•
WORLD AQUACULTURE
•
DECEMBER 2014
45
( C O N T I N U E D O N P A G E 4 6 )
groups. The natural break
points for these data were at
ratios of 1.29, 1.49, 1.89, 2.65,
and 4.02. Jenks Optimization
determined that most farms
had land to water ratios
less than 1.89. Using Jenks
Optimization, a ratio <1.29
can be recommended as a
standard for improved land use
efficiency in pond aquaculture.
Land Area Needed to
Grow Crops for Feed
Ingredients
Based on the mean land
to water surface area ratio of
1.48, the total global land area
devoted to pond aquaculture
farms might be around
24.8 million ha. This area
underestimates the total land
area required for aquaculture
because land is needed to
grow plants as ingredients
included in aquaculture feeds.
The area for plant production
is difficult to calculate because
statistics for aquaculture feed
use are not separated by the
culture system in which they
are used. In addition, not all
species are grown only in
ponds and not all ponds are
supplied with feed.
The total land area
needed for plant meals was
estimated by using average
world yield of crop plants
2
and
typical feed ingredients for
some common aquaculture
species as suggested by Boyd
and Polioudakis (2006); the
average land use was 0.274
ha/t feed. Production per unit
area of water surface area is
reduced in systems that are
more intensive or in which
there is more than one crop per
year. This reasoning does not
apply to land requirements for feed.
The total amount of land needed for aquaculture feed
ingredients can be estimated by making a few assumptions.
According to Alltech’s 2013 Global Feed Summary
3
, 34.4 million t
of aquaculture feed were produced globally in 2011. To estimate the
total amount of feed produced for pond aquaculture, feed production
make precise estimations and
data available to the public are
often a few years behind.
The Ratio of Land Area
toWater Surface Area
A study was conducted
to estimate the total area of
land used globally for pond
aquaculture: the water surface
area of aquaculture ponds plus
land area needed to support
production in ponds. Data were
collected for 100 farms in 26
countries using Google Earth
Pro. Ponds ranged from 0.1 to
26 ha. Ponds were divided into
five categories, with 20 farms in
each category, based on average
pond size for farms: <0.5, 0.5-1,
1-5, 5-10 and >10 ha.
These farms contained
2,783 individual ponds with a
total water surface of 10,923 ha,
placed on 14,091 ha of land, or
1.48 ha of land for each hectare
of water surface. This sample
represents approximately
0.1 percent of the total water
surface area of aquaculture
ponds globally. Despite the
small sample size compared
to the total area of aquaculture
ponds, the results of the current
pond survey should be helpful in
framing land use considerations.
Data were analyzed based
on ratios of farm land area to
water surface area. Farms with
smaller pond size had a greater
average but a more variable land
to water surface area ratio. The
ratio exponentially decreased as
the average pond size on farms
increased. The larger the pond
the less additional land needed
per unit surface area of water.
As average pond size increased,
the change in the ratio with
respect to water surface area
decreased, stabilizing at about 1.25 beyond a pond water surface
area of 3 ha.
The average ratio of total land area to pond water surface
area for all farms was 1.48, with a range of 1.08 to 4.02. Jenks
Optimization (Goodness of Variance Fit) classified breaks in the
data based on distribution of the ratios (Fig. 2). Jenks Optimization
reduces variance within groups while maximizing variance between
TOP.
Farms can be highly concentrated in an area, causing difficulty in discerning
boundaries between adjoining farms through satellite imaging, as at this site in
Malaysia. Screenshot by Lauren Jescovitch.
MIDDLE.
These small-pond farms
in Asia emphasizes the importance of aquaculture in certain regions. Screenshot by
Lauren Jescovitch.
BOTTOM.
This photo illustrates how land can be developed as
an agriculture and aquaculture farm. Photo by Lisa Bott.