SIMEPAR - SINALMET

A HTML based cloud-platform for digital signage of weather information

1. A DIGITAL SIGNAGE SOLUTION FOR SEVERE
WEATHER MONITORING AND FORECASTING
Fábio Sato - Head of IT



[email protected]

1
/simeparparana@simeparpr

2. ABOUT US
2
We are located at Curitiba in the
south of Brazil
We are a self-sustained private non-profit
organization responsible for the operation of the
meteorological and hydrological system of
Paraná state in Brazil

3. INFRASTRUCTURE - STATIONS
3
•Network of 110 automatic
stations: 43 meteorological, 51
hydrological, 16 rain gauges
•Cooperation/Partnership
agreements with many Brazilian and
Latin American institutions for data
exchange and technical cooperation
•We receive and process online
data from more than 3,000
stations

4. INFRASTRUCTURE - RADARS
4
S-Band Single Polarization

Location: Teixeira Soares/PR!
Start of Operation: 1997
S-Band Dual Polarization

Location: Cascavel/PR!
Start of Operation: 2014
•Two doppler S-Band
Doppler radars at Paraná!
!
•Operation, calibration and
maintenance done by our
staff!
!
•Integration and data
exchange with other
institutions in Brazil,
Paraguay and Argentina.!
!

5. INFRASTRUCTURE - LIGHTNING
5
•RINDAT: Brazilian
Lightning Detection
Network!
•35 operational sensors!
•Total Lightning
network on the
western region of PR
state!
!
!

6. ACTIVITIES AND SERVICES
6
Development Operations
METEOROLOGY!
!
HYDROLOGY!
!
REMOTE SENSING!
!
AGRICULTURE!
!
COMPUTER SCIENCE
SOFTWARE DEVELOPMENT
WEATHER AND
HYDROLOGY
MONITORING AND
FORECAST!
!
MAINTENANCE AND
OPERATION OF
MONITORING NETWORKS!
!
IT INFRASTRUCTURE
Research

7. CUSTOMERS
7
Power Companies
Water supply
Hydro-Met Services
Industrial Plants
Construction
Government Agencies
Road Concessionaries
Media
Oil Refineries/Plants

8. SOFTWARE DEVELOPMENT 

EXPERIENCE
Data Visualization
Climate and Hydrology 

Data Management Systems
Weather Forecasting 

and Nowcasting
Early Warning Systems

9. SINALMET
9
Each operation center had
different requirements about
information and area of interest
The problem we were trying to solve
Electric power distribution
company wanted to display our
webpages and applications on
large TV displays deployed at
their operation centers
5 operation centers
geographically distributed
across the state!

10. SINALMET
10
Our webpages and applications were
not appropriate for this purpose!
!
Hard to read texts and to interpret
images from distance!
!
We searched for solutions and didn’t
find specific software for this scenario

TV broadcast solutions were too
expensive
Our initial approach

11. SINALMET
11
\
Our initial approach
We decided to buy and use a general purpose digital signage
solution and setup it to disseminate our content

12. SINALMET
12
We setup 2 large displays (LED
TVs) for each operation center!
!
1 display for applications and the
other for digital signage solution
client!
Our initial approach
•Small computer (Mac Mini) attached to back of the
TVs!
•Computer connected to an extranet of the
customer!
•Content dissemination via dedicated link between

13. INITIAL SOLUTION
13
Digital Signage client software requires computer
reboot sometimes (manufacturers usually
recommends daily reboots)
Technical Problems
Difficult to interact with the data visualization
applications that were not designed for distant
interaction!
Could not be used on wall screens or displays
because of corporate IT policies and security
requirements

14. INITIAL SOLUTION
14
It was hard to scale
!
Every new installation required the
acquisition of dedicated displays,
computer, software licenses, local
installation!
!
Deployment barriers: different
corporate IT policies!
!
New installations are expensive and we
costs don’t scale down

15. INITIAL SOLUTION
15
Technical support was a pain.
!
We used VNC and SSH to remotely control
computers and software, but sometimes we
couldn’t reach the computer!
!
Every new computer had to be continuously
monitored!
!
Slow connections to distant control centers!
!

16. SINALMET
16
Should we develop our own solution for this?
Avoid "not invented here syndrome”: we did new research but we still were not happy with
the solutions we found!
!
We believed that it would be possible to develop a HTML based system for a "Weather Digital
Signage" solution!
!
We developed a prototype in less than 3 weeks!
!
The HTML-based prototype was able to provide better functionality than our original
approach. !
!
We decided that it was worth to develop it

17. SINALMET
17
!
What is SINALMET?
It is a digital platform for the dissemination on large displays of weather,
hydrology and environmental information
It runs entirely on the web browser!
!
It uses new technologies available on HTML5
and CSS3 compatible web browsers!
!
It is scalable and low cost!
!
It doesn’t require dedicated hardware: use
existing computer and low-cost solutions like
Chromecast!
!
It can comply with corporate IT policies!
!
It requires almost zero tech support

18. SINALMET
18
How did we developed it?!
!
HTML5 + CSS + JS

Web formats: XML and JSON!
!
OGC web standards: WMS and WFS!
!
PostgreSQL + PostGIS

Unidata THREDDS!
!
Geoserver and Mapserver!
!
RESTFul APIs: systems integration!
!
Javascript-based data visualization libraries!
!

19. SINALMET - RADAR
19
Assimilates raw volume and product files!
Generates basic radar products: PPI, CAPPI and mosaics/composites!
Integration with lightning data

20. SINALMET - LIGHTNING
20
!
Displays data from a Vaisala lightning processing central!
by connecting to a live UALF feed over Internet for realtime lightning data access

21. SINALMET- SATELLITE
21
Displays infrared channels of GOES or METEOSAT satellites!
Radar and lightning can be overlaid to highlight storm cells

22. SINALMET - FORECAST
22
Presentation of forecast data available in XML or JSON formats

23. SINALMET - STATIONS
23
Automatic assimilation and interpolation of weathers stations data

24. SINALMET - WARNING
24
Dissemination of Early Warnings

25. SINALMET - DEMO
25

26. SINALMET - ROADMAP
26
So, what’s next?
Mobile support (smartphones and tablets)!


Packaging for smart TVs stores!
!
Integration with nowcast systems, e.g.: TITAN, lightning
storm tracking, and satellite based algorithms!
!
Dissemination of CAP (Common Alerting Protocol)
messages!
!
Social media: filter and display relevant information
with natural language and machine learning
algorithms!
!

27. SINALMET
27
QUESTIONS
Visit www.simepar.br/portfolio for more information