Geoscience Reference
In-Depth Information
unpublished. This is partly a recognition of the fact that the assembly of statistical
data is a costly and time-consuming process, so that the independent generation of
an alternative set of regional data for a particular analysis is usually out of the
question. It is also due to the investigator being sufficiently anxious to proceed with
the analysis in question that the validity of the regionalization scheme becomes a
secondary concern. However, because of a myriad of factors, ranging from histo-
rical decisions to politically-motivated pressure, there can be no assurance that
official regions represent a meaningful set of functional economic units.
It is against this general background that we wish to propose a further basis for
determining the extent of a region. Our approach relies on the density function,
which describes the decrease in population density with the distance from a central
point. The concept of the density function, introduced by Clark (
1951
), has proven
valuable in our understanding of urban spatial structure. The density function has
also been applied at the regional level, though less commonly (Barkley et al.
1996
;
Nairn and O'Neill
1988
;Parr
1985
; Wang
2001
). The discussion to follow has two
broad purposes: first, to consider the nature and form of the regional density
function; and second to demonstrate the manner in which it may be employed in
the determination of regional boundaries. Unlike most applications of the regional
density function, no a priori definition of the boundary is assumed. Rather, the
regional boundary is shown to result from a particular application of the density
function. The next section outlines certain key elements of the approach.
6.2
The Economic Region and the Regional Density Function
Although not all writers are comfortable with the notion of an economic region
(Wren
2009
), it continues to represent a frequently-employed unit of analysis. The
significance of the economic region has been discussed in some detail by various
authors, notably Isard (
1956a
), L
ยจ
sch (
1944
/1954, pp. 103-137) and Teitz (
1962
).
6.2.1 The Nodal Region
There are, of course, a number of valid definitions of the economic region (Meyer
1963
), but attention will be confined here to perhaps the most common one, the
nodal region, a region that also lends itself to analysis in terms of the density
function. The nodal region has been widely used in regional analysis (Brown and
Holmes
1982
; Nystuen and Dacey
1961
), as well as in numerous public-policy
interventions. The nodal region is considered here at the scale of a territory
dominated economically by a major metropolitan area having a population above
some minimum level. Such a nodal region thus comprises a metropolitan part or
node, and also a more extensive non-metropolitan part, representing the hinterland.
This latter part contains a rural population as well as an (often much larger) urban
population, located in a hierarchical network of centers. A nodal region of this scale
