Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
1
Incorporating early life stages of fishes into estuarine spatial conservation 1
planning 2
3
Running Head: Ichthyoplankton and estuarine spatial conservation 4
5
Costaa*, MDP; 6
a Laboratório de Ecologia do Ictioplâncton, Instituto de Oceanografia, Universidade Federal do 7
Rio Grande, Campus Carreiros, Avenida Itália Km 8, CP 474, 96201900, Rio Grande, RS, 8
Brazil; [email protected] 9
*Corresponding author, (+55) 53 32336529. 10
11
Possinghamb,c, H.P.; 12
b ARC Centre of Excellence for Environmental Decisions, School of Biological Sciences, The 13
University of Queensland, Brisbane, Qld 4072, Australia. 14
c Imperial College London, Department of Life Science, Silwood Park, Ascot SL5 7PY, 15
Berkshire, England, UK. 16
18
Muelberta, JH 19
a Laboratório de Ecologia do Ictioplâncton, Instituto de Oceanografia, Universidade Federal do 20
Rio Grande, Campus Carreiros, Avenida Itália Km 8, CP 474, 96201900, Rio Grande, RS, 21
Brazil; 22
24
25
26
27
2
ABSTRACT 1
1. Many species of fish depend on estuaries to complete their development. 2
Most of them have a planktonic early life-stage, and adults of the same 3
species often live in a different habitat. 4
2. The aim was to assess the importance of incorporating data on fish eggs and 5
larvae in systematic conservation planning at the Patos Lagoon estuary, 6
Brazil. 7
3. Different scenarios, where fish larvae and eggs were or were not included in 8
the systematic conservation planning process, were investigated. 9
4. An estimate of artisanal fishing revenue was used as an opportunity cost and 10
compared with a spatially homogeneous cost. Cluster analysis was 11
performed to assess the impact of incorporating ichthyoplankton data on the 12
outcomes of estuarine systematic conservation planning. 13
5. Regardless of the opportunity cost, the spatial plans fell into two clusters - 14
those with and without fish egg and larvae data. This shows that egg and 15
larvae data have a large impact on priorities for conservation actions in 16
space. 17
6. This approach is the first to combine artisanal fishery economic spatial data 18
with a conservation plan that incorporates early life stages of fishes. 19
7. In the case of the Patos Lagoon estuary, shallow areas were particularly 20
important for reaching conservation targets in all scenarios. Considering the 21
dynamic nature of these ecosystems, much work needs to be done to devise 22
better methods of spatial planning in estuaries. 23
3
Key words: estuarine conservation; fish eggs; fish larvae; fish conservation; 1
Marxan; systematic spatial conservation planning 2
3
INTRODUCTION 4
Systematic conservation planning addresses the optimal allocation of 5
spatially explicit conservation actions aiming to promote the persistence of 6
biodiversity (Moilanen et al., 2009; Wilson et al., 2009a; Watson et al., 2011). 7
The main aim of this approach, which traditionally has been used for the design 8
of protected areas in the marine and terrestrial realm, is to achieve biodiversity 9
objectives while minimising economic costs (Smith et al., 2006; Watson et al., 10
2011). Few studies have used the systematic conservation planning approach to 11
identify priority areas for conservation in estuaries, and most of them are based 12
on benthic macroinvertebrates and habitat types, which are assumed to be 13
surrogates for all biodiversity (Neely and Zajac 2008; Geselbracht et al. 2009; 14
Shokri and Gladstone 2009, 2013a,b; Shokri et al. 2009). 15
The costs of taking conservation actions in different places are not 16
spatially homogeneous, and there are many types of costs related to conservation 17
actions (Wilson et al., 2009b). Among them, opportunity cost represents the lost 18
benefit when an action precludes a profitable activity that could otherwise occur 19
there; e.g. when a specific site that is used for fishing or navigation is chosen as 20
a no-take marine reserve (Wilson et al., 2009b). The application of opportunity 21
costs in the planning analysis has already been tested in many studies, showing 22
that the incorporation of this information can help to achieve more cost-effective 23
outcomes than when costs are neglected (Naidoo et al., 2006; Klein et al., 24
4
2008a,b; Carwardine et al., 2008; Ban and Klein 2009; Mazor et al., 2014). In 1
the marine environment, most planning studies have used fisheries information 2
to integrate opportunity cost into the analysis (Ban and Klein, 2009). Among the 3
studies that have applied the conservation planning approach in estuaries, only 4
Geselbracht et al. (2009) included opportunity cost in the planning analysis by 5
developing a spatial index of socio-economic factors as opportunity costs. The 6
lack of studies using opportunity cost in estuarine conservation planning exposes 7
a gap in the understanding of the analysis, mainly on the effect of this 8
information on the spatial prioritization analysis in a highly competitive 9
ecosystem such as estuaries. 10
Estuaries are among the most dynamic and productive ecosystems in 11
coastal areas, acting as a nursery ground for a wide variety of species. In 12
addition to nursery, breeding and refuge functions, estuaries are also important 13
for ecosystem regulatory functions, disturbance prevention, sediment retention 14
and nutrient regulation (Barbier et al., 2011). Unfortunately, estuarine and 15
coastal areas are among the most heavily impacted natural ecosystems, and 16
cumulative impacts from human activities can directly affect the benefits 17
provided by these ecosystems (McLusky and Elliott, 2006). Several fishes, 18
mainly marine species, depend on estuaries to complete their development 19
(Elliott and Hemingway, 2002; McLusky and Elliott, 2006). Most of these 20
species have a planktonic early life stage, which means that adults of the same 21
species sometimes live in different habitats and rely on the dynamic nature of 22
estuaries to complete their life cycle. 23
5
The distribution of fish eggs and larvae in an estuary depends on both 1
biological processes (e.g. mortality, spawning and behaviour) (Boehlert and 2
Mundy, 1988; Schultz et al., 2003), and abiotic factors (e.g. salinity and 3
circulation) (Govoni, 2005; Babler, 2000). These processes interact 4
synergistically, influencing the occurrence and distribution of ichthyoplankton in 5
estuaries. These early stages are unable to move against currents, being 6
passively transported in the ocean-estuary interface, and they are retained in 7
specific sites in the estuary by physical processes (Boehlert and Mundy, 1988; 8
Martins et al., 2007). In addition to fish eggs and larvae, the early life history of 9
fishes that use estuaries includes settled juvenile fishes. This late life stage is 10
already capable of directional swimming. Ontogenetic habitat changes during 11
development results from variation in habitat requirements and can structure the 12
spatial distribution of each developmental stage throughout the life cycle (Le 13
Pape et al., 2014). Most estuarine-dependent species complete their 14
development in estuaries and migrate to the ocean as an adult to spawn. Despite 15
the relationship between the survival of fish eggs and larvae and recruitment 16
success (Govoni, 2005; Houde, 2008), no other study has used information 17
about fish eggs and larvae in a systematic conservation plan. 18
The present study incorporates data on the early life stages of fish (eggs 19
and larvae) into a decision support tool to assess the impact of adding these data 20
into estuarine systematic conservation planning. The open question is, will the 21
inclusion of some of the early life history stages of fishes that use estuaries 22
affect the outcomes of conservation planning, and how is this affected by the 23
incorporation of opportunity costs? While fish eggs and larvae are not directly 24
6
impacted by fishing, they are affected by other anthropogenic activities such as 1
habitat destruction and pollution (Allison et al., 1998). These factors are of great 2
importance when identifying and designing protected areas, especially marine 3
protected areas created where the main objective is to sustain marine fish 4
populations and fisheries (Warner et al., 2000). In cases where recruitment is 5
driven by the retention of eggs and larvae, conservation of retention areas is as 6
important as conservation of adult habitat (Warner et al., 2000). In order to 7
answer the main question of the study, we used a large dataset including 8
physical features (e.g., bathymetry, sediment type, and habitat) of the estuary 9
and the spatial-temporal distribution of six important fisheries species. 10
11
METHODS 12
Study Area 13
Patos Lagoon (Figure 1) is a warm subtropical, river-dominated choked 14
lagoon located in southern Brazil (32°S, 52°W - Brazil) with a large estuary that 15
is important to several commercial species. This ecosystem is characterized by 16
its ecological (high biological productivity and biodiversity) and socio-economic 17
importance (ports, industries, agriculture, aquaculture, and fisheries) (Odebrecht 18
et al., 2010). Ichthyoplankton studies in this estuary were conducted to 19
investigate the distribution and ecology of species and the influence of 20
environmental variables on species distribution and abundance (Muelbert et al., 21
2010). In this system, the retention and survival of early stages of fishes are 22
strongly dependent on water exchange and prevailing winds (Muelbert and 23
Weiss, 1991; Sinque and Muelbert, 1997; Martins et al., 2007; Odebrecht et al., 24
7
2010). Furthermore, the occurrence and distribution of fish eggs and larvae are 1
strongly affected by seasonal variation in salinity and temperature (Muelbert et 2
al., 2010). This is the case with many other estuaries around the world 3
(Berasategui et al., 2004; Faria et al., 2006; Ramos et al., 2006; Simionato et al., 4
2008; Primo et al., 2011). The Patos Lagoon estuary was chosen as a case study 5
for estuarine spatial conservation prioritization because of the breadth of existing 6
information and its ecological and economic importance. 7
The majority of the lagoon is freshwater (80%), but brackish water 8
creates an estuarine system of approximately 1000 km² in the southern reach. 9
The deep (15 m) and narrow (800 m) inlet of this system acts as a filter 10
attenuating the advance of tidal waves into the estuarine region (Seeliger, 2001; 11
Odebrecht et al., 2010). Wind and freshwater discharge are the main forces 12
controlling water exchange with the open ocean. Hence, the geographic limits of 13
this estuarine ecosystem are conditioned by climatic factors (Odebrecht et al., 14
2010). The estuarine region of Patos Lagoon (Figure 1) is predominantly 15
shallow (< 1.5 m), although it also has some intermediate waters (1.5 – 5.0 m) 16
and deep waters (> 5.0 m). The estuary is composed of different benthic habitats 17
including unvegetated subtidal flats (300 km²), seagrass beds (120 km²), 18
marginal salt marshes (40 km²) and artificial hard substrates (~ 5 km-long rocky 19
jetties) (Seeliger, 2001). A wide variety of marine fishes and invertebrates 20
depend on these habitats to complete their development and successfully recruit 21
to the adult population (Seeliger, 2001; Odebrecht et al., 2010). 22
23
24
8
Conservation features data 1
The Patos Lagoon estuary was divided into 2,355 hexagonal planning 2
units with sides of 0.5 km using QuantumGIS software (QGIS Development 3
Team, 2013). Distribution data of 64 features were selected as surrogates for 4
biodiversity and included bathymetry, sediment, habitat, and ichthyoplankton 5
(Table 1). Ichthyoplankton data comprise seasonal distribution information over 6
three years for six fish species (Achirus garmani, Brevoortia pectinata, 7
Licengraulis grossidens, Micropogonias furnieri, Parapimelodus valenciennes 8
and Thrichiurus lepturus). These species have the most abundant fish eggs and 9
larvae in the Patos Lagoon estuary (Muelbert and Weiss, 1991; Muelbert et al., 10
2010) and were chosen to represent the estuarine use ecological guilds: marine 11
(T. lepturus), estuarine dependent (B. pectinata, M. furnieri), true estuarine (A. 12
garmani), marine/estuarine resident (L. grossidens), and freshwater species (P. 13
valenciennes) (McLusky and Elliott, 2006; Mai et al., 2014). Because changes in 14
the abundance and occurrence of ichthyoplankton at the Patos Lagoon estuary 15
are determined by fluctuations in salinity (Muelbert and Weiss, 1991; Muelbert 16
et al., 2010), it was assumed that the period of interest is a typical period for fish 17
eggs and larvae recruitment. Seasonal distribution maps applied in the 18
conservation planning were built using the MAXENT software (Phillips and 19
Dudík, 2008) and were based on records of presences of eggs and larvae of each 20
species. The environmental variables used in the distribution modelling were: 21
temperature, salinity, bathymetry, sediment type and aquatic vegetation. We 22
used 25% of the species occurrence data for testing and 75% for training the 23
models. Models were cross-validated and ran with 5,000 iterations and 10,000 24
9
background points. Predictive performance of the species distribution models 1
was analysed according to the Area Under the Curve (AUC), where values 2
closer to 0.5 indicate that model performance is no better than random, and 3
values closer to 1 indicate a better performance (Swets, 1988). Only models with 4
AUC values higher than 0.75 were used in the conservation planning. We used 5
the logistic threshold output, and a 10 percentile training presence logistic 6
threshold was used to build the seasonal distribution maps. Habitat data were 7
determined from the percentage cover of submerged aquatic vegetation of two 8
seagrasses species (Ruppia maritima and Zannichelia palustris) and macroalgae 9
and the meadow height during summer and autumn (B. Gianasi; Table 1). 10
Bathymetry was divided into three classes following a previous study (Plano 11
Ambiental de Rio Grande, 2007): shallow (< 1 m), intermediate (1 – 5 m) and 12
deep (channels, > 5 m) waters; sediment follows the classification proposed by 13
Calliari et al. (1980) based on sand-silt-clay content in 10 classes: silt, silty 14
sandy, silty clay, sandy silty, mixed sandy+silt+clay, clay silty, sand, sandy clay, 15
clay sandy, and clay. Physical features of the Pathos Lagoon estuary were 16
chosen as surrogates for biodiversity according to their availability and 17
relevance to estuarine diversity. Many species prefer particular water depths 18
(e.g. species that prefer shallow water and those associated with deep channels) 19
(Costa et al., 2014). The sediment type can be a good surrogate for benthic fauna 20
and other species associated with the estuarine bottom (Elliot and Hemingway, 21
2002), and submerged aquatic vegetation is an important estuarine habitat, 22
which many species rely on to complete their development (e.g., meadow height 23
10
can be an indicative measure of complexity of this habitat: the higher the 1
meadow the more developed and complex it is) (Gillanders, 2006). 2
3
Opportunity cost 4
In this study, fishing revenue was chosen as the opportunity cost of 5
closing a planning unit to fishing because artisanal fishery is one of most 6
important economic activities in the Patos Lagoon estuary, and it occurs 7
throughout the estuary (Schafer and Reis, 2008). This cost was added to a flat 8
cost of 1 for all planning units. The flat cost ensures that planning units with no 9
fishing effort are not simply added to the reserve system if they have no 10
conservation benefit. The fishing revenue was calculated using the equation for 11
commercial fishing proposed by Mazor et al. (2013). To estimate the annual 12
catch of artisanal fisheries (Ci) in each planning unit, the catch was assumed to 13
be proportional to the estuarine depth (Depth) and to the distance of the nearest 14
landing site (d) weighted exponentially by a constant decay rate α and then 15
multiplied by the area (Area) of the planning unit (i). In essence, it was assumed 16
that artisanal fishermen tend to concentrate in shallow waters near their villages 17
(Freitas and Tagliani, 2009). Four different values for α were used (0.0001, 18
0.001, 0.005 and 0.01) to reflect uncertainty in how far fishers travel to fish. The 19
depth was classified as cited above. The catch was normalized by a measure of 20
total effort (Reffort) which is equal to: 21
𝑅𝑅𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 = ∑𝑖𝑖=1𝑚𝑚 𝐷𝐷𝑒𝑒𝑒𝑒𝑒𝑒ℎ 𝑒𝑒−𝛼𝛼𝛼𝛼 x 𝐴𝐴𝑒𝑒𝑒𝑒𝑟𝑟, 22
where m is the number of planning units in the estuarine area of Patos Lagoon. 23
Then, the final value was multiplied by the total biomass of fish captured 24
11
(Rbiomass, ton) in the Patos Lagoon estuary during 2011 (IBAMA/CEPERG, 1
2012) multiplied by the average price (Cost, $US per ton, IBAMA/CEPERG, 2
2012) of three main target species (whitemouth croaker Micropogonias furnieri, 3
shrimp Farfantepenaeus paulensis, and mullets Mugil spp.) for artisanal fishery 4
in the Patos Lagoon estuary (which was $US 1,000 per ton) which comprises 5
88.5% of landings in the region (IBAMA/CEPERG, 2012). The final equation 6
used to calculate the opportunity cost for artisanal fishing in each planning unit 7
at the Patos Lagoon estuary is as follows: 8
𝐶𝐶𝑖𝑖 = �𝐷𝐷𝑒𝑒𝑒𝑒𝑒𝑒ℎ𝑒𝑒−𝛼𝛼𝛼𝛼𝐴𝐴𝑒𝑒𝑒𝑒𝑟𝑟
𝑅𝑅𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒� x 𝑅𝑅𝑏𝑏𝑖𝑖𝑒𝑒𝑚𝑚𝑟𝑟𝑎𝑎𝑎𝑎 x 𝐶𝐶𝑒𝑒𝑎𝑎𝑒𝑒 9
The four different opportunity cost layers for artisanal fisheries, each 10
with a different emphasis on the value of nearby fishing grounds, in the Patos 11
Lagoon estuary are given in Figure 2. 12
13
Conservation problems and scenarios 14
Two conservation issues were analysed. First, it was assumed that all 15
planning units were available for inclusion in the reserve system. Second, 16
planning units open for navigation, based on the “Framework for Water 17
Classification” (Class 3) from the Brazilian National Council of the 18
Environment (CONAMA, 2005), were locked out and not available for inclusion 19
in the reserve system. In this case, navigation areas include the area adjacent to 20
the port. This framework classifies marine, freshwater and brackish waters based 21
on their usage and divides brackish water into four classes: Special Class (for 22
which the primary purpose is the preservation of aquatic environments and the 23
maintenance of aquatic communities), Class 1 (waters that can be used for 24
12
recreational activities, aquaculture, fishing, human consumption after 1
conventional or advanced treatment and irrigation, and the protection of aquatic 2
communities); Class 2 (waters that can be used by fisheries and for recreational 3
activities); and Class 3 (waters allocated for navigation). Additionally, two 4
conservation targets were set for each issue: first, to protect 30% of each 5
conservation feature and second, to protect 50% of each habitat. Despite the 6
general use of a 10% target (CBD, 2014) for the protection of marine and coastal 7
environments, the present study followed a precautionary approach and tested 8
higher targets for the features. Different scenarios were run based on costs 9
(opportunity cost and flat cost) and ichthyoplankton data for each combination 10
of conservation issue and target. First, all scenarios were run by including 11
ichthyoplankton data in the analysis. Then, all scenarios were re-run by 12
excluding these data to assess the importance of incorporating the data from 13
different life stages. A complete hierarchical cluster analysis was performed in R 14
version 2.15.0 (R Development Core Team, 2013), using the hclust function 15
(stats package), to compare the resemblance within solutions from different 16
scenarios (Linke et al., 2011; Harris et al., 2014). A Jaccard resemblance matrix 17
constructed by the vegdist function (vegan package; Oksanen et al. 2015) was 18
used in this hierarchical cluster analysis. This resemblance method was chosen 19
because it is constructed for binary data (e.g. presence or absence), which is the 20
format of Marxan solutions (Harris et al., 2014). Also, a non-metric multi-21
dimensional scaling (nMDS) was performed to better visualize the solutions for 22
each scenario. This analysis was based on the Jaccard resemblance matrix using 23
the metaMDS function from the vegan package (Oksanen et al. 2015). 24
13
Systematic conservation planning analysis 1
Marxan, a decision support tool, was used to assess the importance of 2
incorporating different life stages of fishes into estuarine spatial planning. This 3
tool uses a simulated annealing algorithm to find alternative solutions for 4
conservation planning problem that meets conservation targets at a low socio-5
economic cost (Ball et al., 2009). Marxan was run 100 times each with 6
1,000,000 iterations. Input parameters were set after calibration, a “boundary 7
length modifier” (BLM) of 10 was adopted to provide reasonable clumping 8
(McDonnell et al., 2002), and the “species penalty factor” was set by increasing 9
the value until all targets were met. These input parameters were constant for all 10
scenarios to enable fair comparison. As each run of Marxan produces a different 11
solution, the selection frequency output, which comprises the number of times 12
that a planning unit was selected across all 100 runs, was chosen to represent the 13
conservation priority of a planning unit in each scenario (Ardron et al., 2008). 14
15
RESULTS 16
Exploratory analysis of spatial prioritization solutions 17
Incorporating fish egg and larvae information into spatial conservation 18
planning has a major effect on spatial priorities. The results of the cluster 19
analysis based on the Marxan solutions showed that most scenarios that used 20
artisanal fishery as an opportunity cost in the Patos Lagoon estuary formed two 21
main groups, splitting the solutions between scenarios that included or excluded 22
ichthyoplankton data, showing that the inclusion of this information produces 23
distinct spatial priorities for conservation. Dissimilarity among groups varied 24
14
within scenarios tested, but in all cases the main split divides solutions that 1
included or excluded ichthyoplankton data. Using a fishing cost decay rate of 2
0.0001, regardless of the conservation target (30% or 50%) and whether 3
planning units are locked out of the reserve system, the priorities form two 4
clusters driven by whether or not fish egg and larvae data are used (Figures 3 5
and 4).Usually, dissimilarity among groups ranged among 20% to 60%. 6
Considering the planning scenarios tested, solutions including ichthyoplankton 7
were most dissimilar to solutions excluding ichthyoplankton when considering 8
conservation target as 50% and planning units open for navigation not available 9
for inclusion in the reserve system (Figure 4C and 4D). Cluster analysis results 10
from different cost layers are presented in Figures S1, S2 and S3 under 11
Supporting Information. In general, it was observed that regardless of the 12
opportunity cost used, the priorities are substantially affected by the inclusion of 13
data on fish eggs and larvae. 14
One exception to this general pattern was found when there is a 15
conservation target of 50% for every habitat, and all planning units are available 16
for inclusion in the reserve system (Figure S2C). It was found that some 17
solutions from this scenario with ichthyoplankton data were grouped with 18
solutions from scenarios without these data. However, when a flat cost was 19
considered instead of an opportunity cost for artisanal fishery, this split 20
patterning disappears, showing that in this case, including fish eggs and larvae 21
data might not be important (Figure 5 and 6). 22
When analysing average cost and the percentage of the target met in each 23
planning design, scenarios considering all planning units available for inclusion 24
15
in the reserve system were cheaper than those excluding planning units used for 1
navigation from the reserve system (Table 2). Additionally, regardless of cost 2
and conservation target, scenarios where all planning units are available for 3
inclusion in the reserve system exhibited a lower total cost and a higher 4
percentage of targets met when incorporating ichthyoplankton data than when 5
they were excluded. However, planning designs excluding planning units used 6
for navigation showed a different pattern. In this case, scenarios excluding 7
ichthyoplankton data showed a lower cost and a lower number of planning units 8
selected in the reserve system than when including these data but with a lower 9
percentage of target met (Table 2). This last pattern is associated with the 10
decrease in sites available for conservation in the estuary, making it more 11
expensive to protect a higher number of features. 12
13
Spatial Prioritization Analysis: Selection Frequency Results 14
The selection frequency of each planning unit in each scenario, run with 15
ichthyoplankton data and using a decay rate from fishing site of 0.0001, is 16
represented in Figure 7. These results show that high percentages of site 17
selection can be found either clustered or dispersed (Figures 7, S4, S5 and S6). 18
Additionally, it was observed that regardless of the conservation target, shallow 19
areas in the upper part of the estuary and in the middle embayment had among 20
the highest selection frequency (Figure 7). However, when analysing the 21
selection frequency results in the case of a flat cost, clustered sites in the centre 22
of the estuarine area had among the highest selection frequency (Figure 8). In 23
general, if we consider only planning units with selection frequency higher than 24
16
80% regardless the planning scenario tested, shallow and intermediate waters are 1
the most important estuarine habitats for conservation in Patos Lagoon estuary 2
(Figure 9). However, their importance to conservation decreases in planning 3
scenarios excluding ichthyoplankton data (Figure 9). When analysing the 4
importance of each depth class in different planning scenarios, we observed that 5
the contribution of habitats varied within cost layer, conservation target, and 6
scenarios including or excluding ichthyoplankton (Figure 10). Despite the 7
variation within planning scenarios tested, shallow and intermediate waters 8
encompasses high priority sites to conservation in the Patos Lagoon estuary. 9
10
DISCUSSION 11
This study shows that information about the early life history stages of 12
fishes and opportunity costs based on artisanal fisheries can help to achieve a 13
more cost-effective outcome in the Patos Lagoon estuary. Furthermore, using 14
fish egg and larvae data in conservation planning substantially changes priorities 15
for estuarine conservation. Usually, planktonic stages have a wide distribution 16
and depend on nursery areas, such as estuaries, and spawning sites. In this case, 17
understanding and including the spatial distribution of these life-stages is 18
extremely relevant to increase the effectiveness of reserves (Allison et al., 1998; 19
Warner et al., 2000). Ichthyoplankton represents the most vulnerable stage of 20
the fish life cycle, and mortality rates during this phase can reach almost 99% 21
for most species (Fuiman and Werner, 2002). The mortality of fish eggs and 22
larvae can be associated with various processes, such as predation, nutrition, 23
diseases, unfavourable environmental conditions, anthropogenic causes such as 24
17
habitat alteration or destruction, and pollution. Protected areas offer a spatially 1
explicit form of protection that enables local control of anthropogenic activities, 2
helping to mitigate numerous pressures such as fishing, pollution threats and 3
habitat disturbance (Allison et al., 1998). Estuaries are among the most 4
important sites for the development of the early life history of fishes. These 5
ecosystems often suffer from anthropogenic impacts that can potentially destroy 6
estuarine habitat (McLusky and Elliot, 2006) and affect nursery functioning for 7
many species (Courrat et al. 2009). The protection of the habitat used by early 8
life stages of fish is essential for ensuring success in the recruitment process and 9
to sustain the adult population. 10
Despite the ecological importance of estuaries, estuarine conservation is 11
in its infancy relative to terrestrial and marine realms. This may be related to the 12
high social and economic values of these areas for humans and also a lower 13
aesthetic appeal compared with other ecosystems (Edgar et al., 2000; Neely and 14
Zajac, 2008). Systematic conservation planning studies in estuaries are restricted 15
to a few papers, most of them using macrobenthic invertebrates or habitats as 16
surrogates of biodiversity (Neely and Zajac, 2008; Shokri and Gladstone, 2009, 17
2013a; Shokri et al., 2009). Until now, ichthyoplankton or fish data have not 18
been used in estuarine conservation planning. The results showed that 19
incorporating ichthyoplankton data into systematic conservation planning can 20
generate different outcomes than a plan without these data (Figures 3 and 4). 21
Therefore, estuarine conservation plans that solely rely on habitat data for 22
surrogates of biodiversity can produce suboptimal results. In a recent study, 23
Shokri and Gladstone (2013a) found that habitat classification schemes that 24
18
represent variation in estuarine biodiversity are inefficient at effectively 1
designing estuarine protected areas. One of the main aspects of estuaries is the 2
interaction between land and sea processes that creates a complex and dynamic 3
spatial pattern for biological, physical, chemical, and socio-economic 4
components (Pittman et al., 2011). This means that an efficient planning scheme 5
must account for ecological dynamics in a spatially explicit framework (Pittman 6
et al., 2011). Future studies should include some analysis relating the 7
distribution of planktonic stages, such as fish eggs and larvae, to other dynamic 8
physico-chemical features of the estuary. Also, future studies should consider 9
identifying the importance of an area by using data on density of fish eggs and 10
larvae rather than range data based on presence/absence data. However, the 11
approach used in this study is useful to managers by showing that it is also 12
relevant to include early life stages of fishes into spatial planning, and also, by 13
offering a start point in evaluating which areas might be important for protecting 14
biodiversity of Patos Lagoon estuary. Furthermore, other socio-economic 15
components with more direct impacts on fish eggs and larvae development, such 16
as water quality, should be examined in future estuarine conservation planning 17
studies. 18
The Patos Lagoon estuary has been subject to substantial natural and 19
human disturbances. The construction of rocky jetties at the entrance of the 20
estuary may have altered the input and output water flux (Seeliger and 21
Odebrecht, 2010), consequently influencing the transport of fish eggs and larvae 22
into the estuary. Despite the strong seasonal pattern in recruitment of fish eggs 23
and larvae to the Patos Lagoon estuary, long-term studies suggest the existence 24
19
of decadal variations in climate due to El Niño Southern Oscillation (Muelbert et 1
al., 2010). El Niño (high precipitation) and La Niña (low precipitation) events 2
interact with wind patterns in the region and influence the salinity in this 3
ecosystem and consequently control the occurrence and abundance of fish eggs 4
and larvae (Bruno and Muelbert, 2009). Despite these disturbances, the study 5
period showed consistent patterns in salinity and consequently in the recruitment 6
of such early life stages of fishes into the Patos Lagoon estuary (Muelbert et al., 7
2010). Therefore, the scenarios tested in the present study that incorporated 8
historical ichthyoplankton assessments represent the general pattern of the 9
recruitment process into the study area. In this sense, historical data may be 10
acceptable for inclusion in studies that aim to develop new methods for spatial 11
planning. However, it is necessary to note that historical data were only used 12
because there was no recent information available for the entire estuarine area. 13
Spatial planning for conservation purposes should be based, where possible, on 14
recent information for the region it intends to conserve. 15
While the present study highlights the advantage of using 16
ichthyoplankton data for estuarine conservation, it also emphasises the 17
importance of incorporating opportunity cost instead of a spatially homogeneous 18
(flat) cost. The results showed that if a flat cost is used (Figures 5, 6 and 8), as is 19
still common in the conservation planning literature, using just abiotic features 20
(e.g. habitat, bathymetry and sediment) is sufficient for achieving conservation 21
targets. In contrast, when a more realistic socio-economic cost was used instead 22
of a flat cost, it was possible to achieve a more effective outcome. Additionally, 23
it was observed that the abiotic features were not adequate surrogates for the 24
20
biotic elements that were the focus in the planning. This confirms the findings in 1
the study by Shokri and Gladstone (2013a), which highlighted that habitat alone 2
is an inefficient surrogate for estuarine conservation. Similarly, Stewart and 3
Possingham (2005) outlined the importance of considering opportunity cost in 4
conservation planning and creating representative, efficient, and practical 5
reserve systems that minimize potential economic losses. In the case of the Patos 6
Lagoon estuary, artisanal fisheries and port activity are the main economic uses 7
of the estuarine area; thus, a conservation plan aiming towards a representative 8
and efficient reserve system must take these activities into consideration. In this 9
sense, scenarios using opportunity cost which also excluded planning units used 10
for navigation from the reserve system represents a more realistic solution for 11
the study area as they minimize fishery loss, avoid prioritizing reserves in the 12
navigation area in the estuary and still achieve conservation targets. 13
The Patos Lagoon estuary is predominantly a shallow estuarine 14
ecosystem (Seeliger, 2001). The results showed that shallow waters (vegetated 15
or not) are among the most selected sites for conservation in the Patos Lagoon 16
estuary (Figures 1, 9, 10). Shallow areas in the upper part of the estuary and in 17
the middle embayment were selected in all scenarios, regardless of the 18
objectives, indicating the importance of these sites for early life fish 19
development. In Patos Lagoon, most shallow waters are colonised by the 20
seagrass Ruppia maritima, which develops during warmer seasons (spring and 21
summer) and plays an important role in feeding, development and protection for 22
the early life stages of many fish and invertebrate species (Castello, 1986; 23
Garcia and Vieira, 1997; Seeliger, 2001). In addition, vegetated shallow waters 24
21
are cited as an important factor structuring fish abundance and composition in 1
estuaries worldwide (Gillanders, 2006). Despite their ecological importance, 2
shallow areas in the Patos Lagoon estuary are also impacted by pollution and 3
contamination from surrounding urban area, dredging, and the loss of seagrass 4
beds and saltmarshes (Seeliger and Costa, 1997; Barletta et al., 2010). 5
Consequently, the efficient management of estuarine shallow waters, vegetated 6
or not, is of paramount importance for conservation and must be accounted for 7
when addressing estuary protection. 8
In summary, if fish egg and larva data are ignored in the planning 9
process, conservation priorities will be very different. Furthermore, including 10
temporal variability in estuarine biodiversity into systematic conservation 11
planning can be beneficial for achieving an effective outcome because it 12
considers the variations that a species can have in their distribution within 13
different scales of time and space. Considering the present case study, shallow 14
waters are of great importance and must be considered for estuarine 15
conservation and incorporated into estuarine protected areas. Additionally, it is 16
suggested that future studies should incorporate information from adjacent 17
coastal and freshwater ecosystems to create a conservation plan for the entire 18
estuarine ecosystem. The present study also shows that opportunity costs can 19
help achieve an efficient and representative reserve system and highlights the 20
need to incorporate social and economic information into reserve design 21
(Naidoo et al., 2006; Carwardine et al., 2008; Klein et al., 2008a, b; Ban and 22
Klein, 2009). More studies should focus on developing methods for planning 23
conservation in estuarine ecosystems with multiple uses and stressors. 24
22
ACKNOWLEDMENTS 1
We thank M.S. Copertino, C.R.A. Tagliani, L.J. Calliari and P.R. Tagliani, who 2
kindly provided spatial information about habitat/vegetation, sediment, 3
bathymetry, and fishery activity in Patos Lagoon estuary; M.E. Watts for 4
helping in the cluster analysis and T. Mazor for insightful comments on previous 5
version of the manuscript. M.D.C.P. was financially supported by the National 6
Council of Scientific and Technological Development (CNPq) with a post-7
graduate scholarship, and J.H.M. received a CNPq grant (Proc. 310931/2012-6). 8
H.P.P. was supported by Australian Research Council fellowships and the ARC 9
Centre of Excellence for Environmental Decisions. 10
11
REFERENCES 12
Allison GW, Lubchenco J, Carr MH. 1998. Marine reserves are necessary but 13
not sufficient for marine conservation. Ecological Applications 8: S79-S92. 14
Ardron JA, Possingham HP, Klein CJ. 2008. Marxan Good Practices 15
Handbook: external review version. Pacific Marine Analysis and Research 16
Association: Vancouver. 17
Babler SJM. 2000. Tropical estuarine fishes: ecology, explotation and 18
conservation. Blackwell Science: Malden, MA. 19
Ball IR, Possingham HP, Watts M, 2009. Marxan and relatives: software for 20
spatial conservation prioritization. In Spatial conservation prioritization: 21
quantitative methods and computational tools, Moilanen A, Wilson KA, 22
Possingham HP (eds). Oxford University Press: Oxford; 185-195. 23
23
Ban NC, Klein CJ. 2009. Spatial socioeconomic data as a cost in systematic 1
marine conservation planning. Conservation Letters 2: 206-215. 2
Barbier EB, Hacker SD, Koch EW, Stier AC, Silliman BR, 2011. Estuarine and 3
coastal ecosystems and their services. In Treatise on estuarine and coastal 4
science, Wolanski E, McLusky, D (eds). Elsevier: Oxford; 12: 109-127. 5
Barletta M, Jaureguizar AJ, Baigun C, Fontoura NF, Agostinho AA, Almeida-6
Val VMF, Torres RA, Jimenes-Segura LF, Giarrizzo T, Fabré NN, et al. 2010. 7
Fish and aquatic habitat conservation in South America: a continental overview 8
with emphasis on neotropical systems. Journal of Fish Biology 76: 2118-2176. 9
Berasategui AD, Acha EM, Fernandéz Araoz NC. 2004. Spatial patterns of 10
ichthyoplankton assemblages in the Río de La Plata estuary (Argentina – 11
Uruguay). Estuarine, Coastal and Shelf Science 60: 599-610. 12
Boehlert GW, Mundy BC. 1988. Roles of behavioral and physical factors in 13
larval and juvenile fish recruitment to estuarine nursery areas. American 14
Fisheries Society Symposium 3: 51-67. 15
Bruno MA, Muelbert JH. 2009. Distribuição espacial e variações temporais da 16
abundância de ovos e larvas de Micropogonias furnieri no estuário da Lagoa dos 17
Patos: registros históricos e forçantes ambientais. Atlântica 31: 49-66. 18
Calliari L, Gomes MEV, Griep GH, Möller Jr. OO. 1980. Características 19
sedimentológicas e fatores ambientais da região estuarial da Lagoa dos Patos. 20
Anais do XXXI Congresso Brasileiro de Geologia 2: 862-875. 21
CBD. 2014. Global Biodiversity Outlook – Marine and coastal ecosystems. 22
Convention on Biological Diversity, UNEP. http://www.cbd.int/ [20 June 2014]. 23
24
Castello JP. 1986. Distribucion, crecimiento y maduracion sexual de La corvina 1
juvenil (Micropogonias furnieri) en el estuário de la Lagoa dos Patos, Brasil. 2
Physis 44: 21-36. 3
Carwardine J, Wilson KA, Watts M, Etter A, Klein CJ, Possingham, HP. 2008. 4
Avoiding costly conservation mistakes: the importance of defining actions and 5
costs in spatial priority setting. PLoS ONE 3: e2586. 6
CONAMA. 2005. Conselho Nacional do Meio Ambiente: Resolução número 7
357. Ministério do Meio Ambiente, Brasília. 8
Costa MDP, Muelbert JH, Moraes LE, Vieira JP, Castello JP. 2014. Estuarine 9
early life stage habitat occupancy patterns of whitemouth croaker 10
Micropogonias furnieri (Desmarest, 1830) from the Patos Lagoon, Brazil. 11
Fisheries Research 160: 77-84. 12
Courrat A, Lobry J, Nicolas D, Laffargue P, Amara R, Lepage M, Girardin M, 13
Le Pape O. 2009. Anthropogenic disturbance on nursery function of estuarine 14
areas for marine species. Estuarine, Coastal and Shelf Science 81: 179-190. 15
Edgar GJ, Barrett NS, Last PR. 2000. The conservation significance of estuaries: 16
a classification of Tasmanian estuaries using ecological, physical and 17
demographic attributes as a case study. Biological Conservation 92: 383-397. 18
Elliott M, Hemingway KL. 2002. Fishes in estuaries. Blackwell Science: 19
Oxford. 20
Faria A, Morais P, Chícharo MA. 2006. Ichthyoplankton dynamics in the 21
Guadiana estuary and adjacent coastal area, South-East Portugal. Estuarine, 22
Coastal and Shelf Science 70: 85-97. 23
25
Freitas DM, Tagliani PRA. 2009. The use of GIS for the integration of 1
traditional and scientific knowledge in supporting artisanal fisheries 2
management in southern Brazil. Journal of Environmental Management 90: 3
2071-2080. 4
Fuiman LA, Werner RG. 2002. Fishery Science: the unique contributions of 5
early life stages. Blackwell Science: Oxford. 6
Garcia AM, Vieira JP. 1997. Abundância e diversidade da assembléia de peixes 7
dentro e fora de uma pradaria de Ruppia maritima L., no estuário da Lagoa dos 8
Patos (RS – Brasil). Atlântica 19: 161-181. 9
Geselbracht L, Torres R, Cumming GS, Dorfman D, Beck M, Shaw D. 2009. 10
Identification of a spatially efficient portfolio of priority conservation sites in 11
marine and estuarine areas of Florida. Aquatic Conservation: Marine and 12
Freshwater Ecosystems, 19: 408-420. 13
Gillanders BM, 2006. Seagrasses, fish and fisheries. In Seagrasses: Biology, 14
Ecology and Conservation, Larkum, AWD, Orth, RJ, Duarte, CM (eds), 15
Springer: Netherlands; 503–536. 16
Govoni JJ. 2005. Fisheries oceanography and the ecology of early life histories 17
of fishes: a perspective over fifty years. Scientia Marina 69: 125-137. 18
Harris LR, Watts ME, Nel R, Schoeman DS, Possingham HP. 2014. Using 19
multivariate statistics to explore trade-offs among spatial planning scenarios. 20
Journal of Applied Ecology 51: 1504-1514. 21
Houde ED. 2008. Emerging from Hjort’s shadow. Journal of Northwest Atlantic 22
Fishery Science 41: 53-70. 23
26
IBAMA/CEPERG. 2012. Desembarque de pescado no Rio Grande do Sul: 1
2011. Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais 2
Renováveis. Centro de Pesquisa e Gestão dos Recursos Pesqueiros Lagunares e 3
Estuarinos: Projeto Estatística Pesqueira. Rio Grande. 4
Klein CJ, Steinback C, Scholz AJ, Possingham HP. 2008a. Effectiveness of 5
marine reserves networks in representing biodiversity and minimizing impact to 6
fishermen: a comparison of two approaches used in California. Conservation 7
Letters 1: 44-51. 8
Klein CJ, Chan A, Kircher L, Cundiff AJ, Gardner N, Hrovat Y, Scholz AJ, 9
Kendall BE, Airamé S. 2008b. Striking a balance between biodiversity 10
conservation and socioeconomic viability in the design of marine protected 11
areas. Conservation Biology 3: 691-700. 12
Le Pape O, Delavenne J, Vaz S. 2014. Quantitative mapping of fish habitat: A 13
useful tool to design spatialised management measures and marine protected 14
area with fishery objectives. Ocean and Coastal Management 87: 8-19. 15
Linke S, Watts M, Steward R, Possingham HP. 2011. Using multivariate 16
analysis to deliver conservation planning products that align with practitioner 17
needs. Ecography 34: 203-207. 18
Mai ACG, Condini MV, Albuquerque CQ, Loebman D, Saint’Pierre TD, 19
Miekeley N, Vieira, JP. 2014. High plasticity in habitat use of Lycengraulis 20
grossidens (Clupeiformes, Engraulididae). Estuarine, Coastal and Shelf Science 21
141: 17-25. 22
27
Martins IM, Dias JM, Fernandes EH, Muelbert JH. 2007. Numerical modelling 1
of fish eggs dispersion at the Patos Lagoon estuary – Brazil. Journal of Marine 2
System 68: 537-555. 3
Mazor T, Possingham, HP, Kark S. 2013. Collaboration among countries in 4
marine conservation can achieve substantial efficiencies. Diversity and 5
Distributions 19: 1-14. 6
Mazor T, Giakoumi S, Kark S, Possingham HP. 2014. Large scale conservation 7
planning in a multinational marine environment: cost matters. Ecological 8
Applications 24: 1115-1130 9
McDonnell MD, Possingham HP, Ball IR, Cousins EA. 2002. Mathematical 10
methods for spatially cohesive reserve design. Environmental Modeling and 11
Assessment 7: 107-114. 12
McLusky DS, Elliott M. 2006. The estuarine ecosystem: ecology, threats and 13
management. Oxford University Press: Oxford, NY. 14
Moilanen A, Possingham HP, Polasky S, 2009. A mathematical classification of 15
conservation prioritization problems. In Spatial conservation prioritization: 16
quantitative methods and computational tools, Moilanen A, Wilson KA, 17
Possingham HP (eds). Oxford University Press: Oxford; 28-42 18
Muelbert JH, Weiss G. 1991. Abundance and distribution of fish larvae in the 19
channel area of Patos Lagoon estuary, Brazil. NOAA Technical Report NMFS 20
95: 43-54. 21
28
Muelbert JH, Muxagata E, Kaminski S, 2010. As comunidades zooplanctônicas. 1
In O estuário da Lagoa dos Patos: Um século de transformações, Seeliger U, 2
Odebrecht C. (eds). Editora FURG: Rio Grande; 67- 75. 3
Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR, O’Hara RB, 4
Simpson GL, Solymos P, Stevens MHH, Wagner H. 2015. Vegan: Community 5
Ecology Package. R package version 2.2-1. http://CRAN.R-6
project.org/package=vegan. 7
Naidoo R, Balmford A, Ferraro PJ, Polasky S, Ricketts TH, Rouget M. 2006. 8
Integrating economic costs into conservation planning. TRENDS in Ecology and 9
Evolution 21: 681-687. 10
Neely AE, Zajac RN. 2008. Applying marine protected area design models in 11
large estuarine systems. Marine Ecology Progress Series 373: 11-23. 12
Odebrecht C, Abreu PC, Bemvenuti CE, Copertino M, Muelbert JH, Vieira JP, 13
Seeliger U, 2010. The Patos Lagoon Estuary, Southern Brazil: biotic responses 14
to natural and anthropogenic impacts in the last decades (1979-2008). In Coastal 15
Lagoons: Critical habitats of environmental changes, Kennish MJ, Paerl, PW 16
(eds). CRC Press: Boca Raton; 433-455. 17
Pittman SJ, Connor DW, Radke L, Wright DJ, 2011. Application of estuarine 18
and coastal classifications in marine spatial management. In Treatise on 19
estuarine and coastal science, Wolanski E, McLusky, D (eds). Elsevier: Oxford; 20
11: 164-205. 21
Phillips SJ, Dudik M. 2008. Modeling of species distribution with Maxent: new 22
extensions and a comprehensive evaluation. Ecography 31: 161-175. 23
29
Plano Ambiental de Rio Grande. 2007. Programa Costa Sul: Plano Ambiental 1
Municipal de Rio Grande, Rio Grande. 2
Primo AL, Azeiteiro UM, Marques SC, Pardal MA. 2011. Impact of climate 3
variability on ichthyoplankton communities: an example of a small temperate 4
estuary. Estuarine, Coastal and Shelf Science 91: 484-491. 5
QGIS Development Team. 2013. QGIS Geographic Information System. Open 6
Source Geospatial Foundation Project. http://qgis.osgeo.org. 7
R Development Core Team. 2013. R: A language and environment for statistical 8
computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-9
90051-07-0. URL http://www.R-project.org/. 10
Ramos S, Cowen RK, Paris C, Ré P, Bordalo AA. 2006. Environmental forcing 11
and larval fish assemblage dynamics in the Lima river estuary (northwest 12
Portugal). Journal of Plankton Research 28: 275-286. 13
Schafer AG, Reis EG. 2008. Artisanal fishing areas and traditional ecological 14
knowledge: the case study of the artisanal fisheries of the Patos Lagoon estuary 15
(Brazil). Marine Policy 32: 283-292. 16
Schultz ET, Lwiza KMM, Fencil MC, Martin JM. 2003. Mechanisms promoting 17
upriver transport of larvae of two fish species in the Hudson River estuary. 18
Marine Ecology Progress Series 251: 263-277. 19
Seeliger U, Costa CSB, 1997. Natural and Human Impact. In Subtropical 20
Convergence Environments: The coast and sea in the Southwestern Atlantic, 21
Seeliger U, Odebrecht C, Castello JP (eds), Springer-Verlag Heidelberg: New 22
York; 197-203. 23
30
Seeliger U, 2001. The Patos Lagoon Estuary, Brazil. In Coastal Marine 1
Ecosystems of Latin America, Seeliger U, Kjerve, B (eds), Springer-Verlag: 2
Berlin; 167-183. 3
Seeliger U, Odebrecht C. 2010. O estuário da Lagoa dos Patos: Um século de 4
transformações. Editora FURG: Rio Grande. 5
Simionato CG, Berasategui A, Meccia VL, Acha EM, Mianzan H. 2008. Short 6
time-scale wind forced variability in the Río de la Plata estuary and its role on 7
ichthyoplankton retention. Estuarine, Coastal and Shelf Science 76: 211-226. 8
Sinque C, Muelbert JH, 1997. Ichthyoplankton. In Coastal Marine Ecosystems 9
of Latin America, Seeliger U, Kjerve, B (eds), Springer-Verlag: Berlin; 51-56. 10
Shokri MR, Gladstone W. 2009. Higher taxa are effective surrogates for species 11
in the selection of conservation reserves in estuaries. Aquatic Conservation: 12
Marine and Freshwater Ecosystems 19: 626-636. 13
Shokri R, Gladstone W, Kepert A. 2009. Annelids, arthropods, and molluscs are 14
suitable as surrogate taxa for selecting conservation reserves in estuaries. 15
Biodiversity Conservation 18: 1117-1130. 16
Shokri MR, Gladstone W. 2013a. Limitations of habitats as biodiversity 17
surrogates for conservation planning in estuaries. Environmental Monitoring 18
Assessing 185: 3477-3492. 19
Shokri MR, Gladstone W. 2013b. Integrating vulnerability into estuarine 20
conservation planning: does the data treatment method matters? Estuaries and 21
Coasts 36: 866-880. 22
31
Smith RJ, Goodman PS, Matthews WS. 2006. Systematic conservation 1
planning: A review of perceived limitations and actual benefits, using a case 2
study from Maputaland, South Africa. Oryx 40: 400–410. 3
Stewart RR, Possingham HP. 2005. Efficiency, cost and trade-offs in marine 4
reserve system design. Environmental Modeling and Assessment 10: 203-213. 5
Swets JA. 1988. Measuring the accuracy of diagnostic systems. Science 240: 6
1285-1293. 7
Warner RS, Swearer SE, Caselle JE. 2000. Larval accumulation and retention: 8
implications for the design of marine reserves and essential fish habitat. Bulletin 9
of Marine Science 66: 821-830. 10
Watson JEM, Grantham H, Wilson KA, Possingham HP, 2011. Systematic 11
Conservation Planning: Past, Present and Future. In Conservation 12
Biogeography, Whittaker R, Ladle R (eds), Wiley-Blackwell: Oxford; 136-160. 13
Wilson KA, Carwardine J, Possingham HP. 2009a. Setting Conservation 14
Priorities. Annals of the New York Academy of Sciences 1162: 237-264. 15
Wilson KA, Cabeza M, Klein CJ, 2009b. Fundamental concepts of spatial 16
conservation prioritization. In Spatial conservation prioritization: quantitative 17
methods and computational tools, Moilanen A, Wilson KA, Possingham HP 18
(eds), Oxford University Press: Oxford; 16 - 27. 19
20
Additional Supporting Information may be found in the online version of this 21 article: 22
23 Figure S1: Relationship among 100 solutions for each scenario with (solutions 1 24 to 100) and without (solutions 101 to 200) ichthyoplankton data run in Marxan, 25 using cost layer with constant α = 0.001. 26 27
32
Figure S2: Relationship among 100 solutions for each scenario with (solutions 1 1 to 100) and without (solutions 101 to 200) ichthyoplankton data run in Marxan, 2 using cost layer with constant α = 0.005. 3 Figure S3: Relationship among 100 solutions for each scenario with (solutions 1 4 to 100) and without (solutions 101 to 200) ichthyoplankton data run in Marxan, 5 using cost layer with constant α = 0.01. 6 7 Figure S4: Selection frequency result displayed for each scenario with 8 ichthyoplankton data, using cost layer with constant α = 0.001. 9 10 Figure S5: Selection frequency result displayed for each scenario with 11 ichthyoplankton data, using cost layer with constant α = 0.005. 12 13 Figure S6: Selection frequency result displayed for each scenario with 14 ichthyoplankton data, using cost layer with constant α = 0.01. 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
33
Tables 1 2 Table 1: Description of the features that were used to assess the role of 3 incorporating fish egg and larva data for estuarine conservation in the Patos 4 Lagoon estuary, Brazil (e: eggs; l: larvae). 5
Feature Data Number of layers Description
Bathymetry Polygon 3 Shallow, intermediate and deep waters
Sediment Polygon 7 Classification based on Calliari et al. (1980)
Habitat Interpolated seasonal survey 16 Ruppia maritima, Zannichelia palustris,
macroalgae, meadow height(*)
Ichthyoplankton species
Output of species
distribution modelling**
(MAXENT)
38
Achirus garmani (e/l), Brevoortia pectinada (e/l), Licengraulis grossidens (e/l), Micropogonias furnieri (e/l), Parapimelodus valenciennes (l), Trichiurus lepturus (e)
(*) meadow height was classified into 5 classes of height (cm): 0 – 10, 11 – 20, 6 21 – 40, 41 – 60, and 61 – 150. 7 (**) only MAXENT outputs with AUC > 0.75 were used in the conservation 8 planning. 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
34
Table 2: Results showing average cost, percentage of conservation target met, 1 average of planning units selected, and main cluster results for each scenario run 2 in Marxan (cost 1: cost layer with constant α = 0.0001; cost 2: cost layer with 3 constant α= 0.001; and cost 3: cost layer with constant α = 0.005 and cost 4: cost 4 layer with constant α = 0.01). Cost 5 represents the cost layer with a 5 homogenous value for all planning units (flat cost) and does not have an 6 economic value (in table is represented by (-)). 7
8
9 10
Planning Scenario Cost (US$ million)
Targets met (%)
Number of PU
selected
Planning Ia – Including Ichthyoplankton 30% conservation target and all planning units available for inclusion in the reserve system
cost 1: 92,455 96.87 686 cost 2: 92,412 95.31 685 cost 3: 92,338 93.75 686 cost 4: 92,307 98.43 686 cost 5: - 100 957
Planning Ia – Excluding Ichthyoplankton 30% conservation target and all planning units available for inclusion in the reserve system
cost 1: 94,939 92.31 684 cost 2: 95,008 88.46 684 cost 3: 94,973 100 684 cost 4: 94,933 96.15 684 cost 5: - 96.15 768
Planning Ib – Including Ichthyoplankton 30% conservation target and planning units open for navigation not available for inclusion in the reserve system
cost 1: 116,955 93.75 670 cost 2: 117,530 89.06 670 cost 3: 116,930 96.87 670 cost 4: 116,873 89.06 669 cost 5: - 96.87 1043
Planning Ib – Excluding Ichthyoplankton 30% conservation target and planning units open for navigation not available for inclusion in the reserve system
cost 1: 112,406 92.30 672 cost 2: 112,325 88.46 672 cost 3: 112,407 88.46 672 cost 4: 112,540 96.30 672 cost 5: - 92.30 907
Planning IIa – Including Ichthyoplankton 50% conservation target and all planning units available for inclusion in the reserve system
cost 1: 98,372 95.31 705 cost 2: 98,420 96.87 705 cost 3: 98,338 93.75 705 cost 4: 98,229 95.31 705 cost 5: - 98.43 1000
Planning IIa – Excluding Ichthyoplankton 50% conservation target and all planning units available for inclusion in the reserve system
cost 1: 98,671 96.15 705 cost 2: 98,595 96.15 705 cost 3: 98,566 84.61 707 cost 4: 98,603 88.46 704 cost 5: - 100 858
Planning IIb – Including Ichthyoplankton 50% conservation target and planning units open for navigation not available for inclusion in the reserve system
cost 1: 128,924 92.18 694 cost 2: 128,848 93.75 694 cost 3: 129,005 92.18 693 cost 4: 129,004 89.06 695 cost 5: - 98.43 1101
Planning IIb – Excluding Ichthyoplankton 50% conservation target and planning units open for navigation not available for inclusion in the reserve system
cost 1: 98,607 88.46 705 cost 2: 96,685 92.30 693 cost 3: 98,457 96.15 704 cost 4: 98,563 92.30 705 cost 5: - 100 872
35
Figure Captions 1 2 Figure 1: Location of the Patos Lagoon estuary in Rio Grande do Sul State (RS, 3 Brazil) and details of the bathymetry in the estuarine area. 4 5 Figure 2: Cost layers of artisanal fishery within each fishing area in the Patos 6 Lagoon estuary: (a) artisanal fishery cost layer with constant α = 0.0001, (b) 7 artisanal fishery cost layer with constant α = 0.001, (c) artisanal fishery cost 8 layer with constant α = 0.005, and (d) artisanal fishery cost layer with constant α 9 = 0.01. 10 11 Figure 3: Relationship among 100 solutions for each scenario with (solutions 1 12 to 100) and without (solutions 101 to 200) ichthyoplankton data run in Marxan. 13 The cost layer used in this analysis represents the opportunity cost of artisanal 14 fishery in Patos Lagoon estuary with constant α = 0.0001. Data are presented as 15 a dendrogram from a complete hierarchical cluster analysis (A and C) and an 16 nMDS biplot (B and D) based on Jaccard resemblance matrix. This scenario 17 represents the conservation problem under a conservation target of 30% for all 18 features and all planning units available for inclusion in the reserve system (A 19 and B); and conservation problem under a conservation target of 30% for all 20 features and planning units open for navigation not available for inclusion in the 21 reserve system (C and D). 22 23 Figure 4: Relationship among 100 solutions for each scenario with (solutions 1 24 to 100) and without (solutions 101 to 200) ichthyoplankton data run in Marxan. 25 The cost layer used in this analysis represents the opportunity cost of artisanal 26 fishery in Patos Lagoon estuary with constant α = 0.0001. Data are presented as 27 a dendrogram from a complete hierarchical cluster analysis (A and C) and an 28 nMDS biplot (B and D) based on Jaccard resemblance matrix. This scenario 29 represents the conservation problem under a conservation target of 50% for all 30 features and all planning units available for inclusion in the reserve system (A 31 and B); and conservation problem under a conservation target of 50% for all 32 features and planning units open for navigation not available for inclusion in the 33 reserve system (C and D). 34 35 Figure 5: Relationship among 100 solutions for each scenario with (solutions 1 36 to 100) and without (solutions 101 to 200) ichthyoplankton data run in Marxan. 37 The cost layer used in this analysis represents the flat cost for each planning unit 38 in the Patos Lagoon estuary. Data are presented as a dendrogram from a 39 complete hierarchical cluster analysis (A and C) and an nMDS biplot (B and D) 40 based on Jaccard resemblance matrix. This scenario represents the conservation 41 problem under a conservation target of 30% for all features and all planning 42 units available for inclusion in the reserve system (A and B); and conservation 43 problem under a conservation target of 30% for all features and planning units 44 open for navigation not available for inclusion in the reserve system (C and D). 45 46 47
36
Figure 6: Relationship among 100 solutions for each scenario with (solutions 1 1 to 100) and without (solutions 101 to 200) ichthyoplankton data run in Marxan. 2 The cost layer used in this analysis represents the flat cost for each planning unit 3 in the Patos Lagoon estuary. Data are presented as a dendrogram from a 4 complete hierarchical cluster analysis (A and C) and an nMDS biplot (B and D) 5 based on Jaccard resemblance matrix. This scenario represents the conservation 6 problem under a conservation target of 50% for all features and all planning 7 units available for inclusion in the reserve system (A and B); and conservation 8 problem under a conservation target of 50% for all features and planning units 9 open for navigation not available for inclusion in the reserve system (C and D). 10 11 Figure 7: Selection frequency result displayed for each scenario with 12 ichthyoplankton data. The cost layer used in this analysis represents the 13 opportunity cost of artisanal fisheries in the Patos Lagoon estuary with constant 14 α = 0.0001. (A: conservation problem under a conservation target of 30% for all 15 features and all planning units available for inclusion in the reserve system; B: 16 conservation problem under a conservation target of 50% for habitats and all 17 planning units available for inclusion in the reserve system; C: conservation 18 problem under a conservation target of 30% for all features and planning units 19 open for navigation not available for inclusion in the reserve system; and D: 20 conservation problem under a conservation target of 50% and planning units 21 open for navigation not available for inclusion in the reserve system). 22 23 Figure 8: Selection frequency result displayed for each scenario with 24 ichthyoplankton data. The cost layer used in this analysis represents the flat cost 25 for each planning unit in the Patos Lagoon estuary (A: conservation problem 26 under a conservation target of 30% for all features and all planning units 27 available for inclusion in the reserve system; B: conservation problem under a 28 conservation target of 50% for habitats and all planning units available for 29 inclusion in the reserve system; C: conservation problem under a conservation 30 target of 30% for all features and planning units open for navigation not 31 available for inclusion in the reserve system; and D: conservation problem under 32 a conservation target of 50% and planning units open for navigation not 33 available for inclusion in the reserve system). 34 35
Figure 9: Mean percentage of priority sites for conservation (planning units with 36 selection frequency ≥ 80%) in each depth class in Patos Lagoon estuary 37 considering all scenarios tested. Percentage number of planning units in each 38 depth class are: shallow waters (< 1.5 m) = 45 %, intermediate waters (1.5 – 5.0 39 m) = 43%, and deep waters (< 5m) = 12%). 40 41
Figure 10: Percentage of priority sites for conservation (planning units with 42 selection frequency ≥ 80%) for different planning scenarios tested including and 43 excluding ichthyoplankton in Patos Lagoon estuary, considering the Percentage 44 number of planning units in each depth class are: shallow waters (< 1.5 m) = 45 45 %, intermediate waters (1.5 – 5.0 m) = 43%, and deep waters (< 5m) = 12%). 46