use change maps provides an indication of the type
and quality of soil lost between 1984 and 1994.
Table 3 FCC soil productivity limitations
Limit Percent of Area | Area km’
None 93.5 3,660.2
Gley 4.8 187.9
Shallow Soil 0.03 1.2
Steep Slope 13.9 544.1
As noted in the land use change description above,
1.9% of the study area was converted to developed
land uses between 1984 to 1992. Table 4 presents the
characteristics of soils underlying the converted areas.
Of all land converted to developed uses, 90.296 was
land without any fertility limitations. Of the land
limited in some way, soil with a gleying condition was
present in 7.696 of the land converted. Shallow soils
and steep slopes were only minimally present in
converted land at 1.2% and .5%, respectively.
Table 4 1984 - 1992 land use conversion by soil
limitation
Limitation Percent of Area | Area km?
None 9 35:7
Gley 08 3.2
Shallow Soils * 0.5
Steep Slopes * 0.2
* indicates less than .01 96
Between 1992 and 1994, 110.0 km? of land were
converted to developed uses. Of this total conversion,
92.5% of the land had no limitations, approximating
the proportion of loss experienced between 1984 and
1992. Similar to the 1984 to 1992 time period, 7.996
of the land converted had a fertility limitation. Land
with a gleyed condition characterized 6.196 of the
converted area. Shallow soils and steep slopes typified
.2% and .6% of converted land, respectively.
Table 5 1992 - 1994 Land use conversion by soil
limitation
Limitation Percent of Area | Area km
No Limitations 0.9 35.7
Gley 0.2 7.8
Shallow Soils 0.01 0.2
Steep Slopes 0.02 0.7
Two major findings are evident from the results
described above. First, the rate of land conversion to
residential, commercial, industrial and extractive
purposes increased dramatically between 1984 and
1994, accentuating the reduction of total acreage
dedicated to agricultural production. If this is
indicative of a national trend then it could indicate a
problem solved by making the remaining land more
productive or new land must be found to replace the
lost agricultural potential. The rate of loss is even
more troubling as a conservative approach was taken
when interpreting the imagery, probably yielding a
significant underestimation of developed land acreage.
The underestimation would be more significant in the
1994 imagery for two reasons.
First, areas in an image which appeared to reflect the
influence of multiple land uses (a combination of
developed and another land use) were classified as the
non-developed group. With more developed land the
1994 imagery would also have more mixed pixels
producing a greater undercount. Second, only pixels
which were developed at the time of imaging were
classified as developed. Based on this rule there were
significantly large areas in the 1994 image which were
being graded in a checkerboard pattern for road and
housing construction, in which the land between the
graded areas was still classified as agriculture. It is
quite possible that the land present in the middle areas
had been taken out of production already or was soon
to be so. This development pattern was seen only to a
minor extent in the 1992 image and not at all in 1984.
The combination of these factors makes it likely the
rate of change was much greater than indicated by the
analysis conducted here. This situation can only be
exacerbated by the completion of the major highway
seen under construction in the 1994 image, which will
permit easier access to agricultural areas. The more
direct access may encourage more development as
people and commerce look for cheaper places to build.
The second major finding from the results is that
development is occurring primarily on agricultural
land which had been rated as having no major
limitations. Once the conversion occurs, the land is
taken out of production. As more development occurs
the proportion of less desirable land increases and the
productive capacity decreases both due to the
reduction in total acreage, as noted above, and in the
quality of land. The combination of reduction in
quantity and quality land implies the actual loss is
greater than a simple subtraction of acreage from the
available supply of agricultural land.
The impact of the land use changes in the study area
extend beyond the borders defined here. The
agricultural census (Jiangsu Province Statistics
Bureau, 1995) indicate that the study area is in a part
of Jiangsu Province that is at best, only moderately
productive in terms of agricultural output. Areas to
the north are much more productive in the raajor crops
of the province, wheat, cotton and rice. However the
498 International Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998
Im
Jia
Jia
