Table of Contents

Title Page

Copyright
Contributors
Foreword
Preface

Part I: Spatial Patterning in the Sea

Chapter 1: Introducing Seascape Ecology
1. 1.1 Introduction
2. 1.2 Landscape Ecology and the Emergence of Seascape Ecology
3. 1.3 What is a Seascape?
4. 1.4 Why Scale Matters in Seascape Ecology
5. 1.5 Seascape Ecology can Inform Marine Stewardship
6. 1.6 Conclusions and Future Directions
7. References
Chapter 2: Mapping and Quantifying Seascape Patterns
1. 2.1 Introduction
2. 2.2 Defining Seascape Applications
3. 2.3 Identifying Scales for Seascape Mapping
4. 2.4 Sensor Selection for Seascape Mapping
5. 2.5 Representing Patterns in Seascape Maps
6. 2.6 Quantifying Seascape Structure
7. 2.7 Applications of Seascape Maps and Spatial Pattern Metrics
8. 2.8 Conclusions and Future Research Priorities
9. References
Chapter 3: Pelagic Seascapes
1. 3.1 Introduction
2. 3.2 Pattern and Process in the Pelagic Realm
3. 3.3 Spatial Pattern Metrics for Pelagic Seascapes
4. 3.4 Spatial Ecoinformatics in the Pelagic Realm: from Physics to Predators
5. 3.5 Conclusions and Future Research Priorities
6. 3.6 Glossary
7. References
Chapter 4: Scale and Scaling in Seascape Ecology
1. 4.1 Introduction
2. 4.2 Expressions of Scale
3. 4.3 Spatial and Temporal Scaling in Estimating Uncertainty
4. 4.4 Spatial and Temporal Scaling in the Pelagic and Benthic Realms
5. 4.5 Looking to the Future: Scaling Concepts and Practice in Seascape Ecology
6. 4.6 From Ceteris Paribus to Dimensional Thinking
7. 4.7 Acknowledgements
8. References

Part II: Linking Seascape Patterns and Ecological Processes

Chapter 5: Ecological Consequences of Seagrass and Salt-Marsh Seascape Patterning on Marine Fauna
1. 5.1 Introduction
2. 5.2 Structural Processes and Change in Coastal Seascapes
3. 5.3 Ecological Consequences of Seascape Structure
4. 5.4 Challenges and Opportunities in Seascape Ecology
5. References
Chapter 6: Seascape Patch Dynamics
1. 6.1 Introduction
2. 6.2 From Patch Dynamics to Seascape Ecology
3. 6.3 Scale
4. 6.4 Factors Influencing Seascape Patchiness
5. 6.5 Mapping and Quantifying Seascape Change
6. 6.6 The Future of Seascape Dynamics Research
7. References
Chapter 7: Animal Movements through the Seascape: Integrating Movement Ecology with Seascape Ecology
1. 7.1 Introduction
2. 7.2 Using Animal Movements to Scale Ecological Studies
3. 7.3 Advances in the Visualisation and Quantification of Space-use Patterns
4. 7.4 Linking Animal Movement Patterns to Seascape Patterns
5. 7.5 Implications of Animal-Seascape Understanding for Marine Stewardship
6. References
Chapter 8: Using Individual-based Models to Explore Seascape Ecology
1. 8.1 Introduction
2. 8.2 Why use IBMs to Study Seascape Ecology?
3. 8.3 Data for Parameterizing Seascape Ecology IBMs
4. 8.4 Challenges and Future Directions in Using IBMs to Explore Seascapes
5. References

Part III: Seascape Connectivity

Chapter 9: Connectivity in Coastal Seascapes
1. 9.1 Introduction
2. 9.2 Global Synthesis of Connectivity Research
3. 9.3 Quantifying Connectivity: Advances in Key Tools and Techniques
4. 9.4 Application of Seascape Connectivity to Coastal Seascapes: Focal Topics
5. 9.5 Integrating Connectivity into Marine Spatial Planning
6. 9.6 Conclusions and Future Research Priorities
7. References
Chapter 10: Networks for Quantifying and Analysing Seascape Connectivity
1. 10.1 Introduction
2. 10.2 Network Models of Connectivity: Representing Pattern and Process
3. 10.3 Modelling Marine Population Connectivity
4. 10.4 Network Analysis of Marine Population Connectivity
5. 10.5 Case Study in Marine Connectivity: Hawaiian Islands
6. 10.6 Conclusions and Future Research Priorities
7. 10.7 Acknowledgements
8. References
Chapter 11: Linking Landscape and Seascape Conditions: Science, Tools and Management
1. 11.1 Introduction
2. 11.2 Landscape Ecology as a Guiding Framework for Integrated Land-Sea Management
3. 11.3 Modelling and Evaluating the Connections between Land and Sea
4. 11.4 Case Studies
5. 11.5 Towards Applying Landscape Ecology to Land-Sea Modelling and Management
6. References

Part IV: People and Seascapes

Chapter 12: Advancing a Holistic Systems Approach in Applied Seascape Ecology
1. 12.1 Introduction
2. 12.2 People as Part of the Seascape
3. 12.3 How Holistic Systems Science can Help Seascape Ecology
4. 12.4 Connecting Seascape Patterns to Human Health, Livelihoods and Wellbeing
5. 12.5 Conclusions and Future Research Priorities
6. References
Chapter 13: Human Ecology at Sea: Modelling and Mapping Human-Seascape Interactions
1. 13.1 Introduction
2. 13.2 Seascape Ecology, Spatial Patterns and Scale
3. 13.3 Human Use Data Types and Geographical Information Systems
4. 13.4 Modelling Human-Seascape Interactions with a Systems Approach
5. 13.5 Conclusions and Future Research Priorities
6. References
Chapter 14: Applying Landscape Ecology for the Design and Evaluation of Marine Protected Area Networks
1. 14.1 Introduction
2. 14.2 Applying Landscape Ecology Principles in the Marine Environment
3. 14.3 Case Study: Applying Landscape Ecology to Evaluate a Network of MPAs in California
4. 14.4 Synthesis
5. 14.5 Conclusions and Future Research Priorities
6. References
Chapter 15: Seascape Economics: Valuing Ecosystem Services across the Seascape
1. 15.1 Introduction
2. 15.2 Habitat Connectivity and Seascape Goods and Services
3. 15.3 Valuing Seascape Goods and Services
4. 15.4 Example of a Mangrove-Coral Reef Seascape
5. 15.5 Conclusions and Future Research Priorities
6. References

Part V: Epilogue

Chapter 16: Landscape Ecologists' Perspectives on Seascape Ecology
1. 16.1 Introduction
2. 16.2 From Landscapes to Seascapes (and Back Again)
3. 16.3 Seascape Ecology and Landscape Ecology: Distinct, Related and Synergistic
4. 16.4 Seascape Ecology
5. References
Index

End User License Agreement

List of Illustrations

Chapter 1: Introducing Seascape Ecology

Figure 1.1 Seascape showing spatial structure in the sea: A. Runoff plume; B. Temperature front; C. Eddies with entrained phytoplankton; D. Thermal front; E. Salinity gradients; F. Surface roughness; G. Plankton patches; H. Thin horizontal layer of plankton; I. Internal wave; J. Thermocline; K. Seafloor terrain morphology from bathymetry (three dimensional); L. Benthic habitat map representing patch-mosaic patterns (two dimensional); M. Geological features (canyons and seamounts); N. Within-patch structure (biological assemblages); O. Surficial sediment and geological strata.
Figure 1.2 (a) Number of publications using ‘seascape ecology’ over time. (b) Comparison of terms used over time.

Chapter 2: Mapping and Quantifying Seascape Patterns

Figure 2.1 Speed and direction of surface currents around the Main Hawaiian islands (MHI). These maps depict the average speed (m/s) and direction (° denoted by arrows) of surface currents in the summer (a, b) and winter (c, d) within the MHI.
Figure 2.2 Process for investigating seascape patterns. The process for choosing how to observe, characterize and apply seascape patterns to support management applications and / or answer specific ecological questions. This process is cyclical because the way in which seascape patterns and metrics are used will influence how they should be developed and vice versa.
Figure 2.3 Data scales collected by different sensors. Spatial and temporal scales of some commercially available sensors used for mapping seascape patterns.
Figure 2.4 Active and passive sensors for mapping seascape patterns. This diagram shows different types and platforms for active and passive sensors. Multispectral, hyperspectral, RaDAR, LiDAR and SoNAR are some of the more commonly used sensors in seascape ecology. Sensor platforms can include satellites, airplanes, ships, UAVs (Unmanned Aerial Vehicles), ASVs (Autonomous Surface Vehicles) and AUVs (Autonomous Underwater Vehicles).
Figure 2.5 Patch-mosaic versus continuous-gradient maps. Examples of how seascape patterns can be depicted as continuous gradients (2D and 3D images) (left) and patch mosaics (2D categorical maps) describing single habitats (middle) or an entire mosaic of habitats (right).
Figure 2.6 The effect of observational and analytical scales on 2D categorical maps. The four map panels show the same geographic location north of St Croix in the US Virgin Islands. The white lines denote distinct habitat patches on the seafloor. The map panels (top to bottom) show how choosing 2 or 3D continuous images with finer spatial and thematic resolutions can increase the number of habitat patches in the seascape. The map panels (left to right) show how choosing a finer resolution analytical scale (i.e., decreasing the MMU size and the delineation scale) can also increase the number of habitat patches in the seascape.
Figure 2.7 Methods for standardizing scales. Commonly used techniques to change and / or standardize the spatial scale of continuous 2D or 3D seascape maps. The neighborhood shape (e.g., square or circle), size (e.g., 3 × 3 m, 9 × 9 m) and statistic (e.g., mean, standard deviation) used can change the patterning seen in the seascape map.
Figure 2.8 Methods for identifying spatial distributions. Six steps involved in analysing multivariate data to determine seascape patterns. The coloured numbers correspond to the coloured list of steps in the middle. The steps go from the top left to the middle right to the bottom left.

Chapter 3: Pelagic Seascapes

Figure 3.1 Interacting temporal and spatial scales in marine ecosystems. (a) Time-space diagram showing the scale of several physical and biological processes from very short-lived, small-scale phenomena such as molecular processes and individual movements to global-scale, decadal climate signals. (b) Approximate time and horizontal space scales of sampling platforms used in the marine environment.
Figure 3.2 Global chlorophyll-a climatology, 1997–2000. Mean primary productivity over the global ocean, showing oligotrophic (nutrient-poor) open ocean waters in blue, and productive waters associated with equatorial and eastern boundary upwelling in green. Extremely productive coastal regions with high sediment concentrations such as the outflow from the Amazon delta are highlighted in warm colours. Surface chlorophyll-a concentration from ocean colour satellites is a useful metric for identifying and classifying pelagic seascapes over broad scales. Image from NASA Ocean Biology (OB), Sea-viewing Wide Field-of-view Sensor (SeaWiFS) Ocean Color Data, 2014.0 Reprocessing. Source: Obtained from NASA OB.DAAC, Greenbelt, MD, http://oceancolor.gsfc.nasa.gov, (accessed 29 May 2017).
Figure 3.3 Fronts as ecotones in pelagic seascape ecology. Example of thermal front mapping using satellite SST imagery, to identify potentially important ecotones in pelagic seascapes. Image shows North Sea waters to the north-east of the Scottish mainland. Transitions between water masses of different thermal characteristics are identified and the relative strength of the temperature gradient is shown by line thickness in (b). Red and blue lines trace the warm and cold sides of each thermal front.
Figure 3.4 Vertical structuring of pelagic seascapes. A schematic example of vertical structuring with depth in pelagic seascapes, from a 3-D biophysical description of Monterey Bay, California. Interactions between thermal fronts at the surface, phytoplankton layers at depth and internal waves – gravity waves that oscillate at depth rather than at the water surface – generate spatial structure in pelagic seascapes. These biophysical coupling processes can be observed and measured using in situ or remote sensing technologies, such as shipboard surveys and satellite oceanography.
Figure 3.5 Broad-scale spatial structuring of pelagic seascapes. Daily Sea Surface Temperature (SST) image, California Current Large Marine Ecosystem, 1 June 2010. Merged Ultra-high Resolution (MUR) daily SST, generated by the Global High Resolution SST (GHRSST) project and distributed by NOAA Coastwatch http://coastwatch.pfeg.noaa.gov/erddap/griddap/jplMURSST (accessed 29 May 2017).

Chapter 4: Scale and Scaling in Seascape Ecology

Figure 4.1 Frequency (n = articles / year) of use of ‘scale’ and use of two restricted uses of ‘scale’ in two oceanographic and three ecological journals. Frequency adjusted as n per 100 articles published in a year. Note threefold expansion of vertical axis for one journal (Ecology) relative to other journals.
Figure 4.2 Modification of the Stommel diagram by Haury et al. (1978), as extended by Godø et al. (2014) to illustrate coverage potential, overlap and uniqueness in time and space of acoustic data from stationary and oceanographic platforms. Vessels include vessel and vessel operated tethered platforms. Stommel diagram shows increasing variance in sea surface height from small scale (lower left) to large scale (upper right). Note large variance at days to week due to tides and ice age variations. Gravity waves, tsunamis, geostrophic turbulence and meteorological effects (in Stommel 1963) replaced with processes relevant to plankton – diel vertical migration, annual cycle and climate change. Phenomena relative to plankton dynamics are shown as circled areas labelled by letters A – K.
Figure 4.3 Spectral density of pelagic (a) and benthic (b) organisms compared to spectral density of water column fluorescence and seafloor substrate on a roughness scale of 1 to 9 (sand through gravel and cobble). Spectral density is the variance per unit of frequency measurement. Periods corresponding to measurement frequency are shown on the upper axis. Fluorescence is a measure of the concentration of passively drifting phytoplankton.
Figure 4.4 Space-time diagrams for the problem of monitoring the effect of chronic release of contaminants in Manukau harbour, New Zealand. (a) Spatial and temporal scale of release and subsequent tidal mixing to the scale of the lagoon; (b) Support and extent of embedded experiment (Legendre et al. 1997) compared to computational model of bedload transport at the scale of a single flat and the scale of the lagoon; (c) Dimensionless ratios that separate space and time scales where demographic rates (recruitment and death) prevail over kinematic rates (locomotion and passive drift). See text for definition of R_juv R_a and R_D/K.
Figure 4.5 Comparison of scaling manoeuvres and applicability. Accumulation, coarse graining and lagging occur by iterative calculation on data, rating occurs by noniterative measurement of units of different size.
Figure 4.6 Cumulative number of species with respect to time, effort, and number of samples. (a) Polychaete species on an arithmetic scale, from 1932 to 2006 from the Bodega Marine Reserve in a fixed area, Shell Beach to Doran Beach. (b) Polychaete species number with respect to effort (number of reports) on logarithmic scale. (c) Diatom species number from Patrick (1968) on a logarithmic scale. Polychaete data from Schneider et al (2007). Figure 6c redrawn from Schneider (2001b).

Chapter 5: Ecological Consequences of Seagrass and Salt-Marsh Seascape Patterning on Marine Fauna

Figure 5.1 Global distribution of (a) seagrasses and (b) saltmarshes; and (c) a close up of Chesapeake Bay to highlight the juxtaposition of seagrasses and saltmarshes in close proximity to the densely populated east coast of the United States.
Figure 5.2 Seagrass seascape dominated by eelgrass (Zostera marina) in SW Finland, northern Baltic Sea.
Figure 5.3 Salt marsh seascape in Bahia de Cadiz Natural Park, Huelva, Andalucía, Spain.
Figure 5.4 Natural horizontal and vertical processes and feedback mechanisms in a salt marsh. Effects of sea-level rise may increase horizontal erosion in shallow intertidal environments or increase vertical accretion in submerged salt marsh seascapes by increased sediment delivery and deposition (a). Examples of local (b), broad-scale (c) and socio-economic (d) drivers that influence above and below-ground accretion and erosion processes are shown in green, blue and yellow boxes, respectively.
Figure 5.5 Vertical view of a seagrass (a) and a saltmarsh (b) seascape illustrating commonly investigated metrics and comparisons influencing distribution and abundance of marine fauna. (i) 1 versus 2: patch size effects, 3 versus 4: patch shape effects (differing edge : area ratios), 5 versus 6 and 7: interior versus edge effects (dots at 6 illustrate a transect sampling across the edge), 8 versus 9: patch isolation; and 10 versus 11: grid to measure percentage cover. (ii) 1: transect for sampling across channel edge, 2: edge : interior comparison in marsh patch, 3 versus 4: patch size effect, 5 versus 6: patch density (percentage cover), 7 versus 8, 9 and 10: effect of larger, higher order channel / corridor versus lower smaller order channels; and 11 versus 12: effects of amount of edge per unit area.
Figure 5.6 Hypothetical Scaling of habitat. Species mobility and behaviour influence organism responses to seascape connectivity and configuration. Habitat perception is species and process specific and a fragmented habitat may appear fragmented for one species but continuous for another species. A = a sessile invertebrate, e.g. a bivalve, perceives a small patch as a continuum, B = semimobile invertebrate, e.g. a shrimp, with a small perceptual range utilizing the edge and the interior of a small patch, C = a fish species with a limited home range can easily cross bare sand and perceives its surroundings as almost continuous vegetation, D = a fish with medium sized home range may effectively utilize habitat edges and perceive its foraging area as either fragmented or continuous, E = a large, highly mobile species may perceive the entire bay as an unvegetated area. Dark green elongated patches with high P : A ratios are suitable only for edge specialists (black triangles), while the rest of the patches (light green) have lower P : A ratios able to support species dependent on core areas (black squares).
Figure 5.7 Conceptual model showing routes and mechanisms of trophic transfers between marine and terrestrial habitats in a temperate ecosystem. References: 1. Kantrud (1991), 2. Ganter (2000), 3. Cottam et al. (1944), 4. Lenanton et al. (1982), 5. Lenenton & Caputi (1989), 6. Irlandi & Crawford (1997), 7. Hanson (1969), 8. Darcey (1985), 9. Mateo et al. (2003), 10. Mateo et al. (2006), 11. Gunter (1967).
Figure 5.8 Shifting interaction zones where marsh nekton may be involved in transfers of intertidal production via predator-prey relationships during (i) high and flooding tide stages and (ii) ebbing and low tide stages. Red lines are the hypothetical ‘hot spots’ of production transfers across the seascape at each tidal stage.

Chapter 6: Seascape Patch Dynamics

Figure 6.1 The perception and responses to a patch are organism specific, one animal's patch is another animal's seascape (adapted from Pittman 2013). Symbols courtesy of the Integration and Application Network, University of Maryland Center for Environmental Science (ian.umces.edu/symbols/, accessed 28 May 2017).
Figure 6.2 Our ability to assess seascape change increases as we consider processes occurring at different scales.
Figure 6.3 Fusion and fragmentation processes within a seascape. A depiction of how different processes of disturbance, colonization and clonal growth in a seagrass meadow can result in similar patch formations within a seascape and the four spatial processes involved in fragmentation: shrinkage, attrition, perforation and subdivision. Source: Adapted from Boström et al. (2011).
Figure 6.4 Examples of types of disturbance magnitudes, causes and examples of different resulting patch configurations using seagrass as an example. Source: Adapted from Lake (2000).
Figure 6.5 Example of a conceptual diagram showing the main steps in a study of seascape dynamics.
Figure 6.6 Conceptual diagram of the interaction of seascape habitat loss and fragmentation and general pathways of seascape pattern outcomes.
Figure 6.7 Illustration of spatiotemporal analysis of seascape change using GIS. Time step differences between areas where submerged habitat gain or loss cover can be used to estimate an index of relative change, and identify areas of concern.
Figure 6.8 A transitional model approach to understand seascape dynamics. (a) A collection of seascape maps are stacked where the transition of different patch size classes (X_n) are follow to estimate change probabilities using (b) transitional matrix models. Then transitional models could be applied to (c) disturbance scenarios to project seascape changes.

Chapter 7: Animal Movements through the Seascape: Integrating Movement Ecology with Seascape Ecology

Figure 7.1 Conceptual model of movement ecology incorporating focal patterning from landscape ecology. Seascape structure and conditions can act as drivers for the internal state of individual animals (physiological and psychological states), can influence their navigation (mental maps, orientation), motion capacity (locomotory morphology/traits) and ultimately the movement process, which results in a specific pathway in time and space. Source: Adapted from Nathan et al. (2008).
Figure 7.2 Relevance of a hierarchical framework to the definition of an environment for highly mobile animals.
Figure 7.3 An operational framework that applies a landscape ecology perspective to the study of connectivity in tropical marine seascapes. The framework consists of three components: developing a conceptual framework that guides the study, determining appropriate scales for analysis and conducting geospatial analyses scaled to the organism of interest. Solid arrows indicate directional flow among the subcomponents. Broken arrows represent directional flow and important feedback loops among and within the three components. Source: Developed through personal communication with C. Jeffrey, NOAA Biogeography Branch.
Figure 7.4 Kernel density estimation with different colours representing iso-surfaces of different kernel densities. The black line represents the home range as defined by minimum convex polygon (MCP) techniques. Red contour lines are 95% kernel home range and the blue contour line is 50% core range. Black dots are GPS locations of an animal. Source: http://gis4geomorphology.com/home-range-kernel/ (accessed 25 May 2017).
Figure 7.5 Stepwise spatial analysis to map and quantify benthic seascape patterning at a spatial extent defined using the diel activity space calculated from tracking an individual fish following the methods of Hitt et al. (2011a).
Figure 7.6 Differences in precision of utilisation distribution using four different home-range estimation techniques. Source: https://www.werc.usgs.gov/ProjectSubWebPage.aspx?SubWebPageID=1&ProjectID=258 (accessed 25 May 2017).
Figure 7.7 (a) Daily tracks of the movement of one lesser black-backed gull (Larus fuscus) collected over one month in the Netherlands; (b) Space-time cube of the same tracks with the z-axis showing time of day in seconds and ranging from midnight at the bottom to midnight at the top; (c) space-time kernel density estimate of the seagull's movements; (d) iso-surface with high intensity value to show two core areas of use separated by space and time. These areas are spatiotemporal hotspots and indicate previously unknown particularities in this gull's movement: consistently spending nights at a mainland location outside the nest and days in and around the nest. Source: Adapted from Demšar & van Loon (2013).
Figure 7.8 Three-dimensional movement-based kernel density estimates (3D MKDE) for an individual satellite tracked dugong overlain on 10 m resolution bathymetry. The probabilities representing 3D space use were mapped to 10 (x) × 10 (y) × 0.5 (z - depth) metre cubes (voxels). (a). The 99% contour volumes for 3D MKDEs based on locations when tidal heights ranged from 0.5–1.0 (red), 1.0–1.5 (orange), 1.5–2.0 (yellow), 2.0–2.5 (light green) and 2.5–3.0 (green) metres are shown. Based on the 3D MKDEs for each tidal height category, the probability that the dugong could have been at different water depths was computed into 0.5 m bins (b). The value on the y-axis is the upper depth value for each 0.5 m bin (i.e., 0 indicates 0.0–0.5 m depth). Source: Reproduced from PloS One (Tracey et al. 2014a).
Figure 7.9 Summary and guidance for analytical tool selection for specific types of research questions when quantifying space use patterns and linking animal movement to seascape structure.
Figure 7.10 Overlay of marine species corridors for 10 marine species (representing fishes, sea turtles and marine mammals) and protected and management areas in the Gulf of Mexico. Source: Provided with permission by Jorge Brenner of The Nature Conservancy (Brenner et al. 2016).

Chapter 8: Using Individual-based Models to Explore Seascape Ecology

Figure 8.1 The basic components of a seascape ecology IBM, based loosely on the blue crab – seagrass model of Hovel & Regan (2008) created using NetLogo software. The seascape is composed of cells that represent seagrass habitat (gray: patch interior; blue: patch edge) and nonseagrass habitat (black). The seascape is populated by prey (blue ‘bugs’), mesopredators (red crabs) and top predators (fish). Each cell is endowed with attributes that, in combination with attributes of individual organisms, influence decisions made by organisms in each model time step. It is possible to include many more attributes.
Figure 8.2 (a) An individual pink abalone (Haliotis corrugata) equipped with an acoustic transmitter that allows the position of the animal to be determined over time steps of minutes (photo credit: Julia Coates). (b) Home ranges of four example abalone tracked in Coates et al. (2013). Dots represent discrete positions in the seascape and gray lines represent statistically determined home ranges. Large dots arranged in a triangle represent the position of stationary acoustic transmitters used to triangulate the position of each animal at each time point.
Figure 8.3 Model procedures for the pink abalone (Haliotis corrugata) IBM testing how abalone movement influences spawning success (Coates and Hovel 2014). Shelters (‘scars’) and abalone are placed in the seascape and abalone make movement decisions at each time step based on an assessment of their surroundings and readiness to spawn. NN = nearest neighbor.
Figure 8.4 Left: aerial photographs of sections of an eelgrass (Zostera marina) seascape in the lower Chesapeake Bay, Virginia, United States, illustrating different patchiness regimes. From Hovel & Lipcius (2001). Right: simulated eelgrass seascapes created in NetLogo. Colored areas represent seagrass (gray: patch interior; blue: patch edge) and black represents unvegetated sediment.
Figure 8.5 The four 100 × 100 m simulated marshscapes used in an IBM to test how marsh seascape configuration influences the survival and growth of the brown shrimp Farfantepenaeus aztecus. Gray is water and black represents vegetation. The four maps represent snapshots in a simplified continuum of marsh disintegration and different combinations of habitat configuration and cover: LE-HV = little edge and high amount of vegetation; HE-HV = high edge and high amount of vegetation; HE-LV = high edge and low amount of vegetation; LE-LV = little edge and low amount of vegetation.
Figure 8.6 Top panel: Map of the Florida Keys, United States and (A) the simulated seascape composed of seagrass and hard bottom habitat used by Caribbean spiny lobsters (Panulirus argus). (B) and (C) represent two simulated cyanobacterial blooms that caused the die off of habitat-forming sponges. Bottom panel: results from the Caribbean spiny lobster IBM comparing lobster abundance between seascapes affected by, or not affected by sponge die off. The percentage difference in postbloom lobster abundance between seascapes affected by blooms or not affected by blooms is shown. The two black bars on the x-axes represent the periods when the algal blooms occurred. Upper panel represents lobster abundance in the entire Florida Keys region and lower panel represents lobster abundance only in the areas of cyanobacterial blooms.

Chapter 9: Connectivity in Coastal Seascapes

Figure 9.1 Spatial linkages that illustrate diversity in the form and function of connectivity effects. Fish move between tropical coastal ecosystems with ontogenetic development (a). Humpback whales migrate between polar feeding grounds and tropical breeding areas (b). Raptors forage on fish stranded on sandy beaches and relay carbon from sea to land (c). Detached kelp washes ashore on exposed beaches where it provides food for invertebrates (d). Mangrove propagules disperse out to sea in coastal currents (e). Flood plumes from rivers transport sediments and nutrients in coastal waters (f).
Figure 9.2 Summary of global research on connectivity in marine ecosystems showing variation in the: types of connectivity assessed (a); regions of study (b); focal seascapes (c); and organisms examined (d). The number of studies conducted is displayed for each: type, region, seascape and organism and bars show this Figure as a proportion of all connectivity studies. Regions: AU, Australia; NA, North America; MS, Mediterranean Sea; CS, Caribbean Sea; SA, South America; OC, Oceania; AF, Africa; AS, Asia and Indonesia; AN, Antarctica; and EU, Europe. Organisms: FI, fish; CO, coral; CR, crustaceans; BI, bivalves; MA, mammals; MO, molluscs; RE, reptiles; EC, echinoderms; AL, algae; SG, seagrass; MH, marsh; and MG, mangroves.
Figure 9.3 Connectivity effects on fish and seagrass ecosystems in the Bahamas. The abundance (b) and secondary production (c) of white grunts (Haemulon plumierii) on artificial reefs is positively correlated with seagrass cover. Fish density declines rapidly with distance from reefs (d), as does the height of seagrass (e), and the phosphorous content of seagrass leaves (f).
Figure 9.4 Combined effects of connectivity and seascape protection on fish, ecosystem functioning and food webs in eastern Australia. Herbivorous rabbitfish (Siganus fuscescens) are most abundant on coral reefs that are both near mangroves and protected in marine reserves (a); these fish feed on algae, which is less abundant on protected reefs near mangroves (b), which promotes coral recruitment (c). Rabbitfish feed on algae from both reefs (d) and mangroves and the contribution of mangrove carbon to their diet declines with reef isolation (e). Adapted from Olds et al. (2012) and Davis et al. (2014b).
Figure 9.5 Great knots (Calidris tenuirostris) (a) migrate annually between feeding areas in Australia and breeding areas in eastern Russia (b). They depend on the tidal flats of the Yellow Sea as the sole resting site on this migration, but 65% of these flats have been lost to land reclamation since 1950 (c). Loss of this critical resting habitat has led to declines in great knot abundance in Moreton Bay (d), and across Australia (Bamford et al. 2008; Wilson et al. 2012; Murray et al. 2014).

Chapter 10: Networks for Quantifying and Analysing Seascape Connectivity

Figure 10.1 An example marine habitat network where functional linkages (edges or arcs) depict directional dispersal among discrete habitat patches (nodes). Direction in the linkages is implied by following arcs in a clockwise direction. Here, habitat area is illustrated by the relative size of the nodes and the colour represents whether the habitat is protected (blue) or unprotected (green).
Figure 10.2 The Hawaiian island seascape where the coral population connectivity was modelled and subsequent network analysis used to identify potential conservation priorities. Coral connectivity was modelled for all reef habitat along the Hawaiian archipelago and Johnston Atoll (see text for biophysical parameters). Reef patches are shown as red nodes (sizes scaled to availability of suitable habitat). The dispersal-based coral population network is drawn with blue links where the directionality is implied by following the arcs in a clockwise direction. True north is aligned vertically and data presented in the Mercator projection (165°W central meridian).
Figure 10.3 Results of the network analysis completed for the system of reefs and coral dispersal dynamics along the Hawaiian islands. The strong demographically significant linkages are shown as a network of blue arcs where the directionality is implied by following them in a clockwise direction. Nodes in all networks represent unique reef habitat patches and are scaled and / or coloured to represent results. (a) The relative rate of local retention. (b) The sourceness of each patch, quantified by calculating the total flow of larvae from the focal patch to all downstream nearest neighbours. (c) Betweenness centrality reflects each site's contribution to the overall cohesion of the coral network and identifies important stepping-stone sites. (d) A network modularity algorithm was used to partition the network into densely clustered communities. Each unique community is shown in a unique colour (node size is shown relative to reef area).

Chapter 11: Linking Landscape and Seascape Conditions: Science, Tools and Management

Figure 11.1 Schematic diagram of pathways through which land use affects nearshore coral reef environment.
Figure 11.2 The Directional Leakiness Index (DLI) is a metric of the spatial arrangement and size of bare, ‘interpatch’ areas. Studies have found DLI is a good indicator of runoff and sediment. Here, conceptual landscapes are arranged from highest to lowest leakiness.
Figure 11.3 Erosion, sediment exposure of coral reefs and recreation in West Maui. (a) Sediment export (tons/year) estimates by subwatershed calculated using InVEST SDR 3.2, coupled with a sediment dispersion model to estimate spatial distribution of sediment dispersed (ton/km²/year) in the coastal zone. (b) Coral cover based on NOAA benthic data and recreation intensity based on InVEST Recreation model.
Figure 11.4 Models of runoff for the Caribbean island of St John (US Virgin Islands). (a) Summit-to-sea hydrological model of runoff showing pour points and predicted exposure of coral reefs to sediment. (b) Landscape development intensity index (LDI) based on land use classified from aerial imagery. (c) Satellite-derived water colour from the MERIS sensor (2003–2011).
Figure 11.5 Local sediment effects ranging from low to high values across the terrestrial realm of the Northeast Corridor of Puerto Rico, as well as sedimentation plumes in the marine realm. The Río Fajardo is show in blue, as well as locations of coral reefs in the marine realm. Decreases in local sediment effects (right-hand map) are observed after land cover changes were made in NSPECT to the polygon outlined in red for our mitigation scenario. The resulting marine plume's accumulated sediment values also decreased.
Figure 11.6 Vista Site Explorer. Two marine elements are noted to occur in this selected site (highlighted in red): manatee and medium density corals. All areas of medium density coral and manatee occurring on this site are incompatible and display a negative response to the scenario components listed below, which include a marina, as well as very high nitrogen and sediment plumes.
Figure 11.7 (a) Postclearance (2009) map of Edgecumbe Bay remaining coastal ecosystems showing where vegetation has been modified or lost. (b) Preclearance (pre-European) coastal ecosystem map of the Edgecumbe Bay sub-basins.
Figure 11.8 (a) Land use 1999 in the sub-basins that feed into Edgecumbe Bay. (b) Land use 2009 in the sub-basins that feed into Edgecumbe Bay.
Figure 11.9 Hydrological connections mapping, ‘Bluemaps’, for the subcatchments.

Chapter 12: Advancing a Holistic Systems Approach in Applied Seascape Ecology

Figure 12.1 Multiscale conceptual framework in which seascape structure is conceptualised within an integrated system with a focus on the nexus of ecosystem services and human wellbeing. Key components, interactions, drivers of change and relevant research topics are listed.
Figure 12.2 The seascape wheel. Example of a holistic definition of seascape integrating social, cultural, psychological and environmental attributes while retaining a distinct societal-centred approach.
Figure 12.3 An example of a causal network of a coral reef-human ecosystem in Mexico, which incorporates connections between ecosystem structure, fishing, conservation, economics and governance. Grey circles represent major socioeconomic drivers. Broken lines represent decisions made by a group of stakeholders. Positive (+) and negative (−) feedback relationships are marked alongside arrows.
Figure 12.4 A heuristic model linking nearshore benthic seascape patterning (mangroves, seagrasses, coral reefs) to fishing communities showing how seascape structure influences functional connectivity, which in turn influences individual biology and population structure with multiple consequences for fishing communities. The Figure provides some examples of relevant patterns and processes and the specialist disciplines that would each offer a different perspective and focal scale(s) through which to study the phenomena.

Chapter 13: Human Ecology at Sea: Modelling and Mapping Human-Seascape Interactions

Figure 13.1 Map from Halpern et al. (2008) illustrating cumulative human impact across 20 ocean ecosystem types.
Figure 13.2 Automatic Identification System (AIS) data showing fishing intensity (purse seine and long line) patterns before (a) and after (b) establishing a marine reserve Phoenix Island Protected Area (PIPA) in Kiribati.
Figure 13.3 InVEST model results mapping the predicted effect of three different mangrove removal scenarios on coastal inundation during storm events. Mangroves are located in all flood potential areas, coloured in dark blue, under the status quo scenario of no mangrove removal.
Figure 13.4 Discrete choice model fit results showing some of the factors that influenced various modelled fisher decisions off the west coast of Florida. The top two panels show the partial probabilities that different wind speeds and catches relative to fish hold capacity affect the decisions to take a fishing trip and return to port respectively. Red points represent the observed values, while black lines are model fits. Bottom panels spatially depict the fishing locations for which parameters for fuel price and wind speed were significant using red circles.
Figure 13.5 Steps to conduct a sound simulation study.
Figure 13.6 Statistical comparison of population level patterns emerging from an agent-based model, with population level patterns from field collected empirical data. Simulated linear fish speeds (Wilcoxon rank-sum test: p = 0.2933) and tagging distance (Wilcoxon rank-sum test: p = 0.780) emergent from the simulation model for red grouper are compared to the same metrics calculated from a field tagging experiment. Alpha and beta are the parameters of the fitted gamma distributions used to approximate the simulated and empirical probability density functions depicted in the graphs.
Figure 13.7 Visual comparison of simulated emergent spatial catch patterns of red grouper predicted by an agent-based model, with actual spatial catch observations of red grouper provided by vessel monitoring system (VMS) data.

Chapter 14: Applying Landscape Ecology for the Design and Evaluation of Marine Protected Area Networks

Figure 14.1 The image on the left shows the California Central Coast Marine Life Protection Act (MLPA) region from Pigeon Point to Point Conception with the marine protected areas (MPAs) outlined in blue (SMCA: State Marine Conservation Area) and red (SMR: State Marine Reserve) and the California State waters boundary outlined in black. Boxes a–e provide zoomed in views of the MLPA habitat along the central coast with the different colours representing each habitat specified in the MLPA based on substrate type and depth overlaid on shaded relief imagery of the seafloor.
Figure 14.2 Examples of MPAs with varying habitat diversity within the Central Coast Marine Life Protection Act (MLPA) region and their locations shown on the map in the centre: (a) Portuguese Ledge State Marine Conservation Area (SMCA), (b) Point Lobos State Marine Reserve (SMR), (c) White Rock SMCA and (d) Vandenberg SMR. The shaded relief imagery of the seafloor is coloured by depth (note: depth range varies between images). The graph provides the distribution of diversity indices in the marine protected areas (MPAs, blue bars) and in the potential MPAs across the region (red line).
Figure 14.3 Example of how the seafloor data can be used to determine the amount of rocky reef habitat the boundaries of marine protected areas (MPAs) intersect. Rocky reef is brown, sediment is light tan and the MPA boundary is outlined in white. The Carmel Pinnacles MPA has a large area of rocky reef habitat within it but the MPA boundary intersects a large portion of the rocky reef (a) while the Piedras Blancas MPA has a large area of rocky reef habitat but the MPA boundary intersects very little rocky reef habitat (b).

Chapter 15: Seascape Economics: Valuing Ecosystem Services across the Seascape

Figure 15.1 Diagram showing the ecosystem connectivity between mangroves, seagrasses and coral reefs. Ecological and physical connectivity between ecosystems is depicted for each ecosystem: terrestrial (brown arrows); mangroves (green arrows); seagrasses (blue arrows); and coral reefs (red arrows). Potential feedbacks across ecosystems from the impacts of human activities on ecosystem services are shown with yellow arrows.
Figure 15.2 Simulation of seascape connectivity, water pollution / sediment control and habitat-fishery linkage in a mangrove-coral reef system.

List of Tables

Chapter 2: Mapping and Quantifying Seascape Patterns

Table 2.1 Universal class-level metrics
Table 2.2 Universal landscape-level metrics

Chapter 4: Scale and Scaling in Seascape Ecology

Table 4.1 Population dynamics of plankton $c04-math-0015$ and nekton $c04-math-0016$ as a function of demographic rates (r = recruitment $c04-math-0017$ ) and as a function of kinematic rates due to locomotory behaviour (F_loc = ∇_h · u_Nekton) and to passive motion with the ocean circulation (F_fluid = ∇_h · u_fluid). An additional kinematic term (F_atmos = ∇_h · u_atmos ) is required for marine birds and fishing vessels, whose motions interact with atmospheric dynamics (Schneider et al. 1992). ∇_h is the horizontal gradient operator, u = organism velocity in directions x and y, with respect to fixed Cartesian grid on the surface of the earth. u = (δx/δt δy/δt)
Table 4.2 Prevalence of terms in four representative journals: Journal of Physical Oceanography (JPO), Marine Ecology – Progress Series (MEPS), Ecology (ECOL) and Landscape Ecology (LECOL)

Chapter 6: Seascape Patch Dynamics

Table 6.1 Different types of seascape models

Chapter 9: Connectivity in Coastal Seascapes

Table 9.1 Common terms relating to connectivity that are adopted in studies of movement biology and seascape ecology. Definitions, examples and citations to relevant published studies are provided for each term

Chapter 10: Networks for Quantifying and Analysing Seascape Connectivity

Table 10.1 Common network terms often applied to ecological and population networks. Also see Galpern et al. 2011; Urban et al. 2009; Minor & Urban (2008) for additional descriptions
Table 10.2 Descriptions of the tools used for the development of habitat networks and for network analysis. References are provided for additional information

Chapter 11: Linking Landscape and Seascape Conditions: Science, Tools and Management

Table 11.1 Studies using modelling approaches to examine patterns of runoff in the Hawaiian islands
Table 11.2 Changes to ecological processes in very frequently connected (VFC), frequently connected (FC), intermittently connected (INT) and infrequently connected (INF) areas of the catchment
Table 11.3 Ecosystem services and actions that can improve the capacity of the Edgecumbe Bay catchment to deliver them. Note: Very frequently connected (VFC), frequently connected (FC), intermittently connected (INT) and infrequently connected (INF)

Chapter 14: Applying Landscape Ecology for the Design and Evaluation of Marine Protected Area Networks

Table 14.1 Landscape metrics used to conduct landscape ecology in terrestrial and marine studies. The first column provides the landscape metric broken into sub categories of landscape composition (2D), landscape composition (3D), spatial configuration (patch based) and spatial configuration (contagion). The second column provides a description of each metric followed by several publications that used each metric in terrestrial (third column) and marine (fourth column) studies
Table 14.2 Examples of studies using metrics of landscape ecology to evaluate the effectiveness of protected areas in both terrestrial and marine systems. The goal of each of the studies is stated along with the question used to evaluate that goal, the landscape metrics used and the source for the study or studies
Table 14.3 Percentage of the MLPA habitat classes within each of the Central Coast MPAs. Those percentages that meet the criteria to be considered a replicate as specified in the MLPA are in bold typeface. The final row in the Table contains the percentage of blocks from the ‘potential MPAs’ in the region that contain replicates of the each of the habitat classes
Table 14.4 The boundary analysis for the Central Coast MPAs. Each MPA is listed along with length of the MPA boundary intersecting reef given in kilometres (HI) and the total area of reef within the MPA (HA) in square kilometres. The final column is a measure of boundary permeability given as HI/HA, which has been multiplied by 1000 for easier comparison between MPAs

Chapter 15: Seascape Economics: Valuing Ecosystem Services across the Seascape

Table 15.1 Seascape connectivity of interconnected habitat types and the consequences for ecosystem services
Table 15.2 Summary of seascape connectivity implications for management of mangrove-coral reef system

Seascape Ecology

Edited by Simon J. Pittman

NOAA Biogeography Branch, Silver Spring, USA and Marine Institute, Plymouth University, Plymouth, UK

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions.

The right of Simon J. Pittman to be identified as the author of the editorial material in this work has been asserted in accordance with law.

John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK

For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com.

Wiley also publishes its books in a variety of electronic formats and by print-on-demand. Some content that appears in standard print versions of this book may not be available in other formats.