Effects of Harvesting on Population Structure of Leatherleaf Fern (Rumohra adiantiformis (G. Forst.) Ching) in Brazilian Atlantic Rainforest

2010·6.5 MB·English

by  Cristina Baldauf

of Harvesting on Population
Effects Structure of
Fern [Rumohra
Leatherleaf
adiantiformis
(G. Forst.)
Ching)
in Brazilian Atlantic Rainforest
Baldauf* and Dos
Cristina Mauri'cio Sedrez
Reis
em
Nucleo de Pesquisas Florestas Tropicais- Universidade Federal de Santa Rndnvia
CataririM
Non-timber products (NTFPs) non-wood
forest are biological resources/
which
products, can be obtained from and managed
natural
forests for
consumption NTFPs
and/or
trade (Peters, 1994). are a source of social and
economic
More
benefits, particularly for local or traditional communities.
adequate management regimes number NTFPs
for the exploitation of a great of
should, however, be determined (Shanley et al, 2002). Information about the
and management
autoecology on
effects populations available only
for a
is
number
limited
of species.
economic
Despite the importance around
of several fern species the world,
there is still surprisingly little information on the impacts of frond harvesting
on fern populations (Ticktin et 2007). One of the most important NTFPs in
al.,
Rumohra
Brazil the pteridophyte Ching
adiantiformis
is (G. Forst.) (leatherleaf
fern). This species is of particular economic importance among ferns and
its
* Corresponding
author, e-mail: [email protected].
BALDAUF HARVESTING OF RUMOHRA
& REIS: ADIANTIFORMIS
149
fronds are sold over the world arrangements.
all for floral Leatherleaf fern
fronds are particularly popular thanks to their long post-harvest
life.
In Brazil, the fronds are extracted mainly from the natural environment and
harvested in Atlantic Rain Forest areas in the south and southeast of the
country (Conte et al, 2000; Coelho de Souza, 2003). More than 50% of the R.
adiantiformis traded in Brazil comes from Rio Grande do Sul (Anama,
2002)
and
the major source of income of about 2000 rural households in the
northeastern region of Rio Grande do Sul based on harvesting and trading
is R.
adiantiformis fronds (Coelho de Souza
et al, 2006).
Although
the plants are not collected (only the fronds), frond harvesting may
cause impacts on
R. adiantiformis populations by reducing rhizome growth,
frond density and size, and effective population size (Milton and Moll, 1988;
Geldenhuys and Van Merwe,
der Baldauf
1988;
et 2008). Detailed
al.,
knowledge
about the impacts of the management system used by
local
harvesters
is therefore required to determine conditions of sustainable
The
extraction of this resource. purpose of this study was to evaluate the
demographic managed
structure of populations
of R. adiantiformis in the
south
of Brazil, as well as to verify the sustainability of the harvesting
activity
in this region.
Material and Methods
Study area and sampled populations—Four populations were studied
in
Maquine,
municipality
a in the northern coastal region of the State of Rio
Grande do
Sul, Brazil (Table This area considered the southern
1). is limit of
the distribution of Atlantic Rainforest in Brazil. Populations one, two, and
were managed
three according knowledge, which
to traditional in harvesting
occurs any month maximum
in of the year, with a of three annual collections
permitted same cm
in the area. In general, fronds longer than 30
all are
harvested.
Population was
four evaluated in an area of species occurrence where fronds
are not harvested; this population served as a reference of the population
demographic without which
structure exploitation, rare in the study
is area.
This population is located in a private area with the same ecological
AMERICAN FERN VOLUME NUMBER
JOURNAL:
100
3
No. of buds (BUD)
No. of young fronds (YOU)
No. of adult fronds (ADU)
No. of dead fronds (DEA)
Total no. of fronds (TOT)
No. of fronds (PER)
fertile
Basal area (BA)
conditions as the others, but owner decided allow regeneration
its to forest to
and
take place maintain was
the area for biological conservation. impossible
It
conduct
to a replication of the unharvested area due to the lack of populations
large enough for experimental studies and free from exploitation.
—
Evaluation demographic Demographic was
of structure. structure charac-
m
X
terized using four 5 5 plots in each of the four study populations. Since R.
adiantiformis is a rhizomatous plant, the fronds were considered individual
The
units. fronds were counted each development
at stage, in four categories:
young
buds, and dead The
fronds, adult fronds, fronds. presence of
fertile
was
fronds verified as well. The fronds were counted quarterly for one year,
May
with evaluations in August 2004, November 2004, February 2005, 2005
and August which
2005. In those cases in harvesting and monitoring occurred
the same month, were
in the evaluations carried out before frond harvesting.
The
respective environments of the sampled units were characterized based
on data of the existing tree and shrub plants in the plots. The parameters
number
evaluated were: of plants (height of over 1.30m, independent of the
-
and (DBH
species) diameter breast height 1.30m), the measured with
at latter
m
a pachymeter. This evaluation was carried out in an area of 8 X 8 in the
center of the plots described above and non-fern species were measured.
all
Data an aiysis.—Correlation analysis and principal component analysis (PCA)
were used demographic and
for the data analysis possible relations with
BUD
The
phytossociological descriptors. demographic descriptors were: (num-
YOU ADU
DEA
ber of buds), (number of young fronds), (number of aduh fronds),
TOT
(number FER
of dead (number and (number
fronds), of total fronds) of
fertile
BA DBH
The
fronds). phytossociological descriptors were: (basal area), (diameter
TRE
and (number The
at breast height) of trees). descriptors that contributed the
most
to axis formation, with correlations equal to or higher than 0.70, were used
in the interpretation. The population structure was compared using the
Kolmogorov-Smirnov
test with Bonferroni correction for multiple comparisons.
analyses were performed using and
Statistical Statistica 6.0 (StatSoft, 1998)
MVSP
d
(Multivariate Statistical Package) version 3.12 (Kovach, 2001).
BALDAUF & HARVESTING OF RUMOHRA ADIANTIFORMIS
REIS:
Results
—
number
Environmental aspects. Large variations in the of fronds, as
fertile
number
well as in the of fronds in each development category, were observed
in the populations studied. The number of fronds was the most variable.
fertile
The coefficients of variation obtained for each category were: buds (75.1%),
young fronds, (51.9%) aduh fronds, (47.7%), dead fronds 39.3%) and
fertile
(
The was
fronds (144.8%). existence of seasonality verified in the formation of
new
fronds (buds). Bud emergence was more intensive the end of winter and
at
A
the beginning of spring (between August and September). correlation
between some climatic factors and frond production in the species was also
The between number buds and mean monthly
observed. correlation the of
= <
was and between number
rainfall negative -0.58; p 0.01) positive the
(r^
= <
buds and months
of rainfall in the three before evaluation 0.67; p 0.01).
(r^
A
number
between and
negative correlation the basal area the average of
fertile
= <
was
fronds throughout the studied period also verified -0.43; p 0.01).
(r^
The Components 76.8%
three axes of the Principal Analysis explained of
first
The showed between
the variation in the data. axis (45.7%) a correlation the
first
R. adiantiformis population structure and the forest succession stage (Fig.
1).
YOU ADU
The (number young
correlations of the descriptors of fronds),
TOT
(number and (number with main
of adult fronds) of fronds) the
total
On
component axis one were high and positive (Table the other hand, the
2).
BA
descriptor (basal area) was high and the correlation with axis one, negative.
number and
Consequently, the of young, adult total fronds higher in
is
environments of lower basal area of tree and shrub species, and vice- versa.
AMERICAN FERN JOURNAL: VOLUME NUMBER
100
3 (2010)
The
second and
third axes explained 19.4% and 11.7%
of variance,
respectively, but the only descriptors with values equal to or higher than
DBH
were
0.70 (Diameter and TRE
Breast Height)
at (No. of trees) in axis two.
Thus, no
associations between demographic and
descriptors
phytossociologi-
shown
cal parameters two and
in axis three were above the value considered
for interpretation
(0.70).
Harvesting and
potential monitoring
indicators ofR. adiantiformis popula-
The
tions.—
patterns of population one and two were
similar in terms of
number
and
proportion of each development phase
during the year
(Fig.
2).
The number
of adult fronds dropped with
the first collection, but in the
following
evaluations the values were
similar those
to prior frond
to
harvesting. Likewise,
in population three, the recovery in frond number and
proportion was Bud
after harvesting production was
fast. intensive in August
sampled
2005, in populations
all (including the non-harvested
area).
In general, frond regeneration was
after the cuts rather managed
in the
fast
populations, evidenced by
numbers
the aduft frond found
after the collections
was
(Fig. also noticed demographic dynamism
2). ft that the of natural
R.
adiantiformis populations
considerable, evidenced by
is as the variations
in
number
frond and
proportion over the study year in different development
phases non-managed
of the
population
(population However,
four)
(Fig.
2).
between
August 2004 and August
2005, the frequency
relative of aduft fronds
decreased
in this population.
No
significant differences were observed among the management
traditional
systems
(populations two and
one,
three) in relation to the proportion of frond
categories (Table However,
populations one and two
revealed
3). differences
compared
as population
to four (non-managed), due
to distinct proportions of
and
aduft dead
fronds. In population two, the number was
of aduft fronds
higher than non-managed
in the population
(population Concerning
four). the
number
dead
of fronds, values were higher in population four (non-managed),
from
differing populations one and The
two. proportion was
of fronds
fertile
lower when
in population one compared
population The
to
four. other
comparisons
no
for this indicated
trait differences.
Environmental aspects.—The
high found
variation in the frond
initial
development
phases has been
reported in other studies on
this species (Milton,
Geldenhuys and Van
1987; Merwe,
der 1994; Conte and may
et 2000),
al.,
reflect the remarkable seasonality of bud emersion in the species. According
to
Geldenhuys and Van Merwe
der bud May
(1988), production begins between
and
South and
July in Africa reaches a peak between September and
November.
South
In the of Brazil, the most intensive period of bud production
in August and September (Anama,
is 2002; Baldauf, bud
2006). Intensive
production months was
in these confirmed
also in the first year of this study.
This occurred in both harvested and non-harvested
areas suggesting
that
this
emergence
not management
is related to the regime employed
(Fig.
2).

AMERICAN VOLUME NUMBER
FERN
JOURNAL:
100
3 (2010)
their experiments were positively associated with rainfall in the months prior
Conte
to evaluation. et (2000) also stated that in the spring of 1999 the
al.
new
production of fronds in the Southwest of Brazil decreased remarkably,
when
due low was
to soil moisture, rainfall less than half of the long-term
same
averages for the period in previous years. Ticktin et (2007) noted that
al.
changes and
in frond density rates of frond production over time coincide
with changes and
closely in precipitation levels over time that seasonal rainfall
were
rates significantly correlated with rates of frond production in two
Hawaiian number
fern species. Thus, the lower and proportion of buds found
in the evaluation carried out in August 2005 (when compared with 2004) are
probably due to the average rainfall value of the previous three months, which
was
approximately measured
half the value in 2004.
In relation to the vegetative propagation of the species, water availability in
months bud
the prior to formation possibly results in water reserves for the
which On
plant, reduces the chances of bud desiccation after emergence. the
new
other hand, Stamps et (1994) claimed that the emergence of fronds
al. is
associated with soil temperature, which increases during the sunny periods;
this explains the inverse and significant correlation between monthly
and
precipitation frond production observed
here.
Our show
results further a probable reduction in adiantiformis
R.
shown
populations caused by
forest regeneration, as in the principal
component showed when
This
analysis. analysis that the basal area of a
number young and
particular location increases, the of adult fronds tends
to
decrease. This tendency was confirmed by Milton and Moll who
also
(1988),
observed that the density of R. adiantiformis populations in South Africa was
negatively correlated with tree height and canopy cover. Local harvesters
mentioned
this aspect as a critical limit for leatherfern harvesting in the
number
Atlantic Rainforest (Baldauf et al, 2008). Additionally, the lower of
fertile fronds registered for population one is probably related to the higher
where The
basal area of the plots this population occurs. results presented here
show common
more more
that fertile fronds are in areas with solar radiation.
This emphasizes need open canopy maintain
fact the for areas to viable R.
adiantiformis populations.
BALDAUF RUMOHRA
HARVESTING OF
& ADIANTIFORMIS
REIS:
and
Harvesting potential monitoring indicators adiantiformis.—One
for R. of
the most remarkable tendencies and
of R. adiantiformis, highly relevant
for
management,
species the rapid frond regeneration
is after harvesting. In
general, frond regrowth of populations as evidenced by the number and
is fast,
proportion
of adult fronds in the evaluation carried out after the collection.
The
results for population three are particularly noteworthy, since they
most commonly
represent the used harvesting frequency in the region under
study (Baldauf
et al, 2007).
The
obtained here from
results differ largely those of populations studied in
South where seem
Africa, R. adiantiformis plants unable maintain frond
to
when
and
density size harvested (Milton, 1987; Milton and Moll, 1988;
Geldenhuys and Van Merwe,
der 1988). According to Milton (1991), a period
of five years is insufficient for R. adiantiformis, under natural conditions in the
indigenous
forests of the southern Cape, recover from
to total or partial
50% may
defoliation. In these forests, only of pickable fronds per plant be
all
harvested and the harvesting cycle was extended to 15 months to give plants
sufficient time to recover (Geldenhuys and Van der Merwe, These
1994).
may
contrasting results be related to climatic differences between the areas.
According Stamps
to et al. (1994), R. adiantiformis grows faster under a high-
temperature regime than under a low-temperature regime. In another study
was new
(Strandberg, 2003), observed that the production of fronds and the
it
time required mature were
for leaves to also strongly associated with degree
days, daylight hours, solar radiation, and soil temperature. However, there are
no
remarkable
differences in the temperature regime of the areas in question.
Thus, other factors might be involved.
One
possibility, already mentioned, the influence of rainfall on the
is
mm
production Annual Maquine
of fronds. rainfall in the area varies from 1400
mm
to 1800 (Hasenack and Ferraro, 1989), whereas the annual rainfall in the
mm mm
South
African area ranges from 920 1250 and
(Milton Moll,
to 1988).
new
Since the emergence of fronds shows a positive correlation with
rainfall,
this could be an explanation for the differences found.
Milton
(1991) attributes the slow recovery South African populations
in in
part to long leaf longevity in this species. In Brazilian natural populations,
fronds are also very long-lived (Anama, 2002; Conte et 2000), but despite
al.
this fact they can recover rapidly. Additionally, the higher harvest
rates
applied to Brazilian populations as compared with the rates employed in
South
Africa could cause reductions in frond production, as reported by
Milton (1991), but such reductions do not occur in Brazil least not when the
(at
traditional management system employed). Geldenhuys and Van der Merwe
is
(1988) point out that frond production in South Africa appears be nutrient
to
and
limited, frond removal also reduces nutrient availability. Therefore,
another would
possibility is that, in Brazilian populations, there be no nutrient
limitation to the species growth. Nevertheless, further studies are necessary
to
confirm
the influences of edaphoclimatic conditions on the population
structure of R. adiantiformis.
AMERICAN FERN VOLUME NUMBER
JOURNAL:
100
3 (2010)
Conte
suggested
et al. (2000) that a cutting cycle of five events each year
two months
(cycles of during the rainy season and months
of three in the dry
season) could maintain
the sustainability of R. adiantiformis populations
in
the Sao
State of Paulo, in the southeast region However, an
of Brazil. in
experimental study conducted in Rio Grande do Sul, in the south of
Brazil,
Anama
(2002) applied different harvesting frequencies and concluded
that
which maximum
traditional harvesting, consists of a of three annual was
cuts,
the only treatment that would not frond number and
affect
size.
Differences in ideal harvesting frequencies between the South and Southeast
regions of Brazil reinforce the influence of rainfall on frond production, since
the annual rainfall in the area where Conte et al. (2000) conducted thefr study
mm
2300 (CEPAGRI,
is 2010), higher than the values observed in other areas
where
the species harvested These
is in Brazil. differences in appropriate
show
harvesting regimes also that monitoring indicators must be determined
on
a regional
scale.
The
Anama
results reported here agree with those obtained in the study by
(2002), since they reinforce the absence of negative impacts
of traditional
harvesting on the species population The dead
structure. differences in frond
among
proportions found populations one and two and
population
four
probably
(control) reflect the fact that the harvesters are removing adult fronds
from populations one and two. Thus, dead would
less fronds be expected
in
these populations than where
in the confrol, fronds die naturally or remain on
the plant for a long time. Nevertheless, no differences were observed on
the
proportions buds and young among
of fronds the populations under
may
which
investigation, be considered
a sign of harvesting
sustainability.
According
to Peters (1996), the first sign of over-exploration of a plant
population expressed
most
is in its size class distribution. In species, the most
visible effect of over-exploration is a reduction in the number of seedlings and
young
plants. Since R. adiantiformis a rhizomatous species, the frond
is initial
development
phases (buds and young
fronds) are understood as equivalent
to
and young
seedlings
plants of tree species in order to monitor the harvesting
activity.
view managed
In of the fact that in populations the variation in the
young
proportion of fronds lower than of buds, the former was considered
is a
better indicator of the demographic dynamism,
characteristic of the species
and
of possible impacts of frond harvesting. The reference values
for this
indicator are variable, due to a period of intensive frond formation the end of
at
the winter (August and September). Based on no
the fact that there
is
difference in the proportion of young fronds in the managed populations when
compared with non-managed
the population, references values were
estab-
lished for monitoring. These values comprise the range of proportions
obtained
in all studied populations plus a safety margin. Therefore, the value
young 10% 20%
of fronds should be between and in the period between
September and December bud
whereas
(after formation), in other periods of the
5%
year a proportion between and 15% would be expected.
BALDAUF RUMOHRA
& HARVESTING OF ADIANTIFORMIS
1 :
The demographic
structure of leatherleaf fern largely influenced by climatic
is
factors and forest succession and must therefore be continuously monitored
over several years. also important to consider that the results and monitoring
It is
indicators presented here are based on studies carried out few and
at sites
should be revised periodically. In cases that require a more in-depth analysis,
it
possible to analyze the genetic diversity of the managed populations, based on
is
By
allozyme markers. main
this technique, the genetic diversity indexes can be
may
determined; in case of population overexploitation, these values be
reduced Once
(Baldauf demographic
et al., 2008). the data are relatively easy to
may
obtain, evaluations include the local harvesters of the collection areas,
helping to evaluate the populations in a participatory monitoring process.
should be emphasized low
Finally, that the ecological impact of the
it
management
traditional systems not maintenance
is sufficient for the of a
leatherleaf fern population, since is a pioneer species that tends to disappear
it
during and
forest succession. Hill Silander (2001) studied the pteridophyte
Jr.
species Dennstaedtia punctilobula (Michx.) Moore and understood any
T. that
event that maintains or results in canopy openness death caused by
(tree
windthrow
disease, or forest harvesting, or the elimination of a shrub layer by
browsing)
favors the persistence of D. punctilobula. Analogously, the
maintenance
of R. adiantiformis populations also depends on further
management
measures would improve example
that productivity, as for the
management
of "capoeira" (naturally regenerated second-growth
forest) areas.
This, along with the farming of perennial agricultural species, could be a way
income and
to guarantee food security to the harvesting families.
Literature Cited
2006. Extrativismo de samambaia-preta [Rumobra
adiantiformis
(G. Forst.]
.
M
rande do Sul: fundamentos para o manejo e monitoramento da atividade.
;rsidade Federal de Santa Catarina, Florianopolis, SC.
Hanazaki and M.
N. S. Reis. 2007. Caracterizagao etnobotanica dos sistemas
,
[Rumohra
tnambaia-preta adiantiformis (G. Forst.) Ching- Dryopteridaceae)
do Acta
Brasil. Bot. Bras. 21(4):823-834.
1
R. BiTTENcouRT, D. K. Ferreira, A. G. Mattos, ScHULTZ, A. Mantovani and
, J.
A
diversidade genetica das populagoes de samambaia-preta {Rumobra adiaw
Ching) no Rio Grande do Sul, in Miguel, L. A., G. P. C. Souza and R. R.
]
,
tivismo da samambaia-preta. Editora da UFRGS, Porto Alegre.(/n press].