Filetype PDF | Posted on 21 Jan 2023 | 2 years ago
Authentication
digital resources
159x Filetype PDF File size 0.72 MB Source: www.interciencia.net
APPLYING ECOLOGICAL DIVERSITY INDICES
WITH ECOSYSTEM APPROACH AT
ECOREGIONAL LEVEL AND PRIORITIZING THE
DECREE OF NEW PROTECTED NATURAL AREAS
LUZ MARÍA CRUZ GARCÍA, JOSÉ ALFREDO ARREOLA-LIZARRAGA,
RENATO A. MENDOZA-SALGADO, PATRICIA GALINA-TESSARO, LUIS
FELIPE BELTRÁN-MORALES AND ALFREDO ORTEGA-RUBIO
SUMMARY
One of the main challenges of the new Protected Natural indices were the most sensitive ones, considering the trends of
Area prioritization is using the appropriate tools to determine their charts, variations, and the distance between their range
which of the areas are more representative at the ecoregional values, which describe ecosystem diversity more accurate-
level. In this work, we used ecological diversity indices (EDIs) ly among different ecoregions. However, in order to describe
with an ecosystem approach as a tool to compare the differ- the richness and heterogeneity of the analyzed ecoregions, the
ences in ecosystem diversity among different ecoregions. After Simpson’s Inverse index was the most useful to define which of
comparing five EDI at coastal and marine ecoregions on North- the regions have greater diversity of ecosystems in comparative
western Mexico, it is concluded that out of the five indices an- studies among them, and therefore the priority to be enacted as
alyzed for ecosystem diversity, the Simpson’s Inverse and Hill a new Protected Natural Area.
cological diversity indices et al., 2014). They usually consider species (Simpson, 1949) are commonly used; but
(EDIs) are indicators em- richness and abundance, including some other indices used in evenness analysis
ployed to describe the further considerations of the relationship be- are those of Gleason (1922), Brillouin
most important and inherent biological tween biotic and abiotic components (1962), Menhinick (1964), Margalef
characteristics of an ecosystem (Izsák and (Jizhong et al., 1991; Soininen et al., 2012; (1968), and Pielou (1969), whose varia-
Papp, 2000). EDIs are also directly used for Lyashevska and Farnsworth, 2014). tions are due to differences in the weight
ecosystem management and conservation, EDIs such as the given to species richness and their even-
and some are used as health, structure, and Shannon-Weaver index (Shannon and ness, as well as a differential sensitivity
performance indicators (Butturi-Gomes Weaver, 1949) and the Simpson index to sample size. Also, in order to compare
KEYWORDS / Ecological Diversity Index / Ecoregion / Ecosystem / Protected Natural Area /
Received: 11/09/2014. Modified: 02/19/2015. Accepted: 02/20/2015.
Luz María Cruz García. Doctoral student in Use, Management and Conservation of Natural
Resources. Centro de Investigaciones Biológicas del Noroeste (CIBNOR), Mexico. e-mail: luzmcg@yahoo.com.mx
José Alfredo Arreola-Lizarraga. Doctor of Science in Management, Use, and Conservation of
Natural Resources, Northwest Biological Research Center, CIBNOR, Mexico. e-mail: aarreola@cibnor.mx
Renato A. Mendoza-Salgado. Doctor of Science in in Management, Use, and Conservation of
Natural Resources, Northwest Biological Research Center, CIBNOR, Mexico. Researcher, CIBNOR, Mexico. e-mail: rams@cibnor.mx
Patricia Galina-Tessaro. Doctor of Science, in Management, Use, and Conservation of Natural
Resources, Northwest Biological Research Center, CIBNOR, Mexico Researcher, CIBNOR, Mexico. e-mail: pgalina04@cibnor.mx
Luis Felipe Beltrán-Morales. Doctor in Environmental Sciences, Universidad de Concepción,
Chile. Researcher, CIBNOR, Mexico. e-mail: lbeltra04@cibnor.mx
Alfredo Ortega-Rubio. Doctor of Science in Ecology, Instituto Politécnico Nacional, Mexico.
Researcher, CIBNOR, Mexico. Address: CIBNOR. Av. Instituto Politécnico Nacional 195, Playa Palo de Santa Rita Sur, La Paz, Baja
California Sur, C.P. 23096, México e-mail: aortega@cibnor.mx
MARCH 2015, VOL. 40 Nº 3 0378-1844/14/07/468-08 $ 3.00/0 179
the similarity in species diversity among (EDIs) to determine the heterogeneity of 2013). The Pacific coast is characterized
sites, a variety of indexes are used, such ecosystems at ecoregional level in the by a wider continental shelf, sandy
as those of Jaccard (1908), of Sorensen coastal zone of Baja California Sur, and coast, alluvial fans, floodplains
(1948) and the Bray-Curtis index (Bray to determine which of them expresses the (Wilkinson et al., 2009) with the excep-
and Curtis, 1957). best results. tion of the Cape Region, which is char-
The EDI used at land- acterized by a mountainous system seg-
scape level is the Shannon-Weaver in- Material and Methods mented into smaller blocks, a complex
dex, because of its wide application to of crystalline igneous and metamorphic
determine entropy and categorization of Study area rocks, mainly granite (López-Blanco and
landscape patches (Yoshida and Tanaka, Villers-Ruiz, 1995; Martínez and Díaz,
2005; Dušek and Popelková, 2012). It The coastal zone of 2011); while the Gulf showed a narrower
has also been used to determine land- Baja California Sur (BCS), Mexico, has continental shelf, abundant islands, rocky
scape structure, including area, shape, a length of 2,131km (Figure 1). An out- coastal cliffs, and small alluvial delta
density and proportions to guarantee standing feature of BCS is the diversity fans (Wilkinson et al., 2009).
their conservation (Kuchma et al., of coastal and marine ecosystem due to The coastal zone of
2013), and it has been applied at eco- the influence of the Pacific Ocean and BCS includes 10 ecoregions (González-
system level as an entropy index that the Gulf of California, both of them with Abraham et al., 2010; Wilkinson et al.,
identifies the number of components in particular geological and oceanographic 2009): Gulf Coast (Gc), La Giganta
an ecosystem and the interactions features (De la Lanza-Espino et al., Ranges (Gr), Sarcocaulescent Shrubland
among them. On the other hand, the
Simpson index has been used jointly
with economic variables to determine
life satisfaction in a specific area
(Ambrey and Fleming, 2014).
No previous records
were found on the use of EDI applied,
not only to know the biodiversity within
an ecosystem but also to determine the
diversity of different coastal ecosystems
within a given ecoregion. The nearest
similar application of the EDI was per-
formed by Lapin and Barnes (1995),
who analyzed the landscape on the basis
of species and ecosystems diversity to
generate a map and a classification of
the area under study, indicating the rich-
ness and heterogeneity of ecosystems. In
our study, EDIs were applied to deter-
mine the variability of different ecosys-
tems that occur in the coastal and marine
ecoregions of the State of Baja
California Sur, Mexico.
Taking into account that
diversity has two components: species
richness and evenness, in this study we
considered species richness as the number
of ecosystems present, and evenness as
the coverage of each ecosystem in each
study site along the coastal zone.
The EDIs that emphasize
on diversity are those of Shannon-Weaver
(1949), Simpson (1949), Simpson’s Inverse
(Williams and Lambert, 1959) and Hill
(1973). They can help to obtain a profile of
ecosystem diversity on a coastal area, for
environmental monitoring and decision
making for conservation and management
(Spellerberg, 1991), and they can also be
applied to monitor the possible effects of
environmental disturbances (Moreno, 2001).
The theoretical founda-
tion of this study is the use of ecosystem
richness as the basic criteria to be applied
in the analysis and the recognition of pri-
ority areas for conservation. Our goal is Figure 1. Study area indicating the ecoregions of Baja California Sur, Mexico. Adapted from
to apply the Ecological Diversity Indices González Abraham et al. (2010) and Wilkinson et al. (2009).
180 MARCH 2015, VOL. 40 Nº 3
(Ss), Tropical Dry Forest (Td), Southern Margalef index Results
Baja California Neritic (Sb), Cape S-1
D= /ln(N) The ecoregions with
Cortezian Neritic (Cn), Magdalena Plains
(Mp), Vizcaino Desert (Vd), Cape where D: ecosystem richness, S> total high surface values (>1000ha) were the
Pacific Neritic (Cp), and Vizcainean number of ecosystem, and N: the sum of Gulf Coast, Magdalena Plains, Vizcaino
Neritic (Vn). As a variable we use the ecosystem i. Desert and Sarcocaulescent Shrubland;
2 with medium surface values (500-1000ha)
surface area (km ) analyzing 11 catego-
ries: bare soil, beach, coastal water body, Simpson index was Cape Cortezian Neritic; and those
mangrove, riparian, salt flat, salt marsh, s with low surface values (<500ha) were
scrub, reef, seagrass, and other vegeta- λ= pi2 Southern Baja California Neritic, La
tion types. To test differences the follow- ∑ Giganta Ranges, Tropical Dry Forest, and
ing statistical analysis were applied: i1= Vizcainean Neritic (Table I).
cluster analysis (similarity measure: where λ: dominance index, p: proportio- Differences between N
Bray-Curtis) and EDI. i values were observed, highlighting two
The ecoregion and eco- nal abundance of ecosystem i, i.e. the to- groups of interrelated ecosystem (Figure 2)
system areas were estimated through the tal surface of ecosystem i divided by the and differentiating the coastal and marine
ni
total surface sum: pi = /
exploration method of satellite imagery N parts. The groups of ecoregions with high-
© er association by their similarity were
with the IDRISI Taiga software Simpson’s Inverse index
(Eastman, 2009), by performing a vector IS = 1/λ 1) Sarcocaulescent Shrubland, Gulf Coast,
transformation of the shapes and ex- Magdalena Plains, Vizcaino Desert; La
pressing the results in hectares; 1,200 Shannon-Weaver index Giganta Ranges, and Tropical Dry Forest;
pixels of Landsat 5TM raster images and 2) Cape Pacific Neritic, Cape
were selected at random to determine the H’=− pilnpi Cortezian Neritic, Southern Baja California
ecosystem type. The surface of each eco- ∑ Neritic, and Vizcainean Neritic.
system was measured for each ecoregion. where H’: diversity and p: surface pro- This was consistent with
To validate their classification, type 973 portion in ecosystem i. i nMDS analysis, which indicated that in the
checkpoints were chosen at simple ran- coastal group Sarcocaulescent Shrubland,
dom sampling to cover the coastal zone Hill index Gulf Coast, Magdalena Plains, Vizcaino
of BCS. Desert, and La Giganta Ranges had higher
H’
Ecosystems in ecore- N1 = e similarity, while in the marine group Cape
gions were classified hierarchically for Pacific Neritic, Cape Cortezian Neritic,
cluster analysis, choosing the correla- where N1: diversity of ecosystems, the and Southern Baja California Neritic had
tion coefficient as a measure of associa- natural logarithm (log to base e), and H’: more similarity (Figure 3).
tion (Sokal and Rohlf, 1962) and the Shannon-Wiener diversity. The behavior of the five
unweighted pair group method using The numerical values indices proposed for this ecosystem diver-
of the EDIs were normalized and sity analysis showed three well defined
arithmetic averages (Sokal and
Michener, 1958) as aggregation algo- grouped according to similar patterns in patterns: 1) similar variation of Simpson´s
rithm. The distortion of the relation- charts, so as to make comparisons Inverse and Hill indices (Figure 4a);
ships was measured by the cophenetic among them. 2) similar variation between Simpson and
correlation coefficient (Cunningham and
Ogilvie, 1972).
In addition to a multi-
variate analysis, nonmetric multidimen- TABLE I
sional scaling (nMDS, multidimensional 2
ECOSYSTEM SURFACE (km) IN TEN ECOREGIONS
scaling of nonparametric transformed OF THE COASTAL ZONE OF BAJA CALIFORNIA SUR, MEXICO
data, fourth root), was applied and stan- Ecosystem
dardized to determine similarities s
among ecoregions. For this analysis we e
p
y y
used Primer v.6 Software (Clarke and d t
o n
Gorley, 2006). b o
i
Ecoregion r t
e a
The surface values ob- t t
e e
l g v sh
tained were analyzed with PRIMER v.6 wa n i e t r
l a a o a ss
a i v h l r a
Software (Clarke and Gorley, 2006) us- b st r so r c f g m f r
u a a e e t n t g
r p r h a l l e a
ing the similarity method (Bray and o i a e e
Curtis, 1957), obtaining the fourth root Sc C R B Ot B Sa Ma Sa R Se N*
for each datum, and thus assessing each Gulf Coast 843 0 1276 176 268 18 18 1 3 0 0 2603
ecoregion, expressed on a dendogram La Giganta Ranges 78 0 124 1 12 0 0 0 0 0 0 215
that defined the relations among the Sarcocaulescent Shrubland 976 0 139 23 234 2 3 1 2 0 0 1380
BCS ecoregions. Tropical Dry Forest 12 0 3 0 1 0 0 0 0 0 0 16
Southern Baja California Neritic 0 358 0 0 0 0 0 1 0 0 1 360
We employed the fol- Cape/Cortezian Neritic 0 946 0 0 0 0 0 1 1 4 0 952
lowing EDIs in this work: Margalef Magdalena Plains 1923 0 55 85 146 198 51 99 1 0 0 2558
(1968), Simpson (1949), Simpson’s Vizcain Desert 163 0 31 1058 77 341 245 15 40 0 0 1970
Inverse (Williams and Lambert, 1959), Cape/Pacific Neritic 0 2018 0 0 0 0 0 28 22 0 0 2068
Shannon-Weaver (1949) and Hill (1973). Vizcainean Neritic 0 5 0 0 0 0 0 1 0 0 0 6
The mathematical expressions of each in- Total area of ecosystem (km²) 3995 3327 1628 1343 738 559 317 147 69 4 1
dex are as follows * N: total area of the ecoregion (km2).
MARCH 2015, VOL. 40 Nº 3 181
0.10 California. In the Pacific Ocean, however, Finding the right method
Vd and Mp were the ecoregions represent- to describe the ecosystem diversity is a
ing the areas with the greatest diversity. challenge, because all indices have differ-
The Hill index had low- ent purposes and each one of them has
er values with a mean difference between several advantages and disadvantages.
range values of 0.18, but it agreed with Therefore, in this work three fundamental
the Simpson’s inverse index in the Vd advantages led us to detect and recom-
ecoregion, both with a value of 1.0 mend the Simpson’s Inverse and Hill in-
(Figure 4a). The Simpson and Shannon- dices as the most appropriate to deter-
Weaver indices showed differences in the mine differences among ecosystem diver-
mean range of 0.07 between the values sity in ecoregions: 1) the theoretical foun-
for each ecoregion but agreed in the high- dations of both indices, 2) the minimum
1.00 x x x x x x est values in Mp and Vd, as well as in
Vn Sb Cn Cp Vd Mp Gc Ss Gr Td the lowest value in Cp (Figure 4b). The
Samples Margalef index had a distinct pattern with
variations between 0.27 in Cp to 1 in Ss 1.00
Figure 2. Classification of ecoregions accord- (Figure 4c). 0.90
ing to cluster analysis. Ordinate shows coeffi- 0.80
cient of correlation. Cophenetic correlation co- Discussion
efficient: 0.94. Ecoregions: Gulf Coast (Gc), 0.70
La Giganta Ranges(Gr), Sarcocaulescent The diversity index applied to the 0.60
Shrubland (Ss), Tropical Dry Forest (Td), x
Southern Baja California Neritic (Sb), Cape ecosystems in ecoregions suggest a useful 0.50
Cortezian Neritic (Cn), Magdalena Plains (Mp), approach, as there is a diversification of Inde0.40
Vizcaino Desert (Vd), Cape Pacific Neritic environments in the coastal zone of Baja 0.30
(Cp) y Vizcainean Neritic (Vn). California Sur that contributes to area
heterogeneity mainly in two strands, the 0.20
Gulf of California and the Pacific Ocean, 0.10 a)
Shannon-Weaver indices (Figure 4b); 3) a where distinct patterns of ecosystems are 0.00
different pattern was shown by the identified according to Lapin and Barnes Inverse Simpson Hill
Margalef index (Figure 4c). In the four (1995) and González-Abraham et al.
cases of Figures 4a and b, the highest (2010). We consider that such patterns are 1.00
value (1.0) was observed in the Vd ecore- associated to climate, physiography, to- 0.90
gion, and the lowest values in Sb and Cn pography and geology of the zone, and 0.80
ecoregions. Both minimum and maximum reflect the complexity of the geomorpho- 0.70
value of the Margalef index (Figure 4c) logical and geological processes that form 0.60
corresponded to different ecoregions. this landscape and in turn the ecosystem. x
According to the EDIs The similarity analysis 0.50
applied, the results indicated that the areas allowed us to observe three groups in the Inde0.40
with higher diversity were distributed in the coastal and marine zones, with a clearly 0.30
coastal zone and those with lower diversity larger ecosystem diversification in the 0.20
were located in the marine part. For the coastal than in the marine zone. These 0.10
coastal zone the EDIs showed that Gc was groups are: 1) Sarcocaulescent Shrubland, b)
the most diverse ecoregion at the Gulf of Gulf Coast, Magdalena Plains, and 0.00
Vizcaino Desert; 2) La Giganta Ranges Simpson Shannon-Wiener
and Tropical Dry Forest; and 3) Cape 1.00
Pacific Neritic, Cape Cortezian Neritic,
Transform: Fouth root 2D Stress:0.01 Southern Baja California Neritic, and 0.90
Resemblance: S17 Bray Curtis similarity Vizcainean Neritic. 0.80
Class Similarity The analysis also indi- 0.70
xCoastal 40
Marine cated that not all the marine environmen- x0.60
tal conditions allow the presence of reefs Inde0.50
Vn and seagrass. Also, it indicated that the 0.40
Gr coastal ecosystems are strongly influenced 0.30
Sb x Ss x by the marine zone, by topography, and 0.20
x Mp by latitudinal species distribution.
Cn x Gc Td 0.10 C)
x Vd x In particular, the terrestri-
Cp al ecosystems have a higher affinity and 0.00 Gc Gr Ss Td Sb Cg Mp Vd Cp Vn
similarity among them, and the analysis Ecoregion
showed that even having the same kind of Margalef
Figure 3. nMDS ordination of the ecoregions vegetation, topography plays a major role.
of the coastal zone of Baja California Sur: For instance, the differences between Gr Figure 4. Variation of the diversity index nor-
Gulf Coast (Gc), La Giganta Ranges(Gr), and Td ecoregions are attributed to topog- malized to ecoregions of the coastal zone of
Sarcocaulescent Shrubland (Ss), Tropical Dry raphy despite the fact that the dominant Baja California Sur: Bare soil (Bs), beach
Forest (Td), Southern Baja California Neritic vegetation in both ecoregions is scrub (Be), coastal water body (Cw), mangrove
(Sb), Cape Cortezian Neritic (Cn), Magdalena (González-Abraham et al., 2010). This was (Mg), riparian (Ri), salt flat (Sf), salt marsh
Plains (Mp), Vizcaino Desert (Vd), Cape consistent when using nMDS analysis. (Sm), scrub (Sc) other vegetation types (Ot)
Pacific Neritic (Cp) y Vizcainean Neritic (Vn). reef (Rf) and seagrass (Sg).
182 MARCH 2015, VOL. 40 Nº 3
no reviews yet
Please Login to review.
