1.0 - Observations

We have access to several Italian observational datasets provided by the University of L'Aquila for:

• temperature
• precipitation
• water level
• discharge

1.1 - Observations

Step 2 - identify station problems:

• short timescale
• shifts
• (out)liers
• errors
• breaks
• fill values

Step 1 - transform the (binary) station datasets in more useful (netCDF) files

Step 3 - fix problems, homogenize datasets, create gridded databases

1.2 - Observations

Outliers

1.3 - Observations

Po discharge database

Fill constant

1.4 - Observations

Dewetra discharge database

Break

Shifts

Outlier

1.5 - Observations

Dewetra waterlevel database

1.6 - Observations

Acqwapo discharge database

Quantization

2.0 - Regional Climate Models

RCMs will be used later in the project to provide gridded input data to the CHyM hydrological model for future projections under climate change scenarios

I have obtained experience with:

• Data analysis of 9 CORDEX RCMs against 9 high resolution European observational datasets (Fantini et al. 2016)
• Model tuning of the RegCM4.6 model
• RegCM4.6 performance assessment with different clusters (Argo, CINECA-Marconi A1/A2)

2.1 - Regional Climate Models

We have assessed the performance of 9 RCMs over 9 regions for precipitation:

• High resolution observations
• Several metrics, with focus on extreme precipitation
• We found consistent added value in higher-resolution modelling (12km vs 50km) for most regions and most models

2.2 - Regional Climate Models

I performed more than 100 model simulations over the EURO-CORDEX domain in both Argo and CINECA's Marconi A1/A2

• Model tuning
• Performance assessment
• PRACE proposal 2016153590 approved

3.0 - Cetemps Hydrological Model

CHyM Is a distributed (gridded) hydrological model. Peculiarities:

• Can build DEM from various sources, smoothing by cellula automata algorithms
• Can use several kind of inputs, such as station observations, gridded model data, etc.
• Designed to work on any domain
• NetCDF output

3.1 - Cetemps Hydrological Model

Successes so far:

• Model now working on all test domains
• Model code and design has been streamlined (thanks to Fabio di Sante!)

• We can reproduce past results (Coppola et al. 2013) on our test domain (western Po basin)
• HR DEMs working (at low model res)

3.2 - Cetemps Hydrological Model

Failures so far:

• River network reconstruction at high resolutions (<500m) fails due to singularities in the DEMs
• Discharge at low resolutions must be rescaled

3.3 - Cetemps Hydrological Model

CHyM default Italian DEM is 300m in resolution, we wanted to try higher resolution DEMs:

• ASTER 30m
• HydroSHEDS 90m

• SRTM 90m
• JAXA 30m (wip)

New HR DEM is able to reconstruct the river network at low model resolutions (900m)

3.4 - Cetemps Hydrological Model

CHyM-OP reproduced domains:

3.5 - Cetemps Hydrological Model

CHyM-OP reproduced domains:

5.0 - LISFLOOD-FP hydraulic model

Widely-used flood inundation model from the University of Bristol (Bates et al., 2010)

5.1 - LISFLOOD-FP hydraulic model

Problems!

• Excessive flooding in low RPs
• Not enough difference between RPs

Solutions?

• Higher resolution (~50m)
• Better DEMs (HydroSHEDS)
• ?

6.0 - Visualization

An R/Leaflet tool for flood, river, DEM, basin and station visualization

6.1 - Visualization

Click_edit: an R/Shiny tool for WYSISYG editing of NetCDF files