I.  Review of Current Monitoring Efforts in Coastal Ecosystems

(STAC Fellows Report)

Review of current monitoring efforts in coastal ecosystems

09 February 2009

Long Island Sound Study
Science and Technical Advisory Committee

Mark Hoover
Santiago Salinas

Section I
Introduction

The scientific community has come to the consensus that anthropogenic emissions of greenhouse gases are changing the earth’s climate.  Climate change will have many detrimental effects on the environment, especially in coastal regions (NERRS 2008).  To mitigate the effects of climate change many environmental organizations are beginning to draft monitoring and adaptation strategies.  It is important to get these programs started now in order to effectively address future climate change impacts.

Currently, there are many programs designed to monitor and protect the environment.  While some programs have had success, climate change will aggravate current environmental problems by adding another dimension to the causes and rates of variability in the ecosystem (GOOS 2003).   One of the biggest issues in managing climate change is the range of variability.  In order to understand how climate change is affecting our coastal resources, there needs to be better long-term, continuous environmental monitoring (CBP STAC 2007, FAO 2005).

In order to understand the effects of climate change on an ecosystem there needs to be long-term continuous data.  Many organizations are using data that has already been collected for other projects over long periods of time (GOM 2004). For accurate mitigation strategies, the data collection needs to be updated frequently, and at multiple temporal and spatial scales (Brush et al. 2007).   Also, it is important to realize that the data collected is not useful if it is not given to the proper authorities.  Therefore many organizations are beginning to promote sharing of data with other agencies to support numerous mitigation plans (GOOS 2008).  While data collection is needed to observe the impacts of climate change, continuous monitoring can also be used to assess the effectiveness of mitigation strategies (GOOS 1998).

Monitoring is crucial for developing mitigation strategies for climate change.  It is impossible to detect what change is occurring if there is no baseline to compare the change to.  In other words, monitoring can be used to provide improved understanding of the earth’s system variability, and thus detect change (FAO 2005). 

Monitoring cannot only be used to detect change but also used to predict what changes will occur.  There are many models that can predict the effects of climate change, but their efficacy can be improved with in situ data (GOOS 2003).  Monitoring will produce a better understanding of dynamic coastal processes, which will enhance the prediction of the effects of climate change and therefore help organizations prepare accordingly (Pyke 2008, GOOS 2008, Barton 2003, NOAA 2007)

Ideally one would like to monitor everything all the time; however, this is clearly impossible. Therefore it is becoming common practice to develop sentinel systems and indicator species.  The goal is that by studying some region or species intensely the effects observed can be extrapolated to entire ecosystems (GOOS 2008).

Funding often limits the scope of many monitoring programs.  To overcome this, many organizations are developing well understood, intensely monitored, sentinel sites which can be used to detect change at broader scales.  By focusing all resources in a localized region, it is possible to develop a better understanding of the ecosystem as a whole.  Sentinel sites can therefore be used as a warning system, because if change starts to occur in these regions it can often be assumed that it will occur in similar ecosystems (FAO 2005).

Another monitoring strategy is the use of indicator species.  Indicator species are organisms that are studied intensely, to predict larger changes in the ecosystem (GOM 2004).  Indicator species are normally organisms that have a large effect on their ecosystem, meaning that a change in them would provide useful information about the condition of the ecosystem as a whole (GOM 2004). Some studies are beginning to establish which environmental indicators are the most sensitive and insensitive to climate change so that the best indicators can be used (Pyke 2008).

Section II
Monitoring programs

There are several large-scale monitoring programs that have been developed to detect climate change in the oceans and coasts.  These programs have gone through much of the same planning that the Long Island Sound Study is currently going through, such as site selection, choosing indicator species, and end product designation.  As such, by looking at these other monitoring programs our organization can learn strategies to use right here in Long Island Sound.

Global Earth Observing System of Systems (GEOSS)

The Global Earth Observing System of Systems is a collection of monitoring programs developed by the United Nations to detect and predict large scale regional change (Christian 2007). There are two main programs in GEOSS: the Global Terrestrial Observing System (GTOS) and the Global Ocean Observing System (GOOS).  For the purpose of Long Island Sound we will be looking at GOOS as well as at a small section in GTOS, the Coastal Global Terrestrial Observing System (C-GTOS) (Christian 2005)

Coastal Global Terrestrial Observing System (C-GTOS)

C-GTOS is a system designed to make global observations that will provide a better understanding of earth system variability and change in the coastal zone (Christian 2007).  The primary goal of C-GTOS is to detect, assess, and predict global and large scale regional change associated with land-based wetland and freshwater ecosystems along the coast.  At several workshops experts derived the following variables of concern for C-GTOS.

C-GTOS is a bridge between the Global Terrestrial Observation System (GTOS) and the Global Ocean Observation System (GOOS).  C-GTOS can be broken down into two phases.

The first phase is to produce a set of readily achievable products that are designed to provide tests of concepts for the mature observation system.  The products were chosen based on the needs of the program and are representative of the topics of concern.

The first objective of the initial phase of C-GTOS is the enhancement of TEMS.  TEMS are current stations used in GTOS.  To enhance them, C-GTOS aims to improve their geographic coverage, add additional coastal monitoring variables and indicators, integrate satellite images with ground truth, and enhance the outreach of TEMS through the web.

The second objective is to estimate the distribution and rate of change of population, urbanization, and land use/land cover in the coastal zone.  It is believed that using current datasets these rates can be estimated within 1 person year. 

The third goal is to assess the vulnerability of ecosystems in deltaic systems.  Deltaic systems are being threatened through anthropomorphic actions as well as sea level rise.  C-GTOS aims to create a series of maps through models showing the vulnerability of deltaic systems.   
C-GTOS will focus on a small subset of 50 deltaic systems for which there is suitable data available.  This product should be completed in 1-2 years and updated with new information every 5-10 years.

The fourth short term goal is the management of conservation of cultural sites in the coastal zone.  These are areas that people have designated as worthy of preservation and protection.    These sites may already have data available and could serve as sentinel sites for global change.  C-GTOS plans on creating a web service that would allow users to access information regarding these cultural sites such as maps, encyclopedic information, and monitoring efforts.  These products should be able to be produced quickly, within 6 months, and at low cost.

The final short term goal is the distribution of sites appropriate for analysis of delivery systems.  For this goal C-GTOS will select a network of sites across a range of geographical, socio-economic, environmental conditions to analyze the delivery systems.  The aim is to provide timely support to C-GTOS.

The development of these five short term goals is the first phase of C-GTOS and is designed to provide tests of concept and garner support for the second phase which is the establishment of a mature monitoring system.   The mature system will provide coastal observation information to address the four indicators of concern.

Currently C-GTOS is still in its first phase of development.  It has been several years since the publication of their implementation strategy and as such they have begun the development of the five short term products.    For example they are currently using a geophysical analysis to produce delta extent maps for 12 deltas worldwide, have a pilot study on the Nile delta using satellite images from 3 decades to record delta loss, and they have begun a preliminary identification of cultural sites that may be suitable to provide the web products.  These are just a few of the projects currently ongoing (GOOS 2005).

Global Ocean Observing System (GOOS)
 
Another component of the Global Earth Observing System of Systems (GEOSS) is the Global Ocean Observing System (GOOS).  GOOS was formally established in 1991 by the Intergovernmental Oceanographic Commission (GOOS, 2008). GOOS is an international, ocean-based monitoring program, designed specifically to detect, assess, and predict change in the ocean.

The initial focus of GOOS was physical measurements as they are the easiest to monitor.  As GOOS develops, more biological and chemical components will be included, but members felt that at the time of development the science was not sufficient to decide what should be measured to meet the needs of coastal resources.  GOOS uses a combination of Remote Sensing sources and in situ measurements

The development of GOOS is done through a pilot project approach.  New projects and programs are developed by the GOOS Scientific Steering Committee.  Upon approval the pilot project runs for 3-5 years after which it becomes a fully functioning operational system.   The pilot projects demonstrate operational techniques, and user demand for products (GOOS 1998b). 

GOOS has become established internationally through several regional alliances.  The developed regional alliances include EuroGOOS, Mediterranean GOOS, Black Sea GOOS, North East Asian Regional (NEAR) GOOS, PL GOOS, Indian Ocean GOOS, IOCARIBE GOOS, Africa GOOS, and US GOS. Each of these alliances are run by partner countries and organizations.  It is through the cooperation of these alliances that GOOS functions as one international organization. For the purpose of simplicity I will now go into further detail regarding two of the previously described regional alliances; Black Sea GOOS, and NEAR GOOS.

Black SEA GOOS

The main objective of the regional Black Sea GOOS is to produce a reliable, international observing network that will collect marine data to observe and predict the variability of the Black Sea.  Existing observation and monitoring networks are the basis for the program.  These monitoring networks were designed to be systematic, routine, and cost effective.

The Black Sea GOOS is an ongoing project and it is currently still in the implementation phase.  They are working on defining an optimal observation network building off existing observation stations and installing some new sensors and equipment when needed.  Also they are trying to establish and automate standards for quality control.

The monitoring networks are operated through two modules: Marine Service Module (MSM) and Regional Management and Analysis Module (RMAM) (IOC 2003) In the MSM module observations will be designed to address specific problems.  The MSM will incorporate many existing programs as appropriate and also incorporate new measurements. 

The RMAM module can be broken down into two components: coastal management and analysis, and open sea management and analysis.  The coastal zone management is designed to monitor pollutants, while the open sea management is designed to study climate change and variability in chemical and physical parameters (GOOS 2003). 

North East Asian Regional (NEAR) GOOS

The primary goal of NEAR GOOS is to facilitate the sharing of oceanographic data gathered by partner countries.   The NEAR GOOS was established in 1996 and has been a product of two phases.

The first phase of NEAR GOOS was to facilitate the sharing of oceanographic data gathered by agencies of the partner countries, using the internet.  This phase has been initiated and been successful by establishing a web server that held 45GB of data in 2007.  Furthermore, the countries of China and South Korea are now updating data in real time.

NEAR GOOS is now undertaking its second phase of development which aims to address the following goals.

The NEAR region is currently working on developing projects to fulfill the goals of the 2nd phase. In particular they are establishing a generic suite of data products, initiating ground truthing for remote sensing platforms, introducing QA/QC protocols, and attempting to reduce the delay in data transmission.  Finally, they are working to conduct a needs assessment for capacity building in NEAR (GOOS 2008).

Coastal GOOS

Under the auspices of GOOS is a smaller organization focused solely on coastal resources.  It is referred to as Coastal GOOS or C-GOOS and overlaps to some degree with C-GTOS.  C-GOOS monitors the physics, chemistry, biology, biogeochemistry, and socioeconomics of the coastal zone. C-GOOS will provide data and information on coastal ecosystems worldwide to address six goals.

C-GTOS is organized into regional coastal ocean observing stations.  These observing stations have been established, but coordination between them has been limited.  Ultimately, the coordination of the regional stations will develop a Global Coastal Network (GCN). 

Using common variables at each regional site is essential for assessment of global change.

These common variables were chosen as indicators of change in the ecosystem.  C-GTOS derived these indicators by developing a checklist for indicator development. 

• does the indicator provide info on the status or condition of important ecosystem, habitat or living resources on appropriate scales?
• is the indicator based on generally accepted models of the structure and function of the system to which it is applied?
• is the indicator reliable and what is the evidence for this?
• have the data requirements for calculating the indicator repeatable at appropriate rates with known accuracy and precision been determined?
• are the required quality controlled data available in real time or post time?
• what technical and conceptual skills must the data providers poses for users to have confidence in the indicator?
• is the indicator comparable or compatible with indicators in use elsewhere?
• is the indicator cost effective in terms of the cost of providing the required data and its effectiveness to decision makers?

Currently C-GTOS is implementing several pilot projects.  These pilot projects are organized, planned activities designed to show promise as potential factors in the overall system.  Current projects are aimed at building capacity in the developing world and improving operational capabilities (GOOS, 2005).

National Estuarine Research Reserve System (NERRS)

Another monitoring program is being implemented by the National Estuarine Research Reserve System (NERRS).   NERRS believes that the impact of climate change will be first felt and severe among coastal regions.  The NERRS program has 27 sites encompassing different bio-geographic regions of the US and has been monitoring weather and water quality for more than 10 years.  Each reserve is managed by a lead state agency or university. 

The fulfillment of the three main goals in NERRS can be broken down into long term and short term objectives.  Most of the projects refer to the System Wide Monitoring Program (SWMP) which is the current monitoring program of NERRS.

To meet the first goal intertidal and emergent habitats are being mapped to provide a baseline for monitoring future climate change induced habitat changes.  Other short term activities are expanding the SETs network, and obtaining high resolution imagery.   Additionally the NERRS system will improve harmful algal bloom forecast models through partnerships with NOAA.  In the long term, the NERRS program will do more biological monitoring through the SWMP, and develop Digital Elevation Models (DEMs) and GIS mapping capabilities. NERRS has undergone an assessment of needs, and three areas have been identified: acquiring Light Detection and Ranging (LiDAR) data, creating higher classification maps, and incorporating the SWMP sites into the National Spatial Reference System.

The second overall goal of NERRS will be achieved through several small steps which include developing monitoring protocols for key ecological components across the reserve system.   Also, the reserves will be used as sentinel sites to assess estuary change throughout the country.  The sentinel sites will use existing monitoring capabilities.  Currently the SWMP is sufficient to monitor changes in the salt wedge extent, precipitation, mean water temperature, and sea level change.  Some long term activities are to produce maps of projected sea level change as well as a data synthesis every five years, focused on the ecological effects of sea level change within the system.

The NERRS system is a robust monitoring program that will aid in the understanding of ecosystem and climate change.  In addition to monitoring, the NERRS program will serve as an educational platform (NERRS 2008). 

There are several other regional- or state-based monitoring programs.  These programs may best represent how a monitoring program would look for Long Island Sound.  The first program is being conducted by the Chesapeake Bay Program.

Chesapeake Bay Program (CBP)

The Chesapeake Bay Program (CBP) developed a monitoring network because they felt that an understanding of the spatial and temporal dynamics associated with the physical processes driving the system is essential in developing mitigating strategies for these challenges.  The Chesapeake Bay Program asserted that climate change can influence the estuary in a number of ways.  The CBP also predicted the physical response of the bay as it will mediate and change through dynamics that will either amplify or buffer the magnitudes of change.  Finally, since climate is a driver and organizing factor in ecological processes it is likely that several areas of living resources will be affected by climate change(Pyke et al. 2008).

The Chesapeake Bay program has developed a base of the physical drivers of change in the Bay.  However, climate change research remains fragmented and incomplete.  The current supply of timely and relevant climate information to support management is inadequate and uncoordinated (Pyke et al. 2008).

Delaware Estuary (Climate Ready Estuaries program)

Delaware has partnered with the Environmental Protection Agency to develop a climate adaptation strategy.  Delaware is one of six pilot studies taking part in the “Climate Ready Estuaries” initiative.   The Delaware project has three case studies: wetlands, drinking water, and shellfish (CRE 2008).

The drinking water could be threatened by sea level and salinity rise.  The shellfish are one of the best sentinel indicators of ecosystem conditions. 

The Delaware pilot study will produce a final report.  This report will summarize the vulnerability, monitoring needs, and potential actions that can be taken to mitigate climate change.  Also, the report will create maps which will show the location of resources deemed vulnerable.

Albemarle-Pamlico Estuary (Climate Ready Estuaries Program)

The Albemarle-Pamlico Estuary, located in the border of Virginia and North Carolina, is also a part of the Climate Ready Estuaries program. The Albemarle-Pamlico Sounds region—30,000 square miles of watershed—is the second largest estuarine system in the United States, second only to the Chesapeake Bay.

The Albemarle-Pamlico Estuary Science and Technical Advisory Committee has based its monitoring efforts from the identification of a number of drivers and responses (APE STAC 2008).

Section IV
Limitations of current indicators and non-linear changes

Limitations of current indicators

After reviewing a variety of monitoring efforts, we concur with Niemi et al. (2004) in finding the current monitoring practices lacking in a variety of ways.

Causal relationships. In most cases, specific indicators are not linked to specific drivers, lacking any possibility of establishing a causal relationship. Without pinpointing a cause, it will be impossible to remediate the state of the indicator. Clever planning (e.g., designation of sites across a stress continuum) may aid in circumventing this problem.

Multiple stressors. Somewhat related to the previous point, it is clear that a number of stressors (anthropogenic perturbations, natural phenomena, climate change-related alterations) act simultaneously. Separating these has proven remarkably difficult because it is unclear how the many potential stressors interact (e.g., additively, multiplicatively, non-linearly). Experimentation and modeling will be likely needed to resolve this.

Spatial and temporal scales. Explicitly addressing how different stressors and responses vary over spatial and temporal scales is seldom considered. This could lead to erroneous conclusions of their interaction.

Statistical analyses. Power analysis and sample size estimation do not appear to be undertaken. Power analysis establishes how large a sample is needed to enable statistical judgments that are accurate and reliable (i.e., how likely a statistical test will be to detect effects of a given size in a particular situation). Not conducting these statistical analyses before the start of monitoring will likely result in a waste of resources (e.g., not sampling an indicator enough to obtain statistically meaningful results, or oversampling an indicator when less sampling would have sufficed).

Thresholds and non-linear changes

Current prediction and adaptation strategies rely heavily on progressive, linear changes, driven by physical phenomena. This, however, appears to be the exception, rather than the norm, in biological systems (Burkett et al. 2005). An ecological threshold (also commonly referred as tipping point, regime shift, or alternative/multiple stable state) is defined as “the point at which there is an abrupt change in an ecosystem quality, property, or phenomenon, or where small changes in one or more external conditions produce large and persistent responses in an ecosystem” (CCSP 2009). Ecological theory and models are now predicting these types of changes regularly (e.g., Sugihara and May 1990, Hanski et al. 1993, Scheffer and Carpenter 2003) and empirical evidence is mounting at an extraordinary pace (e.g., Hsieh et al. 2006, or, for a review, see Burkett et al. 2005). As Wiens (2007) stated, “These thresholds exist because all organisms and species have limits to their environmental tolerances. The physiological tolerances of individuals to heat, moisture, salinity, soil nutrient levels, and the like determine the conditions in which they can function normally. Outside these zones, their performance suffers – thresholds have been passed. These effects on individuals influence population dynamics, in some cases promoting population declines beyond sustainable levels. Individuals and populations have the options to move elsewhere, adapt, or die.”

One well-studied example of this regime-shift phenomenon occurred in the North Pacific, over the winter of 1976-1977 (reviewed in Hare and Mantua 2000). It is believed that small changes in climate over 1-3 years were responsible for the wide-ranging consequences to the ecosystem. The shift led to an increase in Alaskan salmon populations, decreases in Alaskan shrimp and west coast salmon populations, California current zooplankton abundance, and worsening in the oyster condition index. Following the triggering event, the North Pacific ecosystem was rapidly reorganized into a different stable state that included different community structure and organization, food-web dynamics, predator–prey interactions, and distribution and abundance of various species.

More locally, the coastal fish and invertebrate community of Narragansett Bay appears to have shifted from a vertebrate- and benthic-dominated community to one dominated by invertebrates and pelagic fishes (Collie et al. 2008). The shift in species composition can be attributed to the spring-summer sea surface temperature, and is obviously altering food web dynamics and the relative abundance of other species in the Bay (Collie et al. 2008).

To counteract abrupt change, it is essential to recognize that a threshold has been reached and to apply strong pressure against the stressor (or to manage in such a way as to minimize the effect—e.g., if a critical corridor is being hampered by sea level rise, manually create another). If done in a timely fashion, this could tip the system back to its original state (CCSP 2009). Thus, understanding the non-linear patterns and processes of the Long Island Sound ecosystem is essential to effectively adapt to changes brought about by climate change. However, predictability is perhaps the most difficult task when facing abrupt changes; the ability to forecast changes in stable states is in its infancy. Nonetheless, some key concepts have arisen from analysis of case studies. For instance, changes in variance across space and time may be a primary indicator of incipient non-linear change (Carpenter and Brock 2004). Further, monitoring ecosystem stress (in the form of key components of the ecosystem, like abundance of individual species or guilds) may be more informative than monitoring of species of commercial or aesthetic importance (CCSP 2009). To achieve a successful monitoring program that takes into account abrupt changes, monitoring the biological community as a whole is imperative.

Section V
Conclusions

1. Sound-wide indicators in place already may be used to detect biotic and abiotic changes due to climate change
2. CT DEP data on fishes and invertebrates could be analyzed for evidence of non-linear dynamics
3. Special consideration should be given to the planning stage to obtain answers to specified questions (e.g., if interested in disentangling causal effects, location of sites should be considered carefully)
4. Biological responses will be more effectively observed if a community approach is taken

Section VI
References

Barton, J. 2003. Estuarine health monitoring & assessment review. State Water Quality Monitoring and Assessement Committee, Victoria, Australia. 92pp.

Brush, M.J., K.A. Moore, and E.D. Condon. 2007. Synthesis of data from the National Estuarine Resaerch Reserve System-Wide Monitoring Program for the Mid-Atlantic Region. A report submitted to the NOAA/UNH Cooperative Institute for Coastal and Estuarine Environmental Technology.

Burkett, V.R., D.A. Wilcox, R. Stottlemyer, W. Barrow, D. Fagre, J. Baron, J. Price, J.L. Nielsen, C.D. Allen, D.L. Peterson, G. Ruggerone, and T. Doyle. 2005. Nonlinear dynamics in ecosystem response to climatic change: case studies and policy implications. Ecol Complex 2:357-394.

Carpenter, S.R. and W.A. Brock. 2004. Spatial complexity, resilience, and policy diversity: Fishing on lake-rich landscapes. Ecology and Society 9:8.

CCSP. 2009. Thresholds of climate change in ecosystems.  A report by the U.S. Climate Change Science Program and the Subcommittee on Global Change Research [Fagre, D.B., C.W. Charles, C.D. Allen, C. Birkeland, F.S. Chapin III, P.M. Groffman, G.R. Guntenspergen, A.K. Knapp, A.D. McGuire, P.J. Mulholland, D.P.C. Peters, D.D. Roby, and George Sugihara]. U.S. Geological Survey, Reston, VA. 156 pp.

Christian, R.R. and S. Mazzilli. 2007. Defining the coast and sentinel ecosystems for coastal observations of global change. Hydrobiologia 577:55-70.

Collie, J.S., A.D. Wood, and H.P. Jeffries. 2008. Long-term shifts in the species composition of a coastal fish community. Can J Fish Aquat Sci 65:1352-1365.

CRE. 2008. The Delaware estuary climate ready pilot study.  A Report by the Climate Ready Estuaries (CRE) program. Wilmington, DE. 2pp

FAO. 2005. Coastal GTOS draft strategic design and phase 1 implementation plan. By R.R. Christian, D. Baird, R.E. Bowen, D.M. Clark, P.M. DiGiacomo, S. de Mora, J. Jiménez, J. Kineman, S. Mazzilli, G. Servin, L. Talaue-McManus, P. Viaroli, and H. Yap. GTOS Report No. 36. FAO, Rome. 93 pp.

GOM. 2004.  Northeast coastal indicators workshop.  A Report by the Gulf of Maine (GOM). 4pp.

GOOS. 1998a. Implementation of Global Ocean Observations for GOOS/GCOS: First Session. GOOS Report No. 64. UNESCO; Paris. 25pp.

GOOS. 1998b. Implementation of Global Ocean Observations for GOOS/GCOS: Second Session.  GOOS Report No. 65. UNESCO; Paris. 16pp.

GOOS. 2003. Black Sea GOOS strategic action and implementation plan. GOOS Report No. 133.  UNESCO; Paris. 75pp.

GOOS. 2005. An implementation strategy for the coastal module of the Global Ocean Observing System (GOOS). GOOS Report No. 148. 151pp.

GOOS. 2008.  A strategic plan for the NEAR-GOOS in its second phase.  GOOS Report No. 166. 45pp.

Hanski, I., P. Turchin, E. Korpimaki, and H. Henttonen. 1993. Population oscillations of boreal rodents: regulation by mustelid predators leads to chaos. Nature 364:232-235.

Hare, S.R. and N.J. Mantua. 2000. Empirical evidence for North Pacific regime shifts in 1977 and 1989. Progr Oceanogr 47:103-145.

Hsieh, C.H., C.S. Reiss, J.R. Hunter, J.R. Beddington, R.M. May, and G. Sugihara. 2006. Fishing elevates variability in the abundance of exploited species. Nature 443:859-862.

NERRS. 2008. Climate change: Science, education and stewardship for tomorrow’s estuaries.  A Report by the National Estuarine Research Reserve System (NERRS). 16pp.

Niemi, G., D. Wardrop, R. Brooks, S. Anderson, V. Brady, H. Paerl, C. Rakocinski, M. Brouwer, B. Levinson, and M. McDonald. 2004. Rationale for a new generation of indicators for coastal waters. Environ Health Perspect 112:979-986.

NOAA. 2007. NOAA Integrated Ocean Observing System (IOOS) Program: Strategic Plan 2008-2014.  A Report by the National Oceanic and Atmospheric Adminsitration (NOAA). 17pp.

Pyke, C.R., R.G. Najjar, M.B. Adams, D. Breitbur, M. Kemp, C. Hershner, R. Howarth, M. Mulholland, M. Paolisso, D. Secor, K. Sellner, D. Wardrop, and R. Wood. 2008. Climate Change and the Chesapeake Bay: State-of-the-science review and recommendations. A Report from the Chesapeake Bay Program Science and Technical Advisory Committee (STAC), Annapolis, MD. 59pp.

Scheffer, M. and S.R. Carpenter. 2003. Catastrophic regime shifts in ecosystems: Linking theory to observation. Trends Ecol Evol 18:648-656.

Sugihara, G. and R.M. May. 1990. Nonlinear forecasting as a way of distinguishing chaos from measurement error in time series. Nature 344:737-741.

“What is GOOS?”  The Global Ocean Observing System. 25 January 2008. 

Wiens, J. 2007. Ecosystem thresholds and climate tipping points: Implications for policymakers. Environmental and Energy Study Institute, Washington, DC.