How we calculate the BSI (Batgung Smog Index)

  • strict warning: Non-static method view::load() should not be called statically in /nfs/c03/h02/mnt/51424/domains/batgung.com/html/drupaldir/sites/all/modules/views/views.module on line 879.
  • strict warning: Declaration of content_handler_field::options() should be compatible with views_object::options() in /nfs/c03/h02/mnt/51424/domains/batgung.com/html/drupaldir/sites/all/modules/cck/includes/views/handlers/content_handler_field.inc on line 208.
  • strict warning: Declaration of date_handler_field_multiple::pre_render() should be compatible with content_handler_field_multiple::pre_render($values) in /nfs/c03/h02/mnt/51424/domains/batgung.com/html/drupaldir/sites/all/modules/date/date/date_handler_field_multiple.inc on line 185.
  • strict warning: Declaration of views_handler_argument::init() should be compatible with views_handler::init(&$view, $options) in /nfs/c03/h02/mnt/51424/domains/batgung.com/html/drupaldir/sites/all/modules/views/handlers/views_handler_argument.inc on line 745.
  • strict warning: Declaration of views_handler_filter::options_validate() should be compatible with views_handler::options_validate($form, &$form_state) in /nfs/c03/h02/mnt/51424/domains/batgung.com/html/drupaldir/sites/all/modules/views/handlers/views_handler_filter.inc on line 589.
  • strict warning: Declaration of views_handler_filter::options_submit() should be compatible with views_handler::options_submit($form, &$form_state) in /nfs/c03/h02/mnt/51424/domains/batgung.com/html/drupaldir/sites/all/modules/views/handlers/views_handler_filter.inc on line 589.
  • strict warning: Declaration of views_plugin_row::options_validate() should be compatible with views_plugin::options_validate(&$form, &$form_state) in /nfs/c03/h02/mnt/51424/domains/batgung.com/html/drupaldir/sites/all/modules/views/plugins/views_plugin_row.inc on line 135.
  • strict warning: Declaration of views_plugin_row::options_submit() should be compatible with views_plugin::options_submit(&$form, &$form_state) in /nfs/c03/h02/mnt/51424/domains/batgung.com/html/drupaldir/sites/all/modules/views/plugins/views_plugin_row.inc on line 135.

The idea behind the Batgung Smog Index (BSI) is to give a simple measure of whether air pollution in Hong Kong is getting better or worse over time.

It's based on the observation that when the air is smoggy, visibility is low and pollution levels are high.

Fortunately for us, the Hong Kong Observatory publishes monthly records of 'Number of hours of Reduced Visibility observed at the Hong Kong Observatory since 1968' [1]. They define 'reduced visibility' as 'visibility below 8 kilometres when there is no fog, mist, or precipitation'. So if visibility is reduced because there's a heavy rainstorm for example, that isn't counted in these figures.

If it's not fog, mist or rain causing reduced visibility, we can assume it is caused by smoggy air. So now we have a measure of how many hours the air is smoggy each year. Here's a graph covering the period Dec 1968 – April 2008. For each month, it shows the total number of hours of low visibility for the year-to-date.

YTD Hours of Reduced visibility

The figure is relatively steady in the 1970's and 80's, but from the start of the 1990's it starts its climb upwards.

The lowest value was for the period Aug 1973 - Jul 1974, with just 140 hours of poor visibility (dirty air) recorded. The worst period so far was Feb 2004 – Jan 2005, when 1,702 hours of poor visibility were recorded!

Most recently, in the period May 2007 – Apr 2008 there were 1,448 hours of dirty air recorded. Better than the worse, but still a long way away from the best.

You can see that these figures swing up and down a lot, so it can be difficult to identify what is a short term change, and what is a longer trend. For instance in 2005 the government claimed the fall in the amount of hours of dirty air recorded meant their policies to reduce pollution were working. The local environmental groups pushed back that the reduction was just a result of higher than average rainfall.

So to smooth out some of the short-term changes, for the BSI I've chosen to use a three-year average of the annual values. The red line shows the effect of the smoothing:

YTD Hours of reduced visibility and smoothed figures

The disadvantage is that it will be slower to show changes than the 'raw' line, and will also hide the worst extremes. Still, as a way to measure the long-term trend, it seems a reasonable compromise.

Interestingly, it looks as though the government may have been right, and the pollution levels have started falling. Time will tell if we're on a long-term downward (improving) trend, or it is a temporary reversal like we saw around 1995 and 2001.

Finally, to make the index a more manageable number, I've scaled all the numbers so that the value of the BSI for October 1989 is 100:

Batgung Smog Index

Why October 1989? Because that's when I first arrived in Hong Kong, and I'd like my blue skies back! The most recent value for the index is 436 for April 2008, so we've got a long way to go.

MrB

References:
[1] HK Observatory's Number of hours of Reduced Visibility observed at the Hong Kong Observatory since 1968

Comments

Govt Policy at its most cynical

Great stuff. Don't forget the nice long period (last year - 2007) of exceptionally clear skies due to the consistant easterlies blowing throughout most of July and August. The Govt was quick to try and take credit for that one as well *sigh*

Reduced Visibility

"the Hong Kong Observatory publishes monthly records of 'Number of hours of Reduced Visibility observed at the Hong Kong Observatory since 1968'"

Why are some monthly data points have values as many as 1700 hours?
Assuming they continue to, and have ways to take readings after sun-set, then a month has either 672 hours (February for example) or as many as 744 hours (January, for example). There may be a simple explanation to satisfy my curiosity or correct my misunderstanding.

I am interested in knowing what method(s) they have been using through these years to determine reduced visibility. But regardless, the picture looks gloomy, and is steadily getting worse.

I agree that one should not disgard the data points on rainy day, as the pollution on these days could be just as bad or worse. This can be shown by line on the same graph.

re: Reduced Visibility

Hi OldTimer, 

Why are some monthly data points have values as many as 1700 hours?

If you look back at the original data, HK Observatory's Number of hours of Reduced Visibility observed at the Hong Kong Observatory since 1968, you'll see that the highest ever monthly reading was 350 hours.

But since the air pollution varies with the seasons, if we plotted those figures we'd seem to get better each summer and worse each winter. I wanted a figure that would show longer trends, so the blue line on the first graph above "shows the total number of hours of low visibility for the year-to-date." eg the value for Jan 2008 is the sum of all the Observatory's readings from Feb 2007 - Jan 2008.

Regards, MrB

Reduced Visibility

MrB: "so the blue line on the first graph above "shows the total number of hours of low visibility for the year-to-date." I first took that the graph plots monthly values. I missed reading this critical detail. Thank you.
OldTimer