Modeling tools and Web based technologies can support water resources management by: 1) Applying numerical models to study groundwater, surface water, and marine water dynamics through case studies. 2) Addressing challenges in environmental science like complex multi-scale dynamics and data availability. 3) Developing information systems and simulations to analyze pressures, states, and impacts on the environment and identify critical areas.

1. Modeling tools and Web based
technologies to support water
recourses management
Pierluigi Cau
Energy and Environment Program
Center for Advanced Studies, Research and
Development in Sardinia
plcau@crs4.it
CRS4
Sardegna Ricerche, 09010 Pula CA, Italy
http://www.crs4.it

2. The mission of the E &E program
CRS4 Mission and the Grand Challenges in the
Environmental Sciences
• Development of physical and numerical models
implemented on HPC platforms for high resolution
simulations
• Software tools development for the analysis and
management of environmental data, integration of
information systems and numerical applications

3. Expertise: Environmental Science
• Hydrological (SWAT, T-RIBS, MIKE SHE, Qual 2K) – Groundwater (CODESA 3D,
Modflow, Feflow) – Ocean Modeling (GETM, GOTM)
• HPC platforms, Cloud and Distributed Computing, Virtualization technologies in
the field of Environmental management and monitoring
• WEB based information systems that relies on a geographically distributed GIS,
RDBMS, complex models

4. Objectives of the presentation
The aim is to present:
1. the application of ICT numerical tools to study water dynamics for:
- Groundwaters
- the Oristano and the Portoscuso case studies,
- Surface water
- The Cedrino, San Sperate, ….. Case studies
- Marine waters
- The Orosei and Asinara case study
2 .The challenges in the environmental science
3. Future work

5. ISSUES: Environmental Science
Complexity of environmental issues
- multimedia environment,
- multi scale (time and spatial) dynamics
- complexity of the description of the system (lack of quality data)
- characterization of the territory and the interaction with atmosphere:
- complexity of anthropogenic pressures:
• agricultural, zootechnical, civil, industrial pollution
- Complexity of environmental dynamics
- climate change
• The Intergovernmental Panel on Climate Change predicts a further rise
of the air temperature between 1.4°C and 5.8°C by the end of the
century and as a consequence a sea level rise of about 1 to 2 mm/year.
- EU/National/Regional Directives (EU WFD, MSFD, etc.)
There is a need to improve comprehension and modeling technique at scales
relevant to decision making of climate induced changes

6. ISSUES: Environmental Science
Tools
Data, expertise, numerical codes, analysis and visualization tools, etc.
Objectives
Improve the wise management of water and natural resources by:
• Predict the impact of environmental changes, such as climate or land
use changes, on water resources;
• Better comprehend the cause-effect relationship on the local and
large scale (natural and anthropogenic stresses versus ecosystem
responses)
• ….
Improve the usability of models and the interoperability between systems
through mesh up of web applications
Fill the gap between research and production (PA, economic operators,
etc.)

7. From Modeling to Industrial Projects
Environmental issues make necessary a strong integration of
expertise from different disciplines, made possible through the
development of virtual organizations of federated entities
Decision
makers
Problem
definition
Possible
alternatives
DPSIR: a causal framework for describing Development &
the interactions between society and the Implementation
environment: Performance
Driving forces (e.g. industrial production) evaluation
Pressures (e.g. discharges of waste water)
States (e.g. water quality in rivers and lakes)
Impacts (e.g. water unsuitable for drinking)
Responses (e.g. watershed protection)

8. From Modeling to WEB Services
A problem-solving cloud platform for the
integration, through a computing portal, of The virtual organization acts as a
resources for service provider while each
communication partner becomes the recipient of
computation the WEB services
data storage
visualization A cloud is an infrastructure that allows
simulation software the integrated and collaborative use of
instrumentation virtualized resources owned and
human know-how managed by one or more entities
in Environmental Sciences

9. Some Projects: 2002-2010
PdTA – Piano di Tutela delle Acque secondo la 152/99
Decision Support and Information System for water management
http://www.regione.sardegna.it/j/v/25?s=26251&v=2&c=1260&t=1
Datacrossing / Climi Aridi
Web based tools for groundwater management and monitoring
http://datacrossing.crs4.it
Climb
Integration of climate and hydrological model
www.climb-fp7.eu/
EnviroGRIDS - Nuvola
Web based Information System and tools to model superficial waters
http://www.envirogrids.net
MOMAR
Web tools to model the water cycle: from the watershed to the marine environment
http://www.mo-mar.net

10. Groundwaters
Conceptual model – coastal shallow aquifer case
Dirichlet BC Neumann BC

11. Groundwaters
Challenges:
Challenges Model set-up, calibration and uncertainty.
set-
Kh and Kv are assumed deterministic for the phreatic aquifer on the basis of
limited field data
lateral inflow and vertical recharge boundary conditions for the
groundwater model are indirect measure (e.g. calculated by the SWAT code)
the geometry has been built on the basis of heterogeneous data (geologic
map, boreholes and geophysical data)
uncertainty of the interactions between the superficial water bodies and the
groundwater system:
- disconnected (I conceptualization)
- connected or partially connected (II conceptualization)
lack of adequate control data (heads and concentrations)
few control points - few measures

12. Groundwaters: case studies
1 Oristano (Italy)
-Seawater intrusion
2 Portoscuso (Italy)
-Industrial contamination
1 3
2 3 Muravera (Italy)
-Seawater intrusion
5
4 Oued Laou (Marocco)
Aquifer management
4
5 Corba (Tunisia)
Aquifer management

13. Groundwaters : the Oristano Case study
Study the hydrodinamic and the seawater intrusion process
of the aquifer;
Quantify the effect of a possibly discontinuous aquitard on
the salt dispersion process;
Identify contaminated areas more sensitive to aquitard
heterogeneity;
Evaluate the impact of alternative exploitation schemes on
the salt water intrusion;

14. Groundwaters : the Oristano Case study
• soil surface 280 x106 m2 ~ 270 km2;
• aquifer average thickness t =123 m, 18 m < t < 218 m;
• aquifer volume 17.8 x109 m3
•2D surface nodes 1873; 2D surface triangles 3618;
• vertical layers 10;
• 3D nodes 20603; 3D tetrahedra 108540
zoom
A
A

15. Groundwaters: the Oristano Case study

16. Groundwaters: the Oristano Case study

17. Groundwaters: the Oristano Case study
Alternative aquifer exploitation schemes
The Monte Carlo simulation has been run for each of the following
scenarios:
A. A. Pumping from the phreatic aquifer only;
B. B. Pumping from the deep aquifer only;
C. C. Pumping from both aquifers together.

18. Groundwaters: the Oristano Case study
Aquitard hydraulic conductivity K is assumed as the sole source of
uncertainty. K is modeled as a stationary random function with a lognormal
distribution y = ln(K) with K=10-8 m/s, s2(y) = 10 and an exponential
covariance function.
An example of a ln(K) synthetic realization
σ
(σ2 = 10)
Methodology:
1. Generate NSIM synthetic
realizations of the K field by means of
a stochastic (HYDRO_GEN) model;
2. Simulate the NSIM correspondent
pressure heads and concentrations
using the coupled flow & transport
CODESA-3D model;
3. Perform a probabilistic threshold
analysis and evaluate the performance
of the system by means of ensemble
indicators.
Lighter colors represent aquitard “holes”

19. Groundwaters: the Oristano Case study
Monte Carlo iterates to garantee stationarity
normalized avarage of the I moment versus number of iterates
4
normalized avarage of the II moment versus number of iterates
2
0
-2
-4
-6
0 10 20 30 40 50 60 70 80 90 100

20. Groundwaters: the Oristano Case study
Saltwater front ( c = 0.1 [/]) probability map
Pumping schemes: A and B
A B
20

21. Groundwaters: the Oristano Case study
Pumping schemes: A and B
5%<P<95%
A B
21

22. Groundwaters: the Oristano Case study
Contaminated areas sensitive to aquitard heterogeneity
NSIM (cij - c i ) 2
Time evolution of the
concentration nodal variance (4th layer)
σ i2 = ∑j=1 NSIM
10 Years 25 Years 40 Years 50 Years
σ2(c)
Pumping case (A)
22

23. Groundwaters: the Oristano Case study
Main statistical indicators
∆c

24. Groundwaters : the Portoscuso case study
Study the hydrodinamic and contamination of the aquifer;
Set up a numerical procedure to find the most likely pollution
sources;
Identify the area controlled by the monitoring wells
Set up an interactive Information system to view result;

25. Groundwaters: Portoscuso
Computational domain
 ∂ψ ∂c ρ
 σ ∂t = −∇⋅ v −φSwε ∂t + ρ0 q flow

 equation
φ ∂(Swc) = −∇(cv) + ∇ ⋅ (D∇c) + qc* + f
 ∂t
 transport
equation

26. Groundwaters: Portoscuso
Optimal Water Resources Manager: from Field Data to the
Contamination Source (an Inverse Problem)

27. Groundwaters: Portoscuso
Optimal Water Resources Manager: from Field Data to the
Contamination Source (an Inverse Problem)

28. Groundwaters: Datacrossing
Optimal Water Resources Manager: from Field Data to the
Contamination Source (an Inverse Problem)
The most likely
contamination source
The DSS interpolates the simulated
nodal concentrations generated by the
groundwater application and visualizes them
using MapServer and msCross from
Datacrossing

29. Groundwaters: Portoscuso
Optimal Water Resources Manager: from Field Data to the
Contamination Source (an Inverse Problem)
Montecarlo Sim Disk space Total Disk Space
(1 PP) 2238 45 MB/sim 100 GB
Montecarlo Sim CPU time/sim Total CPU Time
(1 PP) 2238 5 min-6 ore about 2 months

30. Groundwaters: monitoring wells
T= 0
T= 12
The model is used to assess the
T= 6
effectiveness of the monitoring network in
detecting contamination. The area of
influence of 41 wells, at different time steps
(from top to bottom: 0 months, 6 months,
12 months) is shown in light blue. Outside
this area, within the same time period,
contamination sources will not affect the
water quality of the wells. The monitored
areas are expected to become larger with
time as shown in this figure.

31. Groundwaters: Datacrossing
Optimal Water Resources Manager: sea water intrusion

32. Groundwaters: Datacrossing /Climi Aridi
The OUED LAOU test case (Marocco)
Objectives of the project
• Increasing the level of knowledge of the Mediterranean coastal
aquifers developing the hydrogeological model of the Oued Lou;
• Developing innovative procedures and tools and improve the
understanding of geographically distributed hydro-geological,
physical, and geo-chemical variables;
• Increase cooperation between Sardinia and Marocco through:
– training for students and advanced training for researchers
– seminars and dissemination events

33. Hydrology: EnviroGRIDS/Nuvola
Modeling Environmental Dynamics
Development and implementation of
Objectives mathematical methods and innovative WEB
• Analyze pressures, states and based ICT tools to support adaptive
impacts on the environment; strategies to face issues of water and soil
• Identify critical areas (e.g. resource vulnerability
affected by desertification);
• Run scenarios on a multi model
& multi scale framework
• produce report on a friendly
environment;
• Improve model usability;
• Improve public consciousness.

34. Hydrology: EnviroGRIDS / Nuvola
THE SWAT Model
It is a hydrological watershed-scale model developed by the
USDA Agricultural Research Service (ARS) and Texas A&M
University.
SWAT aims at predicting the impact of land management
practices on water, sediment, and agricultural chemical yields
in large complex watersheds with varying soils, land use, and
management conditions
over long periods of time.
The water cycle (precipitation, run off,
infiltration, evapotranspiration, etc.),
sediment cycle, crop growth,
nutrient (N, P) cycle are directly
modelled by SWAT.

35. Hydrology: ISSUES

36. Hydrology: Case studyes
The Cedrino (Italy) Watershed The S. Sperate (Italy) Watershed
The Black Sea Watershed The Gange (India) Watershed

37. Hydrology: Cedrino
Virtual river network Land Cover Soil
DAILY PLUVIOMETRIC DATA 1955-2007 DAILY TERMOMETRIC DATA 1955-2007

38. Hydrology: Cedrino
Calibration
NASH-SUTCLIFFE
INDEX [-∞,1]
Calibration period (1957-1964) HRU
Initial K NS -4,4 MULTIPLE
HRU
SWATCUP (1500 runs): NS finale 0,41 DOMINANT
The complexity of the
simulation has been increased

39. Hydrology: Scenarios assessment

40. Hydrology: Soil water stress
Modeling Environmental Dynamics: the agricultural
drought for the Black Sea catchment
The Yellow/orange
indicates
soil water deficit

41. Hydrology: the Black sea Catchment
Modeling Environmental Dynamics: the agricultural
drought for the Black Sea catchment
We assess and quantify complex environmental dynamics through the use of sophisticated,
reliable models.
The Yellow/orange
indicates
soil water deficit

42. Hydrology: The Gange (India) river
Modeling Environmental Dynamics: water quality and
quantity states

43. Hydrology: Climate analysis
The Objective is to:
- check the atmospheric/climate model output and see if they
are consistent with the SWAT model specification
- set up a semiautomatic procedure to gather meteorological
data and produce climatic data fit for the SWAT Model
- analyze the effect of the spatial downscaling on the water
balance for a case study
- Quantify the uncertainty of the meteo-hydrological model
chain. What limitation/uncertainty do we expect to have by
using the meteorological data to feed the hydrological model?

44. Hydrology: Climate analysis
The Objective is to:
- check the atmospheric/climate model output and see if they
are consistent with the SWAT model specification
- set up a semiautomatic procedure to gather meteorological
data and produce climatic data fit for the SWAT Model
- analyze the effect of the spatial downscaling on the water
balance for a case study
- Quantify the uncertainty of the meteo-hydrological model
chain. What limitation/uncertainty do we expect to have by
using the meteorological data to feed the hydrological model?

45. The ensemble climate model
The Ensembles Prediction Systems is based on global
Earth System Models (ESMs) developed in Europe for use in
the generation of multi-model simulations of future climate
The project provides improved climate model tools developed
in the context of regional models, first at spatial scales of 50
km at a European-wide scale and also at a resolution of 20 km
for specified sub-regions.

46. The ensemble climate model
A comprehensive analysis has been carried out.
Complete daily data Incomplete daily data Missing data
Istitution Country Note
CNRM-ARPEGE-new France No data – Only ancillary
CNRM-ARPEGE-old France No data – Only ancillary– Lustrum step
DMI Denmark
DMI-BCM Denmark No data – Only ancillary – Start: 1961
DMI-ECHAM5 Denmark Last time interval: 2091-2099 (9 years instead of 10)
ETHZ Switzerland Last time interval: 2091-2099 (9 years instead of 10)
GKSS-IPSL Germany No Daily step
HadRM3Q0 UK
HadRM3Q16 UK
HadRM3Q3 UK
ICTP Italy
KNMI Netherlands Is present a yearly simulation (1950-1950)
METNO Norway Last time interval:2041-2050
METNO-HadCM3Q0 Norway Last time interval:2041-2050
MPI Germany
SMHI-BCM Sweden Start: 1961-1970
SMHI-ECHAM5 Sweden
SMHI-HadCM3Q3 Sweden
VMGO Russia Last time interval: 2021-2030 (pr); 2011-2020 (tasmin, tasmax)

47. Model result: comparison
SAR-PCP
MPI climate model-PCP

48. Model result: comparison
PCP-SAR
MPI climate model-PCP

49. Ocean dynamics: MOMAR
Modeling Marine Water Dynamics
A multi-model and multi-scale WEB-based
environment for coastal protection
Objectives
• Analyze pressures on coastal
areas;
• Identify major pollution sources;
• Model the bio-geochemical
status of the sea;
• Run scenarios on a multi model
& multi scale framework;
• Produce report on a friendly
environment;
• Improve the monitoring network;
• Improve model usability;
• Improve public consciousness.

50. Ocean dynamics: GETM
General Estuarine Transport Model (GETM)
GETM is a Public Domain, finite difference numerical 3D
oceanographic model, most efficiently used to study shallow
waters and natural processes in natural marine waters.
GETM simulates hydrodynamic
and thermodynamic processes in
natural waters, like currents, sea
level, temperature, salinity, and
vertical / turbulent mixing.

51. Ocean dynamics: GETM
The GETM workflow
• a batch procedure downloads daily:
- updated meteorological/oceanographic
data from regional models:
1. http://nomads.ncep.noaa.gov/
2.http://www.ifremer.fr/thredds/catalog.html
• Boundary (BC) and Initial Condition (IC) are
interpolated on the high resolution GRID from the
above data for the GETM oceanographic model.
• a set of configuration files are updated to match
each new operational condition;
• GETM is run and produce outputs in NETCDF
format (about 4 GB ).
• Each output file is processed to produce a
spatialite db file to be displayed on the WEB
interface .

52. Ocean dynamics: interoperability
FROM MARS 3D to GETM/BASHYT
Orosei Gulf - Forcast 21-03-
2011 18:00 - Salinity
distribution

53. MOMAR (INTERREG)

54. MOMAR (INTERREG)
Oil Spill Model (Lagrangian approach)

55. MOMAR (INTERREG)
River impact

56. MOMAR (INTERREG)

57. The Asinara CASE
ASINARA: Oil spill – Gennaio 2011
Setup GETM 0.0016 con vento GFS

58. The Asinara CASE
ASINARA: Oil spill – Gennaio 2011
Setup GETM 0.0016 con vento MARS3d

59. Conclusion
Environmental issues make necessary a strong integration of expertise from different
disciplines, made possible through the development of virtual organizations of federated
entities
Reliable model prediction is primarily based on the acquisition and the efficient use of large
quality dataset and the development of an interdisciplinary approach to the study.
Today SW technology makes almost transparent the operability of a cloud/grid
infrastructure (network, compute and data resources) for the sharing and the exploitation
of complex applications via Internet
Shifting environmental applications from the desktop oriented approach to the web based
paradigm enhances flexibility in the whole system, extends the use of data and the sharing
of experiences, fostering user participation.

60. Conclusion
With the collaboration of:
Simone Manca, Davide Muroni, Costantino Soru, Marco Pinna,
Giuditta Lecca, Fabrizio Murgia, Antioco Vargiu, Gian Carlo Meloni,
Carlo Milesi, Paolo Maggi, Stefano Amico, Ernesto Bonomi, Michele
Fiori, Elisaveta Peneva, Gian Piero Deidda, and many more!!!
With the support of:
Regione Autonoma della Sardegna, Climb project, Nuvola project,
EnviroGRIDS project, MOMAR project