- Joined
- Jul 24, 2008
- Messages
- 33,627
- Points
- 0
[h=2]Vivian uses misleading statistics to cover himself[/h]
July 29th, 2012 |
Author: Ng Kok Lim
Minister for Environment and Water Resources, Vivian Balakrishnan
Dear Dr Balakrishnan,
I refer to the 23 Jul 2012 Straits Times report of your comments on Orchard Road flood prevention measures. You said those were long-term measures, planned decades ahead and that we always have a long-term view of things:
It was also reported that the expert panel found that our rainfall has increased by about 30% over the past four decades.
The following annual rainfall data was obtained from the government website [1].
<TBODY>
</TBODY>It shows that we have five decades of rainfall data but the expert panel only used four. Plotting the data on a graph below:
We see that rainfall has been fluctuating up and down over the last 52 years. If our period of consideration started on a trough year and ended on a peak year, we would more likely conclude that rainfall has increased. But if our period of consideration started on a peak year and ended on a trough year, we would more likely conclude that rainfall has decreased instead. Since our conclusion depends very much on which years we choose to compare, comparing actual rainfall between years is not a satisfactory way of deciding how much rainfall has increased or decreased over time.
Referring to the graph above, we should instead consider the trend-line over the entire period which shows a 3.4 mm increase each year or an 11% increase over 52 years, far less than the reported 30% increase over four decades. This 3.4 mm increase per year is gradual and doesn’t explain why Orchard Road was suddenly innundated by floods in 2010 and not during the previous peak in 2007. More importantly, the p-value of this 3.4 mm yearly increase is 0.4. In other words, there is 40% chance that this 3.4 mm increase per year is just a random result. In other words, the result is too highly uncertain for us to be reasonably confident to conclude that rainfall has increased over the last four decades, let alone increased by 30%.
Please correct the expert panel and also inform all Singaporeans of their mistaken conclusion that continues to be propagated till this day.
Thank you.
.
Ng Kok Lim
[1] http://data.gov.sg/
.
Dear Dr Balakrishnan,
I refer to the 23 Jul 2012 Straits Times report of your comments on Orchard Road flood prevention measures. You said those were long-term measures, planned decades ahead and that we always have a long-term view of things:
Extract from Straits Times, 23 Jul 2012, “Orchard Road anti-flood measures to meet future needs”:
“Dr Balakrishnan stressed that these were long-term measures, planned for decades ahead.
‘That’s the way Singapore has to be. We always have to have a long- term view on things.’
The country’s rainfall has increased by about 30 per cent in the past four decades, according to the panel’s findings.“
If we planned decades ahead and viewed things in the long-term, how come we were caught flat footed in 2010 and 2011? In between plans that supposedly last a few decades, do we operate on auto-pilot mode?“Dr Balakrishnan stressed that these were long-term measures, planned for decades ahead.
‘That’s the way Singapore has to be. We always have to have a long- term view on things.’
The country’s rainfall has increased by about 30 per cent in the past four decades, according to the panel’s findings.“
It was also reported that the expert panel found that our rainfall has increased by about 30% over the past four decades.
The following annual rainfall data was obtained from the government website [1].
| 1960 | 1961 | 1962 | 1963 | 1964 | 1965 | 1966 | 1967 | 1968 | 1969 |
| 1563 | 1812 | 2287 | 1819 | 2826 | 1853 | 2480 | 2910 | 2077 | 2289 |
| 1970 | 1971 | 1972 | 1973 | 1974 | 1975 | 1976 | 1977 | 1978 | 1979 |
| 2277 | 1608 | 1807 | 2956 | 2066 | 1925 | 2167 | 1775 | 2766 | 2168 |
| 1980 | 1981 | 1982 | 1983 | 1984 | 1985 | 1986 | 1987 | 1988 | 1989 |
| 2326 | 1463 | 1582 | 1994 | 2687 | 1484 | 2536 | 2103 | 2599 | 2463 |
| 1990 | 1991 | 1992 | 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 |
| 1524 | 1877 | 2261 | 2169 | 1942 | 2333 | 2418 | 1119 | 2623 | 2134 |
| 2000 | 2001 | 2002 | 2003 | 2004 | 2005 | 2006 | 2007 | 2008 | 2009 |
| 2371 | 2783 | 1749 | 2391 | 2136 | 1931 | 2753 | 2886 | 2325 | 1921 |
| 2010 | 2011 | ||||||||
| 2075 | 2524 |
<TBODY>
</TBODY>
We see that rainfall has been fluctuating up and down over the last 52 years. If our period of consideration started on a trough year and ended on a peak year, we would more likely conclude that rainfall has increased. But if our period of consideration started on a peak year and ended on a trough year, we would more likely conclude that rainfall has decreased instead. Since our conclusion depends very much on which years we choose to compare, comparing actual rainfall between years is not a satisfactory way of deciding how much rainfall has increased or decreased over time.
Referring to the graph above, we should instead consider the trend-line over the entire period which shows a 3.4 mm increase each year or an 11% increase over 52 years, far less than the reported 30% increase over four decades. This 3.4 mm increase per year is gradual and doesn’t explain why Orchard Road was suddenly innundated by floods in 2010 and not during the previous peak in 2007. More importantly, the p-value of this 3.4 mm yearly increase is 0.4. In other words, there is 40% chance that this 3.4 mm increase per year is just a random result. In other words, the result is too highly uncertain for us to be reasonably confident to conclude that rainfall has increased over the last four decades, let alone increased by 30%.
Please correct the expert panel and also inform all Singaporeans of their mistaken conclusion that continues to be propagated till this day.
Thank you.
.
Ng Kok Lim
[1] http://data.gov.sg/
.
