Environmental Engineering Reference
In-Depth Information
Next, it is important to assess which of the factors
discussed in the previous section are likely to slow the
initial phase of recovery and document the rate of
natural recovery, in order to most effi ciently use the
limited resources available for the large expanses of
degraded tropical lands that should be restored. The
likely rate of natural recovery rate can be assessed by
understanding the ecology of the system, observing
natural recovery in the area to be restored and neigh-
bouring sites, and reviewing similar cases from else-
where, as reported in the literature.
A fi rst question to ask is whether woody species, in
particular those typical of early succession in the
target area, will either resprout in or colonize the site
naturally. There are a growing number of examples of
tropical secondary forests that have recovered similar
biomass, and in some cases similar species composition
- as compared to nearby
primary forests
- within a
few decades (e.g. Dent & Wright 2009; Letcher &
Chazdon 2009). For example, in Puerto Rico, forest
cover has increased from <10% in the 1930s to > 40%
by the year 2000, and forest biomass and species rich-
ness in these regenerating forests are similar to those
of mature forests (Aide
et al
. 2000). If regeneration is
likely to be high, then lands should be left to recover
naturally. Not only is this approach less resource inten-
sive, but also, in cases where there is extensive natural
regeneration, planting trees can either slow recovery
(Sampaio
et al
. 2007) or strongly infl uence the
succes-
sional trajectory
(Murcia 1997; Celentano
et al
.
2011 ).
If recovery is slow, the next question to ask is whether
there is a residual factor affecting seedling establish-
ment that, if removed, would accelerate the rate
of natural recovery (Figure 9.3). If there is suffi cient
colonization through seed dispersal and from sources
within the site, then it may not be necessary to actively
introduce seeds or seedlings, which is usually time con-
suming and expensive. For example, large-scale tropi-
cal dry forest restoration efforts in north-western Costa
Rica have focused on controlling fi re during the dry
season, which has been largely successful in allowing
for natural recovery (Janzen 2002). This approach
of removing barriers to establishment is sometimes
referred to as an
assisted natural regeneration
(Dugan
et al
. 2003 ; Shono
et al
. 2007 ). Shono
et al
.
(2007) discuss an approach for restoring
Imperata
cylindrica -
dominated grasslands in Asia, where they
mark all woody regeneration and then clear surround-
ing grasses to reduce competition and fi re risk. They
also clear fi re breaks around the sites and do follow-up
with enrichment planting of specifi c target species.
These efforts to remove obstacles to natural regenera-
tion without actively planting or seeding are usually
much cheaper and leave less of a human imprint on
the long-term species composition than planting and
maintaining tree seedlings from the outset (Lamb
et al
. 2005), which can strongly affect the successional
trajectory.
If colonization of woody species is limited by lack of
propagules
, due to low seed dispersal and sparse
regeneration within the site, then it is necessary to
intervene to introduce woody species (Figure 9.3). This
is done most commonly by planting tree species and, in
some cases, by direct seeding or creating more favour-
able habitat for seed-dispersing animals. Different
approaches for introducing forest plant species are dis-
cussed in more detail in the following section.
As the naturally regenerating or actively restored
forest matures, one must consider whether the full
complement of later successional plant species recolo-
nize over time (Figure 9.3). Many studies show that
tropical
secondary forests
ranging in age from 30 to
60 years tend to lack large-seeded, mature forest
species (Mart í nez - Garza & Howe 2003 ; Zimmerman
et al
. 2007); this is likely due to dispersal limitation,
which is exacerbated in locations where large seed dis-
persers have been overhunted (Wright 2003). There-
fore, numerous authors have suggested that for former
agricultural lands where pioneer species rapidly colo-
nize, the most effective restoration strategy is to plant
selected mature forest species that do not readily rec-
olonize (Mart í nez - Garza & Howe 2003 ; Lamb
et al
.
2005). In some cases, even if forest species are intro-
duced in the early stages of restoration, it may be nec-
essary to plant later successional species once site
conditions are more favourable (discussed in more
detail below; Lamb
et al
. 2005 ; Kettle 2010 ). Some
later successional species have shown high survival in
recently abandoned pasture (Loik & Holl 1999; Hooper
et al
. 2002), but more research is needed on when in
the successional process these species should be intro-
duced to maximize survival and growth.
9.4 STRATEGIES FOR ACTIVELY
REINTRODUCING PLANT SPECIES
As noted earlier, if some or all native forest species
do not colonize the site within a time frame that is
