Dino Zardi

Via Mesiano, 77 - 38123 Trento
tel. 0461 282682 | 0461 282687
dino.zardi[at]unitn [dot] it
| http://www.ing.unitn.it/~zardi/dz.htm
Caratterizzazione di processi dello strato limite atmosferico mediante analisi di dati da misure di turbolenza (in collaborazione con CNR-ISAC)
martedì 17 agosto 2021

The general frame 

Since long time, the common view of the turbulence in the atmospheric boundary layer (ABL) was strongly related to ‘ideal’ situations: vertical fluxes of momentum and heat drive the mean vertical profiles of wind and temperature and the variances of the fluctuations of velocity and temperature, the flux-gradient relationship holds, turbulent kinetic and temperature variance dissipations are described by the inertial subrange paradigm. In short, small scale turbulence was accounted for, parameterizations in Numerical Weather Forecast (NWF) models was based on it, possibly with semiempirical adjustements, in spectral terms the existence of a spectral gap was a strong hypotesis.. 

Accurate measurements (say in the last 20 years) in a variety of environments and a critical approach to the real world lead to recognize the presence and the importance of some ‘large scale turbulence’, often called submeso motions, which change severely our understanding and description of the ABL.


Broad arguments of the thesis

Using available datasets the work will concentrate on some aspects of the statistical analysis of ABL affected by submeso motions, and on the simulation of some case studies.

The candidate will deal with the ensemble averaged equations, with the various decompositions for specific cases.

Data analysis will provide suitable test of the hypoteses.

NWF MOLOCH model will be used to simulate some (one) cae study, chosen from the investigated data set, in order to understand which level of representation of the  turbulence is present in the model and how far the parameterization can be ‘pushed’ to deal with submeso motions.



- Openness to research on new subjetcs and to work interactions.

- Basic computational skills, including programming languages like Fortran and/or C.

- (Preferred) Use of LaTeX as text editor.


The work will be developed in cooperation with Dr. Oxana Drofa and Dr. Francesco Tampieri, at the Institute for Atmospheric and Climate Science (ISAC) of the National Research Council (CNR) in Bologna.

Proposte di tesi in collaborazione con MeteoBlue
Laureandi magistrali in meteorologia ambientale, ingegneria per l'ambiente e il territorio, ingegneria energetica, fisica. 
martedì 15 giugno 2021

​4000 weather stations worldwide.

• generate Multi-model calculations from a selected range of models based on 2 years of data. (Alternative: received Multi-model calculations from meteoblue for the 4000 weather stations)

• "train " 2 historic models (NEMS30, ERA5) to recalculate temperature based on the multi-model data. The training can be done with 2-3 different methods.

• evaluate the accuracy of the training data on these 4000 weather stations;

• extrapolate the training to the last 10 years.

• evaluate the accuracy of the training data on these 4000 weather stations for the past 10 years.

Bias correction;
Event detection
Risk analysis. 
1.3. CONTACT for data: meteoblue

1.4. SCOPE: could be bachelor if all data provided by meteoblue and only one training method and accuracy needed to be done .


2. Development of methods to backward engineer historical WIND simulation and reanalysis data based on 1 year of hourly and daily measurement data from 4000 weather stations worldwide.

2.1. METHODOLOGY :Same process as with 1.


• Detection of missing values

• Detection of Bias

• Detection of other Errors (snow melt, oscillation , position change)

2.3. CONTACT for data: meteoblue

2.4. SCOPE: could be bachelor if all data provided by meteoblue and only one training method and accuracy needed to be done .

3. Development of automated routines for testing UHI effect of different city planning scenario variations.

3.1. METHODOLOGY :Same process as with 1.


• Detection of missing values

• Detection of Bias

• Detection of other Errors (snow melt, oscillation , position change)

3.3. CONTACT for data: meteoblue

3.4. SCOPE: could be bachelor if all data provided by meteoblue and only one training method and accuracy needed to be done . SINCE precipitation is more complex, this rather seems a Master Thesis.

4. Influence of atmospheric parameters on visual acuity (SEEING) at several astronomic telescope locations 


4.1.0. Use existing Astronomy seeing data from 1 telescopes. 

4.1.1. Factor analysis Examine all individual factors Investigate multi-factorial relationships:
4.1.2. Method analysis. compare different methods. define best practices.
4.1.3. Select analysis on selected locations.
4.1.3. Find 10-30 locations with "Seeing” data
4.1.3. Download data

4.1.4. Analysis for other locations
4.1.5. Identify the main factors.
4.1.6. Applications
4.1.7. Conclusions. 

• Correlation
• Detection of Bias
• others. To be determined.
4.3. CONTACT for data: meteoblue

meteoblue has stored 2 years of atmospheric data for 30 telescope locations
4.4. SCOPE: could be bachelor if all data provided by meteoblue and only one training method and location needed to be done . SINCE 30 location  is more complex, this rather seems a Master Thesis.