FCC workshop on network resiliency 2013

This abstract summarizes recent work we have done to enable third-party measurement of edge-network resiliency for Hurricane Sally.

This abstract summarizes observations about communication infrastructure performance obtained from field damage assessments made by me after notable recent natural disasters and explores ways of reducing communication networks outages intensity and duration. This discussion is made within the context of observed effects of Hurricane Sandy on communication networks from damage assessments I performed shortly after Sandy made landfall. Some of the other recent notable natural disasters which I studied with field damage assessment include hurricanes Katrina (2005), Dolly, Gustav and Ike (2008), Irene (2011), and Isaac (2012), and the Feb. 2010 earthquake and tsunami in Chile, the Feb. 2011 earthquake in Christchurch, New Zealand, and the March 2011 earthquake and tsunami in the Tohoku Region of Japan. A potential presentation about this topic will include extensive photographic evidence of the observations I made during my field damage assessments of these disasters, with particular focus on Sandy. Links for samples of such photographic evidence and a compilation of most of my published work on this subject can be obtained with my bio at http://users.ece.utexas.edu/~kwasinski/BioFCC.pdf

Power outages are one of the main causes of communication network outages during natural disasters and Hurricane Sandy was no exception. Reports issued by Verizon indicate that 300 central offices were affected and information collected during my damage assessment indicates that the majority of them were affected by power outages but did not loose service. Unlike Katrina, there is no indication of central office outages caused by engine fuel starvation originated in diesel procurement and delivery issues. Still, two key central offices in Lower Manhattan had at least some of their operations, if not all, interrupted when flooding damaged power backup equipment including onsite diesel generators and fuel pumps in their basements or first floor. These issues could have been prevented by locating power equipment on higher floors or by using watertight doors instead of conventional doors to access the central office buildings. Watertight doors are found in many of Japan's NTT central offices. In some sites—e.g. Kamaishi—these doors were effective in limiting the damage during the March 2011 tsunami. Power was also the main reason why an average of 25 % of the cell sites in the area affected by Sandy lost service. During the damage assessment I observed that the majority of base stations in cities and towns in the area affected by Sandy are located on buildings roofs and have no permanent gensets. This is different from what is observed in the Gulf Coast where most cell sites are placed on their own parcel and many of them have permanent gensets. Placing base stations on buildings roofs led to power restoration issues because although some of these cell sites have standard power sockets at the ground level and with easy access to connect a portable diesel generator, many sites did not have any pre-prepared way of connecting a generator at ground level so a cable had to be run from the roof to a portable genset on the street. This ad-hoc solution led to service restoration delays because of difficulties in gaining building or roof access. The discussion will also consider the increased need of communication networks users for reliable power, e.g. to charge their cell phones.

Recently, more comprehensive solutions to address power issues have been proposed. In recent years, implementation of smart grid technologies has been seen as a way of improving electric grids performance during natural disasters. Although some of these technologies allow a faster identification of electric outages and general grid condition, power grids damage repairs still requires human intervention. Moreover, as I explained in an INTELEC 2010 paper exploring the effect of smart grid technologies on communication networks, my damage assessments and statistical outage data indicate that power grids are inherently very fragile systems due to their centralized power distribution and control architectures. Since most utility-based smart grid technologies still maintain such centralized architectural and control approaches, these smart grid enabled power supply availability improvements are limited. As I explained in several of my papers, network operators-based solutions and, in particular, microgrids—independently controlled and confined power grids with their own local power generation sources, such as Verizon's Garden City central office that operated during Sandy powered by 7 fuel cells —present a better technical solution. Moreover, for high-power loads, such as central offices, microgrids are cost competitive with respect to conventional power solutions. Still, operation of some microgrid local power generation units may be affected during disasters because of their dependency on other infrastructures, such as roads for diesel supply. Some of the solutions to address dependency on other infrastructures include using technologies that have low probability of being affected by a given hazard—e.g. natural gas in hurricane prone areas—diversifying local power supply technologies, using local energy storage devices—e.g. batteries—and using renewable energy sources that do not depend on other infrastructures, such as photovoltaic (PV) systems or wind generators. Yet, relatively large footprint and output variability limits the application of PV and wind generators. Limited space and higher relative cost makes implementation of alternative power solutions even more difficult in distributed network elements: base stations, outside plant fiber nodes, such as digital loop carrier (DLC) terminals, and CATV amplifiers supporting telephony services. Although fuel cells have been deployed in some cell sites and natural gas generators have been installed in many DLC terminals along the Gulf Coast, different city architecture approaches makes such solutions difficult to reproduce in the Northeastern Atlantic Coast, leading to unsafe ad-hoc solutions observed with Sandy, such as placing portable generators on top of pole-mounted CATV amplifiers.

Although the discussion focuses on power supply issues, problems related with other communication networks infrastructure components are also commented. For example, the discussion will also explore the implications in terms of network vulnerabilities of using fiber optics remote terminal cabinets to restore service to copper cable facilities that were damaged due to a combination of pressurization failures and flooding in the aftermath of several natural disasters (including Sandy).

After causing extensive damage in the Caribbean, superstorm Sandy had devastating effects on the US East Coast, including heavily affecting Internet infrastructure in the region. Researchers at CAIDA have been developing techniques for the detection and analysis of large-scale Internet outages, through correlation of a variety of network measurements, including control-plane signaling, passive traffic collection, and distributed active probing. Experiments with correlating differents sources of data have inspired our current pursuit of new methodologies, with the ultimate goal to develop on operational capability to detect and monitor Internet outages in real-time [This work is partly funded by the NSF SaTC program CNS-1228994].

However, Sandy was a significantly different type of disruption than those we have studied thus far, with characteristics that limited and in some cases prevented us from being able to thoroughly analyze the event. These characteristics include:

• Movement over a large area, with no fixed epicenter like an earthquake has.

• High level of Internet penetration in the affected region, including major hubs for international Internet connectivity.

• Disruption was limited to only a subset of networks/hubs in the affected region, making it harder to identify geographic areas of massive impact.

Nonetheless, we were able to observe some of the impacts of Sandy through Internet measurements, and we also learned some lessons that will allow us to improve our methodologies for studying similar events in the future. We would report on both these aspects at the workshop.

The measurements we would discuss include: analysis of changes in Internet routing to Europe and Asia inferred from traceroutes conducted using the RIPE Atlas network; analysis of traffic reaching the UCSD Network Telescope that originated from the areas affected; and traceroute (forward path) measurements through CAIDA's Archipelago infrastructure.

To support decision-making in the disaster recovery situation, it is critical to collect data about situation and events as they unfold in the area of interest. Consider a scenario when the government agency is trying to assess the flooding situation in a certain area by collecting data, e.g., pictures and text messages from the people in the area. Ideally we want the agency to be able to send a query to mobile phones owned by people in the affected area without requiring the knowledge of who they are, and where they are.

TSCOPE (which stands for telescope) is a user-friendly service for mobile data collection by sending location-oriented queries to the users to get situational data without requiring explicit knowledge of the contact information of the recipients and their whereabouts. This is an ideal service to enable large-scale data collection from willing participants from a region of interest when disaster hits. Using TSCOPE a user can send text-based queries to mobile devices in a particular region, near a point of interest. The area of interest can be precisely specified by using geospatial concepts such as bounding polygons (e.g. Rockefeller center), distance-based boundaries (e.g., within 5 min driving distance from Ground Zero), and trajectory (e.g., 500 meters along the route), and the query will be only sent to the mobiles in that space.

In this talk, we will describe the software architecture and component technologies used to build TSCOPE and present a simple demo in the context of searching for healthcare providers in a foreign area without prior knowledge of their contact information or locations.

In the aftermath of the terrorist attacks on September 11, 2001, the Computer Science and Telecommunications Board (CSTB) of the National Academies formed a committee to assess the Internet's performance and reliability on that day, and to offer suggestions for ways to handle future emergencies. In 2003, the committee published a report "The Internet Under Crisis Conditions: Learning from September 11, 2001" (http://www.nap.edu/openbook.php?isbn=0309087023) that gave a detailed timeline of the impact of the attacks on the Internet infrastructure and on the ways people used Internet services. This talk presents an overview of the CSTB report, and the lessons learned about ways to improve the resilience of the Internet during major emergencies.

Aspects of the Smart Grid concept enhance robustness of the grid to both natural and manmade threats. Many smart grid implementations start with the visibility of the last mile with smart meter deployments and ways to communicate to the last mile. This is an important step to the smart grid but is not the end in making a grid self-healing and the other facets of a smart grid. Making the grid adaptive so it can respond and even anticipate failures is another key component. So also is grid-scale and building/campus scale storage technologies. And ultimately, how do we make the grid smart? How will current approaches be changed to building-out smart grids? Below are some key aspects we have been addressing:

1. An approach to Innvervation -- that is the wiring and sensing of the last mile and distribution level of the grid so that the whole grid is sensed. Disruptive technology will emerge in the world of traditional SCADA systems so that sensors are self-locating, self-organizing, self-identifying, and the data they produce "self-storing". This will be true of all kinds of infrastructure grids -- from transportation to water including the fusion of data across types of infrastructure.

2. In order to be more intelligent about the control of the grid, it needs to be modeled. Learning models from data is an important part of achieving the level of modeling needed. ML/KDD is useful approach complimenting traditional power flow models. In order to react adaptively and intelligently, the tools of multistage decisions under uncertainly, such as approximate dynamic programming or stochastic programming can be applied.

3. In order to accept intermediate sources of power the grid needs "elasticity". This is addressed with electrical storage technologies at different levels of the grid, from the end customer though distribution up to transmission.

I will elaborate these points by discussing both research and policy issues I have been involved with at the Center of Computational Learning Systems and World Team Now.

Recent disasters such as Hurricane Sandy demonstrate that our network infrastructures need to be better prepared for disasters, and they require intelligent and efficient recovery methods during post-disaster events. Note that telecom backbone networks employ optical mesh structures to provide highly-scalable connectivity across large distances; and these networks along with their "higher-layer" (virtual) networks (e.g., IP, MPLS, SONET, Ethernet, ATM) are integral to our economic well-being and national security because they are widely deployed in commercial and defense sectors to support many aspects of our daily life, cloud computing, battlefield surveillance/backhaul, etc. Thus, the need for survivability against disasters is acute, given the scale and criticality of these networks.

Techniques exist (and are implemented in operational networks) to provide fast protection at the optical and other layers, but they are optimized for limited faults without addressing the extent of disasters. Typically, failures caused by disasters are correlated, and cascading, and are much more dynamic than failures recoverable by known techniques. So, there is a pressing need for novel, robust survivability methods to mitigate the effects of disasters on backbone and virtual networks. To address this challenging problem without incurring excessive redundancy, we are conducting research sponsored by the Defense Threat Reduction Agency (DTRA) to investigate methods to provide effective protection against disasters and adaptive network algorithms during disaster events by developing protection techniques for dynamic re-provisioning, multipath routing, and data replication. In particular, we are developing methods for the following: (1) Normal Disaster Preparedness (by accounting for risk of disasters in different parts of the infrastructure); (2) Enhanced Disaster Preparedness (under more-accurate intelligence on potential disasters); and (3) Post-Disaster Service Survivability. Note that while traditional approaches focused on protecting links and nodes (routers, switches, etc.) to provide "network connectivity", the shifting paradigm towards cloud computing/storage require that we protect the data (or content), so we have developed the concept of "content connectivity" and methods to achieve this.

Normal Preparedness: Network operators should proactively take necessary actions to minimize network disruptions and data loss in case of a disaster. Knowledge of possible disaster zones (i.e., risk information) would help to utilize network resources and disseminate data accordingly. For instance, seismic hazard maps would be useful to determine vulnerable parts of the networks and to define risk of traversing connections through these vulnerable parts in case of an earthquake. Similar methods can better prepare our military networks against their threats

Enhanced (Better) Preparedness: If a disaster is predicted through scientific measurements and observations (e.g., the possible path and estimated time of arrival to main land of a hurricane occurring in the ocean can be known days in advance), the network can be better prepared by re-allocation of network resources and re-dissemination of data, and possibly by relocation of hardware resources also.

Post-Disaster Events: After an attack, network resources can become limited, and if full bandwidth cannot be guaranteed, the services should be provided with as much bandwidth as possible (degraded services). Disrupted connections can be reprovisioned on the surviving network resources. The information on network recovery (e.g., through FCC's Disaster Information Reporting System (DIRS)) can give information on network status, and this information can be exploited to better meet the service needs and current network state.

The methods to prepare the network for possible disasters, to better prepare for upcoming disasters, and to provide some minimal level of services after a disaster to support critical operations and recover services while network is recovering can significantly improve network robustness for disasters. If given an opportunity, we would be delighted to share our research activities and findings from our DTRA project at the upcoming FCC workshop on disaster preparedness.

Recent massive failures of the power grid demonstrated that large-scale and/or long-term failures will have devastating effects on almost every aspect in modern life, as well as on interdependent networks. In particular, telecommunications networks are vulnerable due to their strong dependence on power networks. Communication and power networks are vulnerable to natural disasters, such as earthquakes, floods, hurricane, and solar flare as well as to physical attacks, such as an electromagnetic pulse (EMP) attack. Such real world events happen in specific geographical locations and hence cause geographically correlated failures. Therefore, the geographical layout of the network determines the impact of such events.

This talk will focus on our recent results regarding the vulnerability of telecommunications networks and power grids to geographically correlated failure. We will present methods to identify the locations most vulnerable to large scale disasters. Our approach allows for identifying locations which require additional protection efforts (e.g., equipment shielding). Moreover, it may provide input to network and protocol design that could avert geographical disasters or attacks.

The Internet is a critical infrastructure on which we depend, and thus it is essential that it be resilient such that it continues to provide service in the face of various challenges, including attack and large-scale disasters such as earthquakes, hurricanes, tsunami, and coronal mass ejections [1, 2]. A major aspect of our previous and current work on achieving resilience has centered on providing diversity in the network such that when part of the network fails, alternatives will be available to continue operation. This includes heterogeneity and diversity in mechanism (for example wired and wireless) [3], rich topology interconnection, and structural diversity of the network graph such that paths can be constructed that do not share fate when network components fail [4]. We have developed a set of analytical and simulation techniques and tools to generate network topologies, and to analyse the resilience of real and synthetic network graphs [5]. We have also made our data and topology viewer publicly available at http://www.ittc.ku.edu/resilinets/maps. A key aspect of this ongoing work is the multilevel nature of the analysis, in particular, attacks against the physical infrastructure must be modelled on the physical layer graph (fiber interconnection for the typical backbone) but its effects analysed on the IP network layer graph overlay. Under new funding from NSF NeTS (in collaboration with Deep Medhi at UMKC), we are exploring geographic diversity and its impact on traffic load, such that networks can be designed to survive large-scale disaster of a given scope. For example, an application should be able to specify: give me three multipath routes over which communication can be erasure coded such that the paths are no closer than 100 km (except at the source and destination). This example defends against a disaster with a diameter of less than 100 km in diameter. While we work to understand how to generate networks with desired graph-theoretical, diversity, and resilience properties, the reality is that even if adopted, there will be cases where disasters will partition the network, either because the area is greater than anticipated, or cost constraints have not permitted the deployment of sufficiently resilient infrastructure.

Thus, we are beginning research on how to optimally and rapidly deploy infrastructure after a disaster, in particular, to restore services outside the disaster area, to rapidly deploy assets to permit assessment of the damage to the environment and network, and to rapidly and optimally deploy infrastructure to restore critical network infrastructure to the affected area. This is joint work with Chinese instutions (Jiannong Cao at Hong Kong Poly and Jinyao Yan at CUC Beijing) as a result of our participation in the US NSF / China NSFC Workshop on Environmental Monitoring or Public Health and Disaster Recovery.

References:

[1] James P. G. Sterbenz, Rajesh Krishnan, Regina Rosales Hain, Alden W. Jackson, David Levin, Ram Ramanathan, and John Zao. Survivable mobile wireless networks: issues, challenges, and research directions. In Proceedings of the 3rd ACM workshop on Wireless Security (WiSE), pages 31-40, Atlanta, GA, 2002.

[2] James P. G. Sterbenz, David Hutchison, Egemen K. Cetinkaya, Abdul Jabbar, Justin P. Rohrer, Marcus Schöller, and Paul Smith. Resilience and survivability in communication networks: Strategies, principles, and survey of disciplines. Computer Networks, 54(8):1245-1265, 2010.

[3] James P. G. Sterbenz, David Hutchison, Egemen K. Cetinkaya, Abdul Jabbar, Justin P. Rohrer, Marcus Schöller, and Paul Smith. Redundancy, Diversity, and Connectivity to Achieve Multilevel Network Resilience, Survivability, and Disruption Tolerance (invited pa- per). Springer Telecommunication Systems, 2012. (accepted April 2012).

[4] Justin P. Rohrer, Abdul Jabbar, and James P.G. Sterbenz. Path diversification for future internet end-to-end resilience and survivability. Springer Telecommunication Systems, 2012. accepted April 2012.

[5] James P.G. Sterbenz, Egemen K. Cetinkaya, Mahmood A. Hameed, Abdul Jabbar, Qian Shi, and Justin P. Rohrer. Evaluation of Network Resilience, Survivability, and Disruption Tolerance: Analysis, Topology Generation, Simulation, and Experimentation (invited paper). Springer Telecommunication Systems, pages 1-32, 2011. Published online: 7 December 2011.

Bio:

James P.G. Sterbenz is Associate Professor of Electrical Engineering & Computer Science and a member of technical staff at the Information & Telecommunication Technology Center at the University of Kansas, and is a Visiting Professor of Computing in InfoLab 21 at Lancaster University in the UK. He has previously held senior staff and research management positions at BBN Technologies, GTE Laboratories, and IBM Research. His research interests include resilient, survivable, and disruption tolerant networking, future Internet architectures, active and programmable networks, and high-speed networking and components. He is director of the ResiliNets Research Group, PI in the NSF-funded FIND and GENI programs, the EU-funded FIRE ResumeNet project, leads the GpENI international programmable network testbed project, and has led a US DoD project in highly-mobile ad hoc disruption-tolerant networking. He received a doctorate in computer science from Washington University in 1991. He has been program chair for IEEE NGNI, GI, GBN, and HotI; IFIP IWSOS, PfHSN, and IWAN; is on the editorial board of IEEE Network, and is chair of IEEE ComSoc TCCC and formerly TCGN. He is principal author of the book High-Speed Networking: A Systematic Approach to High-Bandwidth Low-Latency Communication.

The convergence of the cellular industry on a common standard, 4G LTE, offers an opportunity, heretofore unavailable, to build robustness across competing carriers during man-made and natural disasters. Aggregating resources, be they on the airlink or the backhaul, would allow emergency communications to be provided to customers. On the airlink, we will show that if two identical carriers pool their resources, they can double the capacity delivered to each customer, even as their combined pool of customers doubles. Equivalently, this means near normal service can be maintained even if half the cell towers are lost during a disaster. For backhaul, opening residential femtocells to all cellular users would allow islands of service even if all macrocells and/or their backhaul networks failed. These are two examples of using the diversity across wireless and wireline carriers to provide inherent robustness in service. Clearly a regulatory and policy environment that encourages cooperation between competing service providers would be needed to bring this vision to fruition. An advantage of this approach is the relatively modest incremental cost.

This talk will discuss the benefits of fiber to the home (FTTH) as a reliable and energy efficient solution for broadband access. Fiber optic cables are more robust against floods than copper cables. The replacement of damaged copper cables and considerations to install utility cables underground in the aftermath of super storm Sandy provide an opportunity to roll out a future proof wireline infrastructure. A passive optical network (PON) does not require active equipment in the outside plant, which avoids the need for powering remote nodes and reduces the risk of failures. PON is the most energy efficient broadband access technology thanks to the low signal attenuation of the fiber transmission medium and the sharing of a single interface in the central office by multiple subscribers.

We will discuss research directions pursued in the GreenTouch research consortium to drastically reduce the power consumption even further. This will extend the service availability for a given power back-up capacity at the central office and customer premises in case of a power outage. An improved energy efficiency of the customer premises equipment (CPE) enables new power back-up approaches, such as easy-to-replace consumer batteries or small solar cells.

As interoperability tools and new technologies such as LTE and FirstNet deploy, responders gain great flexibility. But with that flexibility comes new responsibility.

As we alleviate the traditional hard-wiring of public safety communications into fixed assignment of channels and talkgroups, we also remove a great deal of implicit structure and context that made network management and use easier. Things like namespace management (i.e., callsigns) and network protocols (procedures and codes) can no longer be treated as merely local concerns.

Suddenly--when we can connect, for example, the FBI and local police instantly and seamlessly by radio--we then must address a variety of new issues. Do the two groups of users know each others's callsigns and assignments? Do they understand each others radio codes and procedures? For that matter, how do they find each other in the first place? Where's the phone book? Or is it more like Google? And who gets access to what, and who says so?

While this is a relatively new class of problems in the Land Mobile Radio domain, it's familiar ground in the Information Technology discipline of Embedded Systems and Web Services. Challenges of "service discovery" and "directory services," for example, have been addressed in a variety of ways in data systems, some of which may serve as examples for needed support services in our emerging emergency communication networks.

As a direct consequence of the Loma Prieta earthquake in 1989, 154 of 160 telephone central offices in Northern California lost primary power, and some of these also lost their backup power. The impacted communities had an expectation of rapid restoration of communication services - in that day and age, this meant voice calling over POTS circuits. Amateur radio operators were able to provide supplementary voice communications during the recovery efforts because individuals had taken steps to become licensed, build skills and prepare equipment that would operate in the absence of communications infrastructure.

But since that time, both the expectations and the infrastructure have changed in ways that have reduced communications resilience for communities. The rise of the internet and, more recently, smartphones and social networks, have shifted expectations toward mobile IP-based data services (e.g., exchanging health and welfare information from phones via text and images; vesting the storage of critical banking and health information in online services). And the nature of this new infrastructure is inherently less robust than central-office-based POTS. Cell sites are constructed with limited backup power (reportedly, more than 25% of the cell sites in the affected area went down during Sandy). Wireless technologies have unique vulnerabilities. Cellular signaling is particularly prone to overloading. Consumer-grade internet services are not engineered to pre-divestiture AT&T standards.

While full restoration of internet access will be important, it stands to reason that restoring communication infrastructure within an impacted community using familiar-feeling communications tools will be among its top priorities in an emergency situation.

The Survivable Social Network (SSN) project at Carnegie Mellon University is developing a solution to this problem. Our approach addresses, among other things, (a) rapidly-deployed replacement infrastructure that can meet basic communication needs and (b) a rich and flexible means for authorities to communicate with citizens.

The presentation will give an eye-witness account of the loss of power and connectivity in downtown Manhattan, as well as an observation of social behavior to find wireless connectivity in Manhattan the morning after.

Weather related natural disasters occurring over the past two years have created an opportunity for an examination of assumptions regarding communications networks, interdependencies of communications platforms, and risk probability at a regional level. Sprint Nextel Corporation's ("Sprint") planning for network continuity and recovery is a constant process. Sprint maintains internal teams of personnel dedicated to ensuring continuity of service and expeditious network recovery. These expert teams adapt their forward-looking readiness and disaster response plans from lessons learned from past events, and have adapted their plans based on the lessons learned from the weather events of 2011 and 2012.

Hurricanes Irene and Sandy were unique due their extremely large footprints and severe impacts. Hurricane Irene made landfall in North Carolina and impacted every state from there through New England, while Hurricane Sandy impacted every state from the mid-Atlantic through New England. The frequency of such storms impacting the northeast is unprecedented. The scale, volume and frequency of these storms present a tremendous opportunity to analyze and plan for future events. Sprint's readiness, mobilization and deployment protocols were adapted between these storms, and Sprint continues to adapt its approach based on lessons from Hurricane Sandy. Sprint's presentation will focus on how wireless networks are affected differently by each disaster, and the vulnerabilities, risks and mitigation strategies that are involved in preparation and response.

Diversity, or as it is some times called, redundancy, is a well-known approach for designing resilient systems. In the context of communications networks, it is typically synonymous with the availability of multiple, disjoint paths to a destination. In wireless settings, the availability of multiple paths arises naturally, and the challenge is in properly exploiting them. In large-scale wired networks such as the Internet, the distributed design decisions behind their deployment often make it difficult to even ascertain the presence of path diversity, and the many policy constraints under which the Internet operates can further limit it.

In this talk, I will first briefly report on an investigation seeking to establish the level of path diversity available in the Internet, and review simple solutions to improve it. The results hint at the fact that Internet resiliency could be enhanced by modifying its protocols to better leverage existing diversity.

Next, I will outline a simple multipath protocol aimed at wireless mesh networks with variable link conditions, and show how its reliance on multiple paths can significantly improve transmission stability even under adverse link conditions.

Finally, I will outline why in spite of the significant benefits that multipath solutions can afford in both wired and wireless settings, fully realizing those benefits may call for coordinated changes across networks and end-systems.

A brief summary of the climate projections for the NYC metropolitan region for the 2020s, 2050s and 2080s will be given with particular emphasis on hurricane frequencies, and how sea level rise will drastically increase the frequency of severe coastal storm surges. These updated findings will be discussed in the context of the general findings in the New York State issued, pre-Sandy climate change adaptation report: ClimAID, Chapter 10: Telecommunications, accessible at http://www.nyserda.ny.gov/climaid.

Red Hook Initiative WiFi is a collaboratively designed community mesh network. It provides Internet access to a small area in the Red Hook section of Brooklyn, NY, and serves as a platform for developing local applications and services. Red Hook Initiative has built the network in partnership with the Open Technology Institute, putting human-centered design and community engagement at the core of the project. The community expanded the network with the help from recovery volunteers and FEMA significantly following Superstorm Sandy in November 2012. This case study shows how lightweight community built infrastructure can facilitate rapid emergency response due to the social and technical infrastructure already in place.