Page 1
PM2.5 Interpolation andVisualization based on Wind Fields
ked2016/05/152016/06/13 update
Page 2
Our system
PM2.5 http://www.cmlab.csie.ntu.edu.tw/~ked/pm25.htm
PM2.5 + radar http://www.cmlab.csie.ntu.edu.tw/~ked/
pm25_4.htm
Page 3
Why a different PM2.5 vis. system
2016/04/12
Page 4
The visualization of
PSI result
2016/04/12
Page 5
The visualization of
2016/04/20
Page 6
The visualization of
PSI result Averaged index
2016/04/12
Page 7
The visualization of
PSI result Averaged index Sparse distribution
2016/04/12
Page 8
Other systems for PM2.5
http://env.g0v.tw/air/
2016/04/15
Page 9
Other systems for PM2.5
http://env.g0v.tw/air/ http://env.childgrowth.info/air/
2016/04/15
Page 10
Other systems for PM2.5
http://env.childgrowth.info/air/
天下雜誌 2015.03.17
Page 11
How to interpolate the sparse data
Page 12
Wind & PM2.5
天下雜誌 2015.03.17
Page 13
Wind & PM2.5
東森新聞 2016.01.01
Page 14
Our system
2016/04/15
Page 15
Our system
2016/04/22
Page 16
Outline
System description Data capture Interpolation Rendering Conclusions
Page 17
Outline
System description Data capture Interpolation Rendering Conclusions
Page 18
Features
Interpolation based on winds x 3D interaction
x Auto-renewal
Page 19
Architecture
Our server
環保署
NCEP
Capture PM 2.5
Interpolate
Capture wind field & height field
Page 20
Architecture
Web browserOur server
環保署
NCEP
Capture PM 2.5
Interpolate
Get data
Visualize &update
Capture wind field & height field
Page 21
Outline
System description Data capture Interpolation Rendering Conclusions
Page 22
PM2.5 data
Location ?
Page 23
PM2.5 data + coordinate
Page 25
.grib2 format – records
Page 26
How to read .grib2 file
Refer to https://www.google.com.tw/url?
sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwjCvc7I2Y3MAhVKGJQKHVC4AxgQFgglMAA&url=http%3A%2F%2Fjou-p3.as.ntu.edu.tw%2FP3%2Ffile%2Fteaching%2Ff1386923093.pdf&usg=AFQjCNF0Z5dLY08WxC5pl_0BUxfTvxUvuQ&sig2=eLDe7X0wv0Ol1KG_Dnr9YQ
Page 27
.grib2 format – regular data
https://www.google.com.tw/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwjCvc7I2Y3MAhVKGJQKHVC4AxgQFgglMAA&url=http%3A%2F%2Fjou-p3.as.ntu.edu.tw%2FP3%2Ffile%2Fteaching%2Ff1386923093.pdf&usg=AFQjCNF0Z5dLY08WxC5pl_0BUxfTvxUvuQ&sig2=eLDe7X0wv0Ol1KG_Dnr9YQ
Page 28
Outline
System description Data capture Interpolation Rendering Conclusions
Page 32
Flow in the field
Page 33
Flow in the field
Page 34
Transition
1/3
5/9
1/9
Page 35
Transition – intensity decay
1/3 x 1/2
5/9 x 1/2
1/9 x 1/2
1/2
Page 36
Transition matrix
Refers to the stochastic process in which the transition probabilities depend only on the current state
1/3 x 1/2
5/9 x 1/2
1/9 x 1/2
1/2
Page 37
Irreducible, aperiodic, recurrent
Stationary Markov chain
Page 38
Stationary Markov chain
Page 39
Outline
System description Data capture Interpolation Rendering Conclusions
Page 40
Implementation
Javascript + webgl
Page 41
Rendering
Taiwan
PM2.5
Wind vector
Page 42
Scientific visualization
"it’s infovis when the spatial representation is chosen, and it’s scivis when the spatial representation is given“
-- wikipedia
Page 45
Billboard
http://www.markmark.net/clouds/
Page 46
Outline
System description Data capture Interpolation Rendering Conclusions
Page 47
Case – 2016.03.14 " 在臺灣西南側一條繩索狀的白雲,那是南海吹過來微弱西南風和東側高壓迴流的交界面,因為大氣穩定,所以只在海面上形成一長條的索狀雲。 "
Page 48
Case – 2016.04.08
Page 49
Thx.
站在推廣環保和教育的立場,我從電腦圖學的角度出發,做了一個互動系統,希望民眾可以重視 PM2.5 的問題,更希望的是,我們的下一代,可以因為我們的覺醒,而呼吸到更好的空氣。
Page 50
Update Streamlines for wind field Weather radar for rainfall
Anaglyph 3D
Page 51
Streamline placement
Refer to
Page 52
Weather radar data
Download from