaborators: Yang, Niu (UT), Chen (NCAR), Ek (NCEP/NOAA), and others
Noah-MPLand Model for Weather
Forecasts
A new paradigm in land-surfacehydrological modeling
In a broad sense,o Multi-parameterization Multi-physics Multi-
hypothesis
A modular & powerful framework foro Diagnosing differences
o Identifying structural errors
o Improving understandingo Enhancing data/model fusion and data assimilation
o Facilitating ensemble forecasts and uncertaintyquantification
18
Niu et al. (2011); Yang et al. (2011)
G it R d Cli t E i t
-
7/23/2019 Water Forum2 Yang
19/21
19
Gravity Recovery and Climate Experiment
(GRACE)
10 years of missionoperation (Tapley etal., 2004)
First-time global dataof gravity (~100 km,monthly to 10-day)
Unprecedentedaccuracy of mass
variations
Allowing a betterunderstanding of theglobal water cycle
-
7/23/2019 Water Forum2 Yang
20/21
Center for Integrated EarthSystem Science
http://www.jsg.utexas.edu/ciess
Formed inAugust 2011
Director: Liang Yang Associate Director:
David Maidment
A cooperative
effort between
-
7/23/2019 Water Forum2 Yang
21/21
Summary
21
CIESS was formed to integrate UTs
expertisein earth system science andcollaborate with national/internationalcommunitiesfor the betterment of society.
As high-resolution seasonal to decadalclimate and hydrologic forecasts areemerging as a new paradigm for modelingand prediction research, CIESS is positioned
to develop an integrated atmosphereland-surfaceriver network modeling system,applicable to Texas for water resourceapplications.