Loading...
HomeMy WebLinkAbout01.28.20 Email from Jillian Beck - Policy Analysis on Wildfire Alert and Warning System DATE: December 9, 2019 TO: Butte County Board of Supervisors FROM: Jillian Beck Master of Public Administration Student Cal State LA RE: Policy Analysis of Butte County Wildfire Alert and Warning System Executive Summary During the November 2018 Camp Fire in Paradise, Butte County, the deadliest fire in state history, thousands of residents failed to receive timely evacuation warnings, preventing many from fleeing before flames reached their homes. This policy analysis examines and provides recommendations for alert and warning system. Criteria of efficiency, effectiveness, and cost were used to evaluate policy options regarding whether alert and warning systems should use warning message templates, privately operated and/or federal wireless warning systems, and a combination of non-wireless alternative warning methods. Prior to the start of wildfire season in August 2020, the Butte County Board of Supervisors should direct county staff to ensure use and review of warning message templates for adherence to best practices; the use of federal and privately operated wireless warning software systems that are tested quarterly; the development of community watch groups to assist with neighborhood evacuation warnings; and the establishment of public education and outreach campaigns to ensure the effectiveness of recommended warning methods. Problem Definition Californias recent deadly wildfire seasons have brought renewed attention to failures in public alert and warning systems during natural disasters. During the November 2018 Camp Fire in Paradise, Butte County, thousands of residents failed to receive timely evacuation warnings, preventing many from fleeing before flames reached their homes (Cal. Legis., 2019). The wildfire became the deadliest in state history with the deaths of 85 people (Cal. Legis., 2019; California Department of Forestry and Fire Protection, 2018; California Department of Forestry and Fire Protection, 2019). As the fire quickly spread, local officials utilized a county-contracted, privately owned emergency notification system to send staggered evacuation warnings by landline and cellphone to previously designated evacuation zones, but thousands of impacted residents did not receive the warnings either due to failed connections or because they did not subscribe to the service, and no citywide evacuation order for Paradise was ever sent (St. John & Serna, 2018; St. John, Serna, & Lin, 2018; Butte County Grand Jury, 2019). Evacuation warnings were not successfully sent through a widespread federal wireless warning system, and fire damage to cellular network and internet infrastructure eventually shut down all cellular and internet warning capabilities (Cal. Legis., 2019; St. John, Serna, & Lin, 2018). 1 came one year after similar warning system failures in Sonoma County during the deadly Wine Country Fire in October 2017 (Cal. Legis., 2019). This policy analysis will examine and provide recommendations for alert and warning system with a focus on components related to communication, including preparing warning message templates and the selection and use of alert and warning methods, including wireless warning systems. Background During the first hours of the Camp Fire on Nov. 8, 2018, the wildfire rapidly became uncontrollable (Butte County Grand Jury, 2019). ecause of the speed, a lot of the normal systems that we have been using and have been effective for decades just kind of went out the window,recalls Jim Broshears, retired Paradise fire chief and emergency operations coordinator for the town. At the same time, Butte County faced resource challenges that impacted its ability to notify residents of evacuation warnings. Only several staff members were trained to send out emergency notificationstwo were occupied working dispatch, one was out of the area, and the who serves messaging for the first 24 hours (S. Collins, personal communication, November 14, 2019). Wireless Emergency Notifications and Templates During the Camp Fire, opt-in CodeRED notifications were used to alert residents of evacuations; however less than 40% of residents at the time were registered for the notifications and only approximately 7,000 of 52,000 evacuees received warnings (Butte County Grand Jury, 2019). According to Lt. Collins, Butte County attempted to send out evacuation warnings as Wireless Emergency Alerts (WEA) through System (IPAWS) (S. Collins, personal communication, November 14, 2019; FEMA National Advisory Council, 2019; Cal. Legis., 2019). The CodeRED software is used to send both CodeRED messages and WEA messages through IPAWS (S. Collins, personal communication, November 14, 2019; FEMA, 2019a). Launched in 2011, IPAWS is operated by the Federal Emergency Management Agency (FEMA) and is an internet-based tool that authorities at the federal, state, local, tribal, and territorial levels can utilize to disseminate public warnings and alerts (State of California, 2019; FEMA, 2018). Wireless Emergency Alerts are emergency messages sent to mobile devices by authorized alerting authorities through the major mobile carriers; they can be targeted to specific geographic areas and are accompanied by a repeated unique alert tone and vibration that distinguish the messages from other notifications (State of California, 2019; FEMA, 2018). Wireless Emergency Alerts are opt-out, meaning users of mobile devices will receive the warnings unless they deactivate the messages on their mobile device (State of California, 2019). This system allows for alerting authorities to send emergency warnings to all people with mobile devices in a specified area, including visitors from out of county or out of state, unlike CodeRED, which is opt-in and only sends messages to members of the public who seek out and sign up to receive alerts, which likely leaves out residents of neighboring jurisdictions or out-of-town or out-of- state visitors in the event of a disaster (Cal. Legis., 2019; State of California, 2019). 2 During the Camp Fire, an emergency official attempted to send the evacuation warnings through CodeRED to the opt-in system and the opt-out IPAWS system at the same time (S. Collins, personal communication, November 14, 2019). The message appeared to send through both avenues, Lt. Collin said; however, the IPAWS message never went out (S. Collins, personal communication, November 14, 2019). Butte County officials have not received guidance from FEMA regarding why sending the messages at the same time does not work, Lt. Collins said; instead, since the Camp Fire, Butte County officials send the CodeRED and IPAWS messages separately and verify the message was logged in the IPAWS system subsequently (S. Collins, personal communication, November 14, 2019). There is currently a lack of training and continuing education for alerting authorities regarding IPAWS, which FEMA identified as a priority to remedy, set to be implemented over the next 24 to 36 months (FEMA National Advisory Council, 2019). The California Office of Emergency Services (2018) identified a lack of training and knowledge of the IPAWS Wireless Emergency Alerts system as a driving factor as to why Sonoma County officials failed to utilize the system as a means of issuing timely evacuation warnings during the Wine Country Fire in 2017. A forthcoming training being developed by the California Office of Emergency Services, as mandated by Senate Bill 833, is required to include information on the evaluation, purchase, and operation of Wireless Emergency Alerts system and Emergency Alert System equipment and software (S. Bill No. 833, 2018). Though IPAWS certification is free, alerting authorities must still purchase the software used to operate the Wireless Emergency Alerts system; as of May 2019, 25 software providers have demonstrated capabilities for supporting the system (FEMA, 2019a). Technical limitations also exist within IPAWS Wireless Emergency Alerts. Currently, there is a 90-character limit for warning messages sent through the system (FEMA, 2019b); however, FEMA has a goal to increase the limit over the next several years to 360 characters (FEMA National Advisory Council, 2019). Such limited space for a warning can be challenging and it is highly unlikely for emergency officials to be able to write effective, brief mobile warning messages in the midst of an emergency (Bean, Sutton, Liu, Madden, Wood, & Mileti, 2015). message is preferable to a t (2019) new guidelines aim to address this issue through its suggested templates. Another limitation of IPAWS Wireless Emergency Alerts, and other wireless alert methods, is their reliance on internet and/or cellular network connectivity due to the possibility of such infrastructure failing during natural disasters or a lack of connectivity in rural areas (Li, Cova, Dennison, Wan, Nguyen, & Siebeneck, 2019). The Wireless Emergency Alerts system froze during Hurricane Harvey in Texas when officials attempted to alert a large geographic area, and damage to communication infrastructure in the Caribbean prevented the system from being used during Hurricane Maria (FEMA National Advisory Council, 2019). Similarly, cell towers were damaged during the Camp Fire, which prevented many residents from receiving warnings or information through their mobile devices (Cal. Legis., 2019). Best Practices In response to communications failures during recent wildfires, including the Camp Fire, the California Legislature passed Senate Bill 833 in 2018, mandating the Office of Emergency 3 Guidelines, which were released in March 2019 (S. Bill No. 833, 2018; Cal. Legis., 2019; State of California, 2019). The guidelines provide best practices, guidance, and expectations for localities throughout the state to use to update, revise, or create specialized alert and warning plans and procedures for their communities (State of California, 2019). Research into public natural disaster warnings, dating back to the mid-19th century, coalesces around the idea that to be effective warnings must be timely and reach all those who are at risk regardless of where they are located or what they are doing (Anderson, 1969; Working Group on Natural Disaster Information Systems Subcommittee on Natural Disaster Reduction, 2000; Steelman & McCaffrey, 2012). Many components go into ensuring the timeliness, accuracy, and thus, effectiveness, of disaster warning systems. Warning Messages. In a review of 200 publications studying public disaster warnings, Mileti and -8). Disaster warning messages must be clear, consistent, accurate, and credible (Mileti & Sorenson, 1990; State of California, 2019), should be locally-specific, and should include guidance of what protective action to take that can rey, Velez, & Briefel, 2013, p. 21). During the Wine Country Fires in Sonoma County in 2017, officials often had to improvise warning message text in the midst of the crisis due to a lack of pre-written templates (California Office of Emergency Services, 2018). Prior to the Camp Fire, Butte County did not use warning ency manager (S. Collins, personal communication, November 14, 2019). However, in the year since the wildfire and following the templates into its process. The county was able to implement the templates into its system (S. Collins, personal communication, November 14, 2019). In its new guidelines, the State of California (2019) recommends the utilization of templates and provides examples for localities so they can craft effective messages prior to disaster striking. The templates include identifying the credible source the warning is coming from, describing the hazard, identifying the impacted location, advising specific protective actions, and alerting the public to when the warning expires or when new information can be expected (State of California, 2019). Alternative Warning Methods. Recent public disaster warning literature emphasizes the need for multiple methods of disseminating warning messages to the public during a natural disaster (Working Group on Natural Disaster Information Systems Subcommittee on Natural Disaster Reduction, 2000; Steelman & McCaffrey, 2012; Mileti & Peek, 2000). A report by the U.S. Working Group on Natural Disaster Information Systems Subcommittee on Natural Disaster Reduction Warnings (2000) recommended that warnings during natural disasters be distributed risk can receive warnings regardless of their location or present activities (p. 7). Mileti and Peek (2000) have found that warnings disseminated across multipl 4 Mileti and Sorenson (1990) identify several potential communication methods for disaster warnings: personal notifications or door-to-door warnings, portable loud speakers and public address systems, radio and tone alert radio, television and cable overrides, signs, reverse-911 telephonic warnings, sirens and alarms, and aircraft. Over time, the possible methods of communicating warnings to the public have vastly expanded, especially with advances in technology, and now include social media, text messages and apps, digital billboards, email, and websites (State of California, 2019). The need for multiple communication avenues is necessitated by limitations that exist within each method, especially those that rely on internet or cellular network connectivity (Li et al., 2019). In light of the unique limitations of each method, the California Legislature and state emergency officials, in line with the literature, have During the Camp Fire, Butte County used various warning methods, including wireless emergency notifications, door-to-door notifications, traditional media, ham radio operators, and social media (S. Collins, personal communication, November 14, 2019). In the year since the fire, - low tone that will only be used for evacuation warnings (S. Collins, personal communication, November 14, 2019; Mensch, 2019). Some neighborhoods in the county have also started community watch groups that focus on emergency warnings (S. Collins, personal communication, November 14, 2019). A liaison with these groups is in touch with county emergency officials through radios and can assist with disseminating evacuation warnings when needed (S. Collins, personal communication, November 14, 2019). alert and warning system, especially as it relates to the Camp Fire (Howle, 2019). The state auditor recommended Butte, and other high-risk counties, improve its preparedness and urged the state to commit more resources, support, and oversight to ensuring localities are adequately prepared to alert and warn their residents in the event of a disaster (Howle, 2019). California Board of Forestry and Fire Protection, 2019). In the early days of the Camp Fire, then-Gov. Jerry Brown referred to the sifying fires caused by factors including the effects of climate change (California Board of Forestry and Fire Protection, 2019; Ashton, 2018). Thus, uncontrollable and fast-moving wildfires like the Camp Fire are likely to reoccur. Criteria for Policy Analysis For the purposes of this policy analysis, several criteria were identified to be used to evaluate policy options for evacuations. These criteria include the effectiveness of reaching the affected population of residents, effectiveness of communicating the warning message, efficiency of resources and time spent by emergency management staff, and cost. These criteria will be used to evaluate options related to warning message templates, selecting among wireless alerting methods, and utilizing additional alternative warning techniques. 5 Effectiveness Effectiveness, whether the policy option achieves its intended effect (Patton, Sawicki, & Clark, 2013), will be valued as the most important criteria in this analysis. The ability to reach as close to all residents and visitors of Butte County during a wildfire in order to deliver evacuation warnings is imperative, as evidenced by state legislative reports on failures during the Camp Fire as well as recommendations by local accountability boards and interviews conducted with local emergency officials (Cal. Legis., 2019; Butte County Grand Jury, 2019; S. Collins, personal communication, November 14, 2019). Butte County Lt. Collins that we can reach as much of the public as possible and get them the information they will need in order communication, November 14, 2019). Effectiveness includes two outcomes: reaching all of the affected people and communicating the necessary warning. Efficiency Efficiency, or maximizing the benefits from the use of resources (Patton et al., 2013), will be evaluated as a criterion in this analysis. In the context of wildfire evacuation warnings in Butte County, efficiency relates to the speed of warning methods and the use of resources, including staff and time. This criterion was chosen due to Butte Countys finite amount of resources and staff time and the necessity to deliver warning messages quickly as discussed previously. Efficiency will be prioritized over cost due to the urgency of the policy outputs and outcomes. Cost Cost will be used to evaluate policy options related to wildfire alert and warning systems. These costs include tangible costs, expenses that can be feasibly counted such as spending on software or infrastructure and staff time, as well as further monetizable costs that can be accounted for with dollar amounts (Patton et al., 2013). Adopting and implementing new software, conducting trainings, and exploring new methods will require different levels of county expenditures and use of resources. In recent years, Butte County has had to cut spending due to reductions in revenue in the form of state and federal funding, which has resulted in job cuts (Schoonover, 2018). Therefore, cost will be of importance to the county when evaluating any policy recommendations. However, due to the political saliency of this policy issue (Butte County Grand Jury, 2019), cost will be weighed as the least important policy criterion. Policy Options for Alert and Warning System This policy analysis is divided into three subsections: warning message templates, wireless warning methods, and alternative warning strategies. An outline of options follows for each area of the warning and alert system. Warning Message Templates. The options for warning message templates include using templates or not using templates. Warning message templates are pre-scripted messages that are written prior to a disaster striking; they include clear directions on the protective actions the public is being asked to take (State of California, 2019). 6 Wireless Warning Methods. The options for wireless warning methods are CodeRED, or other privately operated opt-in emergency notification systems, Wireless Emergency Alerts through -out mass notification system, or both. Alternative Warning Techniques. The options for alternative warning techniques include sheriff and police patrol car sirens, permanently installed siren systems, door-to-door notifications, and community watch programs. Based on the evaluation, some, all, or none of these options may be recommended. Methods of Analysis A Criterion-Alternative Matrix Analysis and cost-benefit analysis will be used to evaluate the policy options using the previously established criteria. Qualitative interviews, government record research, and a literature review have also been conducted and will be used when assessing policy alternatives. Triangulation, or using multiple research methods, can help avoid potential bias and strengthen the validity of research findings (Newcomer, Hatry, & Wholey, 2010). Criterion-Alternative Matrix Analysis Three separate Criterion-Alternative Matrix (CAM) Analyses will be used to evaluate the proposed policy options: one regarding the use of warning message templates, another concerning wireless warning methods, and the final to evaluate alternative warning methods. Each of the alternatives within the wireless warning methods and the alternative warning strategies are analyzed individually; however, the options may be adopted in varying combinations dependent upon recommendations. The criteria of effectiveness, efficiency, and cost have been weighted from 1 to 3 for this analysis, with 1 being least important and 3 being most important. Effectiveness, for both reaching all of the affected population and communicating the necessary message, is weighted at 3, efficiency at 2, and cost at 1. These weights were selected based on qualitative interviews with local officials, state and local government reports, and best practices from emergency communications literature that put the most importance on reaching all people with accurate protective messaging over efficiency or costs (S. Collins, personal communication, November 14, 2019; State of California, 2019; Mileti & Sorenson, 1990). Each policy option is weighted from 1 to 3, with 1 representing low, 2 representing medium, and 3 representing high. In the case of the criterion cost, 1 equates to a high cost, 2 a medium cost, and 3 a low cost. In the attached Appendices A, B, and C, raw scores and weighted scores are displayed. Total scores were calculated by multiplying the column weight (criteria) with the row weight (alternative). Higher scores indicate favorable policy options based on the identified criteria. CAM Analysis of Warning Message Templates Appendix A illustrates the CAM Analysis of the policy options for the use or non-use of warning message templates. Not using pre-written templates, or writing emergency messages entirely after the onset of a wildfire, received a low efficiency score, a low effectiveness score, and a 7 medium cost score. Using templates received a high efficiency score, a high effectiveness score, and a medium cost score. In studies of alert and warning systems, the warning messages one of the most important factors in determining the effectiveness of Mileti & Sorenson, 1990, p. 3-8). In past wildfires in other localities, having to improvise the text of warning messages during the onset of a wildfire hindered communication efforts (California Office of Emergency Services, 2018). Additionally, the State of California (2019) recommends the utilization of templates and provides examples for localities so they can craft effective messages prior to disaster striking. Cost will be examined in closer detail in the subsequent cost-benefit analysis. Not using pre- written templates was scored at a 7 and using templates was scored at a 17. Templates were not used by Butte County officials during the Camp Fire; however, Lt. Collins reported that templates have been drafted in the year since (S. Collins, personal communication, November 14, 2019). CAM Analysis of Wireless Warning Methods Appendix B illustrates the CAM Analysis of wireless warning methods, including the use of CodeRED, or other privately operated emergency notification systems, and the use of IPAWS WEA, the federal wireless warning system. CodeRED received a low score for effectiveness of reaching all populations, a high score for effectiveness of communicating emergency warnings, and a medium score for cost. IPAWS received a high score for effectiveness of reaching all populations, a medium score for communicating emergency warnings, and a medium score for cost. The opt-in nature of CodeRED limits its ability to reach all populations, as illustrated by the low registration rate of residents during the Camp Fire, and all but eliminates the capability to reach out-of-town visitors (Butte County Grand Jury, 2019; J. Broshears, personal communication, November 18, 2019). Alternatively, IPAWS was scored as effective for geographic area without prior registration (State of California, 2019; FEMA, 2018). However, CodeRED allows for 160-character messages, while IPAWS currently only supports 90-character messages, which county emergency officials have found to be extremely limiting (S. Collins, personal communication, November 14, 2019). effectiveness score for communicating warnings was also due to previous technical difficulties experienced by Butte County officials when attempting to use the system (S. Collins, personal communication, November 14, 2019). Overall, CodeRED received a weighted score of 14 and IPAWS received a weighted score of 17. CAM Analysis of Alternative Warning Methods Appendix C illustrates the CAM Analysis of alternative warning methods, in other words methods to mitigate gaps left by the wireless warning methods discussed previously. Patrol car sirens were scored at 14, a permanent siren system at 13, door-to-door notifications at 17, and community watch programs at 24. Patrol car sirens, which have recently been implemented in Butte County, received a medium score for effectiveness of reaching all populations, effectiveness for communicating warning message, and cost because some patrol cars may be limited in reaching all populations during a 8 The permanent siren system received a medium score for effectiveness of reaching all populations, a medium score for effectiveness of communicating messages, and a low score for cost. If sirens are placed throughout the county, it is likely most populations will receive the messages; however, this is dependent upon education of the public about the siren meanings (Cal. Legis., 2019; Kuligowski & Wakeman, 2017). Some emergency officials dispute the effectiveness of sirens of reaching all communities (such as those who are hearing impaired) and warn of negative unintended consequences due to loud noise levels (Berkeley Disaster and Fire Safety Commission, 2019). Additionally, the cost is high as shown in Appendix C. Door-to-door notifications received a medium score for effectiveness of reaching all populations, a high score for effectiveness of communicating the warning message, and a medium score for cost. Door-to-door notifications, which are currently used as a supplemental method in Butte County, are likely to clearly get a warning across but divert personnel during the crisis to notifications (S. Collins, personal communication, November, 14, 2019). Community watch programs received a medium score for effectiveness of reaching all populations, a high score for effectiveness of communicating warning messages, and a high score for cost. This method is limited to no cost to the county to maintain due to its grassroots nature and has a high likelihood of effectiveness due to its reliance on already existing social networks. Cost-Benefit Analysis A cost-benefit analysis will be conducted on the policy options for warning message templates, wireless warning methods, and alternative warning techniques. A cost-benefit analysis is a tool that can help assess the economic feasibility of various policy options, which provides a determination on whether or how much the benefits of a particular alternative outweigh the associated costs (Patton et al., 2013). Some costs outlined can be explicitly monetized or include estimated monetary costs and others are designated as non-monetary costs, such as staff time or unknown future costs. Warning Message Templates Using Templates: The costs of using templates include the staff time used to craft the messages; however, these costs can likely be mitigated by using sample templates created by the State of California (2019) or using free trainings provided by OnSolve, the company behind CodeRED, with which Butte County has an existing contract (Butte County Grand Jury, 2019; OnSolve, 2019). The benefits of using templates outweigh the minimal costs and include improving the accuracy and consistency of warning messages by drafting them prior to the high- pressure environment of a wildfire or other disaster (State of California, 2019; Mileti & Sorenson, 1990). Not Using Templates: Not using pre-written templates provides minimal costs on the front end because staff time is not needed to develop or curate templates. However, potential costs exist with diverting staff time to drafting messages after the onset of a crisis as well as the potential steep costs associating with repairing public trust or assuaging public confusion if inconsistent or inaccurate messages are sent out. 9 Wireless Warning Methods CodeRED/privately operated emergency notifications: contract with OnSolve for the CodeRED software is $37,500 annually with an additional cost of $900 per year for a toll-free phone number that can host pre- recorded messages for the public to supplement 160-character wireless emergency notifications (Butte County Board of Supervisors, 2017). Benefits that balance these costs include the ability to send wireless emergency notifications and provide expanded recorded messages to residents, though limited to those who sign up due to opt-in nature of software (Butte County Grand Jury, 2019). IPAWS WEA: The costs of using IPAWS are approximately the same as CodeRED (subtracting the additional $900 annually) because the CodeRED software through OnSolve is an approved developer by FEMA (FEMA, 2019a). The benefits of IPAWS ability to reach all cellphone users in the targeted geographic region without prior registration far outweigh the costs. However, limitations to the benefits exist because of the 90-character limit and technical difficulties experienced by emergency officials in the past (S. Collins, personal communication, November 14, 2019). Both CodeRED and IPAWS WEA: Using both CodeRED and IPAWS WEA produce negligible cost differences when compared with using the two options separately since the same software is used to disseminate both wireless alerts (S. Collins, personal communication, November 14, 2019; FEMA, 2019a). Benefits of using both can include capturing a wider group of people due to redundancies. Alternative Warning Methods Patrol Car Sirens: vehicles was approximately $3,000 (Mensch, 2019). Additional costs are associated with this option for an outreach campaign to educate the public about the sirens. Benefits, if outreach is effective, include a recognizable and fast warning option. Permanent Siren System: Costs of a permanent siren system installed throughout Butte County can be estimated to be approximately $800,000 for 21 sirens and an additional $100,000 for a public outreach campaign (Berkeley Disaster and Fire Safety Commission, 2019). This is a conservative cost estimate based on one provided to the City of Berkeley, which is much smaller in area than Butte County, so the costs may be higher if additional sirens are required. Benefits of a permanent siren system include permanency and ability to be installed in neighborhoods that may be difficult to reach via patrol vehicle. However, some negative consequences associated with noise level of siren system and effectiveness is disputed (Berkeley Disaster and Fire Safety Commission, 2019). Door-to-Door Notifications: The cost of door-to-door notifications delivered by emergency personnel consist of staff time, specifically the use of personnel resources that may be needed in other areas during crises. Benefits include clear messaging from interpersonal communications. Community Watch Programs: There are limited costs associated with community watch programs due to the grassroots nature of the method. Benefits of such methods are great due to the use of existing networks and social ties. 10 Recommendations As part of its alert and warning system, Butte County should use pre-scripted warning message templates to ensure clear communication of protective warnings to the public during a disaster. The use of templates follows emergency management best practices and can help county staff avoid being rushed as warnings are assembled during future disasters, as was the case during the Camp Fire. Since the Camp Fire, templates have been drafted. Ensuring the messages are kept up-to-date and follow best practices should be of minimal cost to the county and can be implemented immediately. The Butte County Board of Supervisors should: Direct county staff to have warning message templates reviewed by California Office of Emergency Services officials to ensure they meet FEMA and emergency management best practices. Require state-mandated After Action Reports, which following local emergencies, include an assessment of the effectiveness of warning message templates, allowing for evaluation and adjustment as needed. individuals with cellphones in Butte County with a unique and recognizable alert tone during a disaster regardless of cellular congestion issues. However, the 90-character maximum for wireless messages sent through IPAWS presents a significant challenge for adequately communicating vital protective instructions and warnings to the public. Thus, it is recommended that Butte County use both Wireless Emergency Alerts and its CodeRED notification system. The Butte County Board of Supervisors should: detail in its draft Alert and Warning Plan quarterly exercises to test the IPAWS and CodeRED notification systems. Direct the Butte County Alert and Warning Group to develop a formalized public outreach campaign to increase registration for the CodeRED opt-in notification system. Campaign may require additional funding to be allocated by the Board. A widely agreed upon principle in emergency warnings is redundancy. Because no one method can currently reach all populations effectively, it is recommended that Butte County use a combination of warning methods to supplement the wireless warning methods. Currently, the Plan stipulates the use of IPAWS and CodeRED notifications, conventional media, social media, high frequency radio, local radio stations, changeable message signs, mobile public address systems and vehicle sirens, and in- person door-to-door notifications ( With a substantial public education campaign, a permanent siren system installed throughout the However, successful implementation will be unlikely without significant financial support from the state or federal government, so such a system is an optional recommendation for future exploration. The new two-tone vehicle sirens that were recently installed in patrol vehicles must be accompanied with a public education campaign. It is also recommended that the county explore the establishment of a formalized community watch program specifically geared toward emergency alerts and warnings. 11 The Butte County Board of Supervisors should: Direct the Butte County Alert and Warning Group to develop a comprehensive public education campaign to inform residents about the evacuation warning vehicle sirens, including but not limited to social media advertisements, community events, and physical signage. This effort must involve key stakeholders including representatives from communities with access and functional needs to ensure siren messages can be adequately communicated or that necessary accommodations can be identified. Direct the Butte County Alert and Warning Group to begin working with residents in each county evacuation zone to establish community watch groups with liaisons who are contacted in the event of a disaster. This recommendation may need to be implemented in a phased approach, beginning with the establishment of a public subcommittee of residents that works with the alert and warning group to assess unique neighborhood needs and feasibility. Implementation of these recommendations will be contingent on available Butte County resources as well as any additional funding provided by the state as a result of the recent state season in 2020, the Butte County Board of Supervisors review, approve, finalize, and publish the ; ensure review of warning message templates for adherence to best practices; require wireless warning exercise tests with software; and establish public education and outreach campaigns to increase resident registration in opt-in emergency notification system and ensure understanding of new vehicle sirens. Attachments Appendix A: CAM Analysis of Warning Message Templates Appendix B: CAM Analysis of Wireless Warning Methods Appendix C: CAM Analysis of Alternative Warning Methods 12 References Anderson, W. (1969). Disaster Warning and Communication Processes in Two Communities. Journal of Communication, 19(2), 92-104. https://doi.org/10.1111/j.1460- 2466.1969.tb00834.x Ashton, A. (2018). Brown swings back at Trump: Climate change is governor says. Sacramento Bee. Retrieved from https://www.sacbee.com/latest- news/article221518685.html Bean, H., Sutton, J., Liu, B. F., Madden, S., Wood, M. M., & Mileti, D. S. (2015). The Study of Mobile Public Warning Messages: A Research Review and Agenda. Review of Communication, 15(1), 6080. https://doi- org.mimas.calstatela.edu/10.1080/15358593.2015.1014402 Berkeley Disaster and Fire Safety Commission. (2019). Commission Referral: Recommendation to Install an Outdoor Public Warning System (Sirens) and Incorporate It Into a Holistic Emergency Alerting Plan. Retrieved from https://www.cityofberkeley.info/Clerk/City_Council/2019/09_Sep/Documents/2019-09- 10_Item_33_Commission_Referral_Recommendation.aspx Butte County Board of Supervisors. (2017). Agenda Transmittal: Contract with Onsolve LLC For Emergency Mass Notification System. Retrieved from http://buttecounty.granicus.com/MetaViewer.php?view_id=2&clip_id=411&meta_id=71 487 Butte County Grand Jury. (2019). Annual Report 2018-2019. Retrieved from https://www.buttecourt.ca.gov/GrandJury/reports/2018- 2019%20Grand%20Jury%20Report.pdf 2019). Retrieved from S. Collins. California Board of Forestry and Fire Protection. (2019). Responding to Wildfires in the Era of Retrieved from http://californiasaf.org/wp- content/uploads/2019/03/Responding-to-Wildfires-in-the-Era-of-the-New-Abnormal.pdf Cal. Legis. Joint Legislative Committee on Emergency Management, & Assembly Committee on Communications and Conveyance. (2019) Joint Informational Hearing Background Paper. Emergency Alerts and Evacuations Lives Saved or Lives Lost?. Retrieved from https://acom.assembly.ca.gov/sites/acom.assembly.ca.gov/files/Background%20Paper%2 0-%20Emergency%20Alerts%20and%20Evacuation.final_.pdf California Department of Forestry and Fire Protection. (2018). Camp Fire. Last updated Nov. 15, 2019. Retrieved from https://www.fire.ca.gov/incidents/2018/11/8/camp-fire/ 13 California Department of Forestry and Fire Protection. (2019). Top 20 Most Destructive California Wildfires. Retrieved from https://www.fire.ca.gov/media/5511/top20_destruction.pdf California Office of Emergency Services. (2018). Public Alert and Warning Program Assessment for Sonoma County. Retrieved from https://sonomacounty.ca.gov/WorkArea/DownloadAsset.aspx?id=2147555230 FEMA. (2018). An Introduction to IPAWS. Retrieved from https://www.fema.gov/media-library- data/1543325778432- 9b4f87f4e3621e225c7b3f53e9c896a9/2018_An_Introduction_to_IPAWS_final.pdf FEMA. (2019a). Alert origination software providers that have successfully demonstrated their IPAWS capabilities. Retrieved from https://www.fema.gov/media-library- data/1558451891984-2c0dd8fa4474f0ff4c4dce730b6619ed/1.pdf FEMA. (2019b). Frequently Asked Questions: Wireless Emergency Alerts. Retrieved from https://www.fema.gov/frequently-asked-questions-wireless-emergency-alerts FEMA National Advisory Council. (2019). System. Retrieved from https://www.fema.gov/media-library-data/1550587427456- 30d4179ee4fa8b97ecf4ab6bee76ace6/NAC_IPAWS_Subcommittee_Final_Report.pdf Howle, E. (2019). California Is Not Adequately Prepared to Protect Its Most Vulnerable Residents From Natural Disasters. Emergency Planning Report Number: 2019-103. California State Auditor. Retrieved from http://auditor.ca.gov/reports/2019- 103/index.html Kuligowski, E. D., & Wakeman, K. (2017). Outdoor Siren Systems: A review of technology, usage, and public response during emergencies. National Institute of Standards and Technology. Technical Note 1950. Retrieved from https://nvlpubs.nist.gov/nistpubs/TechnicalNotes/NIST.TN.1950.pdf Li, D. Cova, T. J., Dennison, P. E., Wan, N. Nguyen, Q., Siebeneck, L. K. (2019). Why do we need a national address point database to improve wildfire public safety in the U.S.? International Journal of Disaster Risk Reduction, 39, 1-11. https://doi.org/10.1016/j.ijdrr.2019.101237 McCaffrey, S., Velez, A., & Briefel, J. (2013). Differences in Information Needs for Wildfire Evacuees and Non-Evacuees. International Journal of Mass Emergencies and Disasters, 31(1), 1-3. Retrieved from https://www.fs.fed.us/nrs/pubs/jrnl/2013/nrs_2013_mccaffrey_001.pdf Mensch, J. (2019). In Case of Sirens. Fox 40. Retrieved from https://fox40.com/2019/07/23/in-case-of-wildfire-butte- county-sheriffs-vehicles-now-have-evacuation-sirens/ 14 Mileti, D. S., & Peek, L. (2000). The social psychology of public response to warnings of a nuclear power plant accident. Journal of Hazardous Materials, 75, 181-194. Retrieved from http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.463.2441&rep=rep1&type=pdf Newcomer, K. E., Hatry, H. P., & Wholey, J. S. (2010). Handbook of Practical Program Evaluation (3rd ed., p. 446). San Francisco: Jossey-Bass. OnSolve. (2019). Online Product Training. OnSolve. Retrieved from https://www.onsolve.com/training/ rd Patton, C., Sawicki, D. & Clark, J. (2013). Basic Methods of Policy Analysis and Planning, 3 ed. New York, NY: Routledge. S. Bill No. 833, California Legislature, 2017-2018. Retrieved from http://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=201720180SB833 Schoonover, S. (2018). Butte County budget balanced, but with job cuts. Chico Enterprise- Record. Retrieved from https://www.chicoer.com/2018/06/10/county-budget-balanced- but-with-dozens-of-job-cuts/ St. John, P., & Serna, J. (2018). Camp fire evacuation warnings failed to reach more than a third of residents meant to receive calls. Los Angeles Times. Retrieved from https://www.latimes.com/local/california/la-me-ln-paradise-evacuation-warnings- 20181130-story.html and became a deathtrap. Los Angeles Times. Retrieved from https://www.latimes.com/local/california/la-me-camp-fire-deathtrap-20181230-story.html State of California. (2019). Alert & Warning Guidelines. Retrieved from http://calalerts.org/document Working Group on Natural Disaster Information Systems Subcommittee on Natural Disaster Reduction. (2000). Effective Disaster Warnings. National Science and Technology Council Committee on Environment and Natural Resources. Retrieved from https://digital.library.unt.edu/ark:/67531/metadc25972/m2/1/high_res_d/EffectiveDisaste rWarnings2000.pdf 15 Appendix A: CAM Analysis of Warning Message Templates Total Efficiency Effectiveness Cost Score 2 3 1 Write Hinders speed of Variable due to quick Low cost on front-end; messages disseminating turnaround; accuracy however, potential after warnings; staff may be inconsistent; costs associated with wildfire resources and time could lead to correcting inaccurate onset, as used to draft and get confusion with public messages if sent out incident approval for and distrust develops messages in their entirety Weights Raw Weighted Raw Weighted Raw Weighted 1 2 1 3 2 2 7 Pre-write Reduces time spent More time to vet Some costs due to staff warning during incident; more messages; ensures time used to craft message time and resources consistency and messages; however, templates can be dedicated to accuracy free trainings provided for a other purposes via OnSolve/CodeRED variety of and examples readily scenarios available from state ahead of and other localities incident Weights Raw Weighted Raw Weighted Raw Weighted 3 6 3 9 2 2 17 16 Appendix B: CAM Analysis of Wireless Warning Methods Total Effectiveness for Effectiveness for Cost Score reaching all communicating 1 populations message 3 3 CodeRED or Opt-in nature 160-character limit $37,500 annually other reduces for content (more for OnSolve privately effectiveness, with than IPAWS) allows software; additional operated low percentage of for additional $900 per year for emergency public registered; information and phone number with notifications unable to capture clarity; expanded pre-recorded out-of-town feature includes toll messages visitors; reliance free number to call on cellular for additional infrastructure and information impacted by jammed networks; ability to geotarget Weights Raw Weighted Raw Weighted Raw Weighted 1 3 3 9 2 2 14 IPAWS WEA Opt-out system; 90-character limit $37,500 annually system captures residents hinders message for OnSolve and out-of-town content; gets software (used for visitors; ability to attention with loud both CodeRED and geotarget; relies on alert tone; past IPAWS) cellular failure due to infrastructure but unresolved IT not affected by problems network congestion Weights Raw Weighted Raw Weighted Raw Weighted 3 9 2 6 2 2 17 17 Appendix C: Alternative Warning Methods Effectiveness for Effectiveness for Cost Total reaching all communicating 1 Scores populations message 3 3 Patrol car Does not rely on Variable due to need $3,000 for siren sirens cellular technology; for education and installation; plus could be limited by outreach to be additional cost for hard to reach effective outreach and regions education of public and training of officers Weights Raw Weighted Raw Weighted Raw Weighted 2 6 2 6 2 2 14 Permanent Siren System Does not rely on Relies on outreach Estimates at $800,000 cellular technology; and education for for 21 sirens can be stationed in efficacy; hard to (conservative estimate hard to reach educate out-of-town due to possibility of neighborhoods and visitors may have more sirens); plus throughout county unintended negative estimated $100,000 for externalities due to public outreach noise level campaign Weights Raw Weighted Raw Weighted Raw Weighted 2 6 2 6 1 1 13 Door-to-Does not rely on Person-to-person Cost of staff time Door cellular technology; message provides during wildfire Notifications could be limited by limited room for hard to reach areas misinterpretation or fast spreading wildfire; reliant on staff resources Weights Raw Weighted Raw Weighted Raw Weighted 2 6 3 9 2 2 17 Community Success relies on Likely to be highly Limited costs due to Watch widespread use; effective once put in grassroots nature of Programs would not capture place due to use of method out-of-town visitors strong informal and friendship networks and social ties Weights Raw Weighted Raw Weighted Raw Weighted 2 6 3 9 3 9 24 18