Mike,

That’s a very good hypothesis – it makes complete sense of what we found in the London study where the pedestrian model (no traffic management) correlated better with flows than the model with oxford street cut to vehicular movement. My analysis of this was that what we were in fact observing was a long term historical evolution of land uses and densities according to the un-traffic managed space structure. Now, in the case of Kowloon, I am not so sure that it is as evenly distributed in terms of density and use as you suggest, certainly at the large scale, and I suspect that there may be issues to do with data consistency, mapping and size of the map all of which have been raised, and should be checked for. In particular, at about 5pm every evening a whole series of streets get cut to traffic by the setting up of street markets. These produce crowds of pedestrians and severely disrupt neighbouring streets. One could see these as both attractor destinations – I am sure that some people drive to get to them, and as blockages in the space structure to vehicular movement. Given this kind of complexity I am not surprised that a first cut analysis doesn’t show a correlation, however you have provided a very nice working hypothesis for Chengke to test. I wonder whether data on land use and development density are available?

Alan

You know, I think the problem with this is that there is very little spatial autocorrelation in HK. This is because of the density of the place and because the traffic pattern is not radial like organically growing western cities that grow from a core, usually.

In western cities there is a lot of spatial autocorrelation which means that more central points in the city are naturally the most accessible like the centre of a circle and therefore there is more traffic travelling there due to more people there. Some would argue that this shows that any such correlation with traffic going to the centre and accessibility is spurious and it is true that it doesn't explain much at all, forthe centre of any bounded space is the most accessible by definition.

In fact in a lot of spatial analysis, accessibility or integration  is factored out before analysis begins - this is like getting rid of spatial autocorrelation  before the modelling begins. We wrote a couple of papers on this in the 1970s and here they are:

1976 (with L. March) The Method of Residues in Urban Modelling, Environment and Planning A, 8, 189-214.

1978 (with L. March) Urban Models based on Information-Minimising, in R. L. Martin, R. J. Bennett and N. J. Thrift (Editors) Towards Dynamic Analysis of Spatial Systems (Eds., Pion Press, London, 127-155.

There is a good book on this - in fact the original book by Andy Cliff and Keith Ord (1981) Spatial Processes: Models and Applications, Pion  which tells you why this effect is important but there are a lot of more intelligible primers such as Spatial Regression Analysis on the PC: Spatial Statistics Using Sas (Resource Publications in Geography) by Daniel A. Griffith, Assn of Amer Geographers (Nov 1993).

So all this means that what one might do is once one has computed the integration accessibility values then use these to explain how traffic patterns differ from these values and hence all the explanation is heaped on the residuals. So in a sense in HK this is less of a problem and it means that to explain traffic we need new independent variables such as density, land use type, trip demand and so on.

If you have a density of population that is not declining inversely with distance from the centre which is probably the case in places like HK and also the place is hemmed in by mountains - Lion Rock to the north of Kwowloon and the Peak to the south of Victoria, then it must be almost impossible to satisfy the conditions for high accessibility at the centre in terms of just the street geometry.

Mike

 At 12:03 22/01/2007 +0000, Alain Chiaradia wrote:

Dear Chengke,


To add to Alan comments:

a/ What kind of road counts do you use? Are they mainly expressways, trunks road, main roads? If so try global measure. Traffic flow level is also dependant on capacity (useable width or lane number) so this may impact on your analysis.

b/ if there are global movement roads it might be that your model coverage is too small. Large scale movement require a regional model size.

________________________________________

Alain Chiaradia
________________________________________


-----Original Message-----
From: Chengke LIU
Sent: 21 January 2007 10:18
To:
Subject: Poor correlation between local integration and traffic flow in Kowloon, Hong Kong

Deal all,

I have finished drawing axial lines of one district of Hong Kong - Kowloon,
totally 3004 axial lines. Also I pin-pointed all the counting stations on
those axial lines. I calculated local integration of axial lines and
compared it with the natural logarithm of the average traffic flow of that
corresponding axial line. To my disappointment, the correlation is quite
poor, r square value is only 0.10. I am not sure why the correlation is so
poor, the possible reasons are listed as follows:

1.The street network in Hong Kong is very complex, I digitized some road
lines with only one axial lines, since they represents different lanes for
different traffic direction. And even, somewhere a cluster of roads are
represented for only one axial line for simplification.
2.Due to lacking of time and manpower, I did not take those overpass and
underpass roads into consideration when I calculated the connectivity value
of axial lines. When I digitized the axial map, I only referred to the
digital street and building maps of Hong Kong at hand. Some lines are
apparently intersected on digital map, but actually they are not, since one
of them is overpass or underpass road.
3.When I assigned the traffic flow the axial lines, I first pin-pointed
counting stations on roads to corresponding axial line, and then averaged
the traffic flow.
4.The spatial configuration (space syntax) is not suitable in Hong Kong to
predict traffic flow.

The above factors are responsible for the poor correlation from my point of
view. What do you think about this problem? Since Kowloon is only one
district of Hong Kong, I need to continue with the whole Hong Kong
territory. If something wrong in above 4 factors, please point it out, so I
can correct it early. Your comments are highly appreciated!

Michael Batty Director Centre for Advanced Spatial Analysis (CASA)
University College London - 1-19 Torrington Place - London - WC1E 6BT UK
________________________________________________________________

tel 44 (0) 207 679 1781 New CASA Web Site at http://www.casa.ucl.ac.uk/
Personal Web Page: http://www.casa.ucl.ac.uk/people/MikesPage.htm