1. Introduction
Since 1980, extreme precipitation events have caused an average of $4 billion in damage per year in the United States (NCEI 2020). These events impact public health (e.g., Exum et al. 2018) and sectors of the economy, including transportation (e.g., Suarez et al. 2005), agriculture (e.g., Rosenzweig et al. 2002), and tourism (e.g., Craig and Feng 2018; Meseguer-Ruiz et al. 2021). Floodplain managers make decisions during daily operations related to development and infrastructure in their jurisdictions and are uniquely situated to be impacted by extreme precipitation events. Floodplain managers are responsible for understanding physical processes related to flooding, managing anthropogenic changes in floodplains, developing policies and plans for their jurisdictions, and overseeing infrastructure design and construction (Association of State Floodplain Managers 2010).
The decisions floodplain managers make depend on meteorological forecasts because, in the short term, these forecasts can provide advance notice of extreme precipitation events that may impact, for example, daily road-paving activities. In the long term, projections indicate what conditions may be like in warmer climates. As meteorological predictions become more skillful, especially at longer lead times such as those associated with the subseasonal or seasonal time scales (from about 2 weeks to 3 months), floodplain managers may be given more advanced notice of impending precipitation events, empowering them to take action ahead of time and prepare their communities.
There are multiple types of predictions available related to extreme precipitation events. Short- to long-range weather forecasts generally are any forecast with a lead time of less than 10 days and are produced by both National Weather Service (NWS) local offices and national forecast centers. These forecasts provide specific information about precipitation amount and timing, as well as impacts across the country, and can be either deterministic (i.e., exact conditions at a specific time) or probabilistic (i.e., probability of an event occurring at a specific time or range of times). Forecasts for slightly longer lead times, usually between 10 and 30 days, are considered subseasonal (S2S) forecasts. In the United States, NOAA’s Climate Prediction Center (CPC) produces subseasonal forecasts that communicate which portions of the country may experience above- or below-average temperatures and precipitation. Seasonal forecasts, also produced by the CPC, have a lead time of 0.5–12.5 months and are solely probabilistic. These products communicate how likely it is for a given area to experience conditions that differ from its normal climate.
Given the variety of weather and climate information available for decision-making, there are many potential applications to water and floodplain management. In 2013, the U.S. Army Corps of Engineers compiled a report that details different types of information water managers use for making decisions. On the short-term time scale, this information includes river gauge observations or precipitation predictions. It was common for water managers to obtain this information from observational networks such as state mesonets or the Snowpack Telemetry (SNOTEL), which is used to monitor snowpack, precipitation, and temperature (Schaefer and Paetzold 2001). Information on the subseasonal time scale was used less frequently, although water managers did use quantitative precipitation forecasts from the Weather Prediction Center. On longer time scales, such as those associated with seasonal forecasts, water managers used the U.S. Drought Monitor or outlooks produced by the CPC to make decisions regarding long-term water supply. Overall, forecasts used by water managers covered a range of time scales and were used for decisions such as responding to flooding or planning for long-term water availability in their jurisdictions.
Even though there is a wealth of weather and climate information available to inform decisions, many barriers exist to using these forecasts. For example, practitioners may be unsure of where to get the exact information they need (e.g., Carbone and Dow 2005; Rayner et al. 2005; Bolson et al. 2013; Bruno Soares and Dessai 2016). When practitioners can access the information they need, it is not always easy to understand (e.g., Pagano et al. 2001; Coelho and Costa 2010; Taylor et al. 2015). Potential users also can view forecasts as not trustworthy enough to be used for decision-making (e.g., Carbone and Dow 2005; Bruno Soares and Dessai 2016; Crochemore et al. 2021).
In addition to problems accessing and understanding forecasts, floodplain managers often work in institutional systems that add complexity to their decision-making. One challenge is the fragmentation associated with water and floodplain management, because water is often overseen by multiple agencies or departments within a jurisdiction (Rayner et al. 2005). Additionally, managing floodplains is often one component of someone’s job, and they will have other responsibilities to balance. Another challenge in floodplain management is a lack of flexibility in policies and procedures (Steinemann 2006) or reactive policy changes that arrive after a community has experienced a disaster (Pagano et al. 2001). Last, budgets and financial resources in a jurisdiction significantly limit which floodplain management strategies are implemented and how likely a community is to invest in long-term protective actions (e.g., Ramos et al. 2013; Arnal et al. 2016). Because of the multiple challenges that a floodplain manager may experience in their position, it can be difficult for them to act on weather and climate information, even if they are able to access it and interpret it for their jurisdiction.
The purpose of this study was to understand how floodplain managers in Oklahoma currently use weather and climate information to make decisions, especially because floodplain managers and their use of this information are understudied in the scientific literature. Building on an existing decision model, we developed and addressed three research questions: 1) What forecast information do floodplain managers use to plan for extreme precipitation events and where do they obtain this information from? 2) How do different forecasts influence the floodplain manager’s decision-making timeline? 3) What types of forecast information do floodplain managers wish they had when making decisions in their jurisdictions?
2. Methods
a. Context
In 2018, the Prediction of Rainfall Extremes at Subseasonal to Seasonal Periods (PRES2iP) project hosted a 3-day workshop at the University of Oklahoma with water and emergency managers and tribal environmental professionals from around the country, as described in VanBuskirk et al. (2021). During this workshop, the PRES2iP team learned how different practitioners plan for and make decisions regarding extreme precipitation events. According to workshop participants, practitioners are likely to focus on short-range forecasts, even if long-range forecasts are available to them. Based on this response to extreme precipitation events, we investigated if other practitioners may be more likely to use forecasts on the subseasonal to seasonal time scale. We chose to focus on floodplain managers, as their job responsibilities may affect long-term development and planning in their jurisdictions (Association of State Floodplain Managers 2010) and are affected by extreme precipitation events. Floodplain managers are professionals dedicated to reducing flood losses and are relatively understudied in the literature on their use of weather and climate information for decision-making (Rasmussen et al. 2017). To understand how Oklahoma floodplain managers are making decisions using weather and climate information, we conducted interviews with eight floodplain managers across the state.
Floodplain managers across Oklahoma deal with a climate that varies widely, both spatially and temporally. Western Oklahoma receives an average of 508–762 mm of precipitation annually, whereas eastern Oklahoma receives 1016–1270 mm of precipitation per year (Oklahoma Climatological Survey 2011). Within each of the nine climate divisions (not shown) and for the entire state (Fig. 1), interannual precipitation variability is large. Downscaled climate projections indicate that by midcentury (2036–65), the year’s largest 1- and 5-day precipitation events are expected to have higher precipitation totals than in the recent past for eastern Oklahoma and no change for western and central Oklahoma, regardless of representative concentration pathway (South Central CASC 2022).
Annual precipitation history [black diamonds; in. (1 in. = 2.54 cm)] for Oklahoma from 1895 to 2021 (Oklahoma Climate Survey 2021). Shading represents the 5-yr weighted average of the annual values as compared with the long-term average, green denotes above-average periods, and brown denotes below-average periods.
Citation: Weather, Climate, and Society 15, 3; 10.1175/WCAS-D-22-0080.1
Rivers in Oklahoma generally flow from northwest to southeast (Fig. 2), ultimately draining into the Mississippi River; all major lakes are reservoirs that are used for flood control and other purposes. Flash floods can occur anywhere but are most common in the eastern two-thirds of the state, including Oklahoma City and Tulsa. Causes of flash floods typically are back-building thunderstorms (e.g., Yussouf et al. 2016), stalled outflow boundaries or fronts (e.g., Dahl and Xue 2016), or remnants of tropical storms (e.g., Arndt et al. 2009). Extended periods of rainfall, sometimes associated with El Niño conditions, also have caused widespread river flooding (e.g., Wang et al. 2015), particularly in the Arkansas, Canadian, Cimmaron, Kiamichi, and Red Rivers.
Location of rivers, major streams, and operations of interview participants in Oklahoma. The base map was provided through the courtesy of the Oklahoma Water Resources Board.
Citation: Weather, Climate, and Society 15, 3; 10.1175/WCAS-D-22-0080.1
In Oklahoma, floodplain management is often only one area of responsibility that a government employee might have. In general, those overseeing floodplain management have other job duties, such as emergency management, administration of public works projects, or municipal or town planning (Siebeneck et al. 2022). Floodplain managers are challenged by the extensive knowledge required to do their job, a lack of clarity regarding what level of government (i.e., federal, tribal, state, local) is responsible for different aspects of the work, alterations in precipitation and land-use patterns resulting from climate change and urbanization or agricultural production, and other work demands in their position (Siebeneck et al. 2022).
b. Protective action decision model
For this study, we selected the Protective Action Decision Model (PADM; Fig. 3), developed by Lindell and Perry (2012), as our theoretical framework because it encompasses multiple factors accounting for decision-making, it aligned with concepts noted by floodplain managers in a prior workshop (VanBuskirk et al. 2021), and it has been applied in similar studies (Tyler et al. 2021). The PADM is conceptualized as an iterative process that begins with environmental and social cues (e.g., experiencing street flooding or observing the behavior of other people), as well as warnings (e.g., receipt of a weather forecast). The next step is the predecisional processes: exposure, attention, and comprehension (Lindell and Perry 2012). First, a floodplain manager must receive the cues or warnings, then decide to heed the information, and take protective action within the decision-making portion of the model. After the predecisional processes, someone taking a protective action will face the protective action decision-making portion of the model, which consists of five stages: risk identification, risk assessment, protective action search, protective action assessment, and protective action implementation. In theory, by the last stage, a decision-maker will have assessed the risk they face, decided among possible actions to take, and implemented one of those actions.
The Protective Action Decision Model, as described in Lindell and Perry (2012). Environmental and social cues (left section) are two types of information that can initiate protective action decision-making. Message recipients will face psychological processes (center section) to assess their level of risk to a given hazard and determine which action they should take to protect themselves. Then, they will respond to the risk assessment (right section) by seeking more information or heeding a warning message. The sequence repeats iteratively until a protective action is taken.
Citation: Weather, Climate, and Society 15, 3; 10.1175/WCAS-D-22-0080.1
This process of searching for a protective action will repeat, driven by the communication action assessment portion of the PADM. In many cases, people will search for more information to guide how they respond to a hazard. The types of information that floodplain managers search for vary widely and is not always related to weather forecasts. For hazards such as extreme precipitation events on the subseasonal or seasonal time scale, a lack of urgency about hazard impacts could lead floodplain managers to continue searching for information and wait for more certainty in the forecast, instead of acting ahead of time when long-term forecasts become available.
The PADM was developed to understand how individuals respond to environmental hazards and threats. Previous research has applied the PADM to multiple short-range hazard predictions, including tornado warning response (e.g., Ripberger et al. 2019; Sherman-Morris et al. 2020), flash flooding warning response (e.g., Companion and Chaiken 2017), or wildfire warning response (e.g., Kuligowski 2021; Santana et al. 2021). Although the model has historically been used to understand individual decision-making, it is beginning to be applied to the ways people respond to risks in other contexts (Liu et al. 2019), such as the types of hazards that S2S products are intended to predict (White et al. 2022). Additional studies have applied the PADM to flood risk management (Terpstra and Lindell 2013), or used the theory to study how floodplain managers make decisions for their community (Tyler et al. 2021).
In this study, we applied the main components of the PADM to understand how floodplain managers are receiving, understanding, and acting on weather or climate information to make decisions for their communities about environmental hazards. Thus, the PADM is a useful theoretical model for this study as it provides a framework for a decision-maker receiving forecast information, interpreting it, and then taking some form of protective action. Although floodplain managers make decisions on behalf of their communities, they are still individuals who are acting based on a variety of economic, institutional, and environmental factors. Other models, such as trust in forecast information (Shafiee-Jood et al. 2021), cost-loss econometric approaches (Nelson and Winter 1964), or the decision-analytic framework (Hilton 1981), do not consider some of the factors that have been found to influence floodplain managers’ decisions, such as prior experience with an extreme precipitation event (VanBuskirk et al. 2021). The PADM is a well-established theoretical model that describes how people make decisions about environmental hazards. Because it has been applied on multiple time scales and hazards, we felt it was appropriate to apply its core principles to an understudied population (floodplain managers) and an understudied time scale (subseasonal–seasonal).
c. Sampling
Our goal was to recruit participants from across Oklahoma to reflect the west-to-east gradient of precipitation. To do so, we used both purposive sampling (Tongco 2007), recruiting participants who were active floodplain managers, and snowball sampling (Goodman 1961), using recommendations from previously interviewed floodplain managers to make additional connections. Initially, we emailed county floodplain managers to recruit potential interview participants, using a list maintained by the state water agency. After exhausting this list of county floodplain managers, for a second round of recruitment we emailed county emergency managers to ask if they could connect us with floodplain managers in their community. Additional participants were recruited from a list of presenters at a recent Oklahoma Floodplain Managers Association conference and personal recommendations. Our goal was to interview 5–10 people, as recommended by the literature (e.g., Boddy 2016; Creswell and Poth 2017; Vasileiou et al. 2018) for qualitatively assessing phenomena—in this case the decision-making process of floodplain managers; ultimately, we conducted eight interviews (Fig. 2).
d. Interview guide
We designed our interview guide—composed of questions to directly answer our research questions—on the basis of findings from the PRES2iP Research Priorities workshop (VanBuskirk et al. 2021) and the PADM (Lindell and Perry 2012). For instance, at the stakeholder workshop described above, participants indicated they used a variety of weather and climate information to make decisions, so interview questions asked participants to state what types of information they use [Table 1, research question 1 (RQ1)]. In addition, the PADM addresses how recipients utilize forecasts to make decisions in response to natural hazards, so we wrote questions to understand how floodplain managers use and act on different weather and climate information (Table 1, RQ2 and RQ3). Our goal was to understand what information floodplain managers use to make decisions and how they come to those decisions. Table 1 displays the research questions.
Research questions (RQ) and associated questions from interview guide.
e. Data collection and sample description
Prior to contacting potential participants, our study was reviewed and approved by the University of Oklahoma’s Institutional Review Board (study 13548). To recruit participants, we emailed individuals from the sources listed earlier, describing the purpose of the study, commitment required, timeline for scheduling interviews, and required paperwork to be completed. Interviews were semistructured, leaving room for follow-up questions to participants’ responses as needed. All interviews took place via Zoom software and were transcribed afterward. Then, participants were given an opportunity to review the transcript of their interview and clarify any statements. We did not compensate the participants for their time.
f. Analysis
The first author analyzed the responses using thematic analysis, first developing codes, then creating code categories, and synthesizing codes into themes about the data (Fig. 4). First, they read each transcript, coding text relevant to our research questions and developing codes as they read, a process known as open coding (Braun and Clarke 2006). To enhance trustworthiness while coding (e.g., Elliott et al. 1999), the first author kept a notebook to document their coding process, following suggestions in the literature (e.g., Maxwell 2012). The first round of coding was conducted on the basis of the content of the interviews; statements about different topics were coded together. This process resulted in 35 codes of two types: characteristics of the data and descriptive statements. Characteristics of the data included 1) types of information that floodplain managers used, 2) sources of weather and climate information, and 3) types of decisions being made. Descriptive statements included 1) people explaining their role and experience in floodplain management and 2) challenges floodplain managers faced in their jobs.
The coding and thematic analysis process. Interview quotations were categorized into codes describing different characteristics of the data, such as statements about budgets, challenges managers faced in their positions, or specific information sources such as NWS email or professional colleagues. These codes were then grouped into six categories on the basis of shared characteristics, and the categories were then collapsed into five themes describing research results.
Citation: Weather, Climate, and Society 15, 3; 10.1175/WCAS-D-22-0080.1
After initial coding, the first author completed a second round of coding to combine the 35 codes into six different categories: types of weather or climate information, sources of weather or climate information, practitioner partners, types of decisions, decision-making factors, and desired information. All of the authors then inspected and analyzed these categories in light of our research questions and theory, collapsing them into five themes (Saldaña 2009). These themes described findings to address our research questions and added new insight from the interviews regarding the PADM.
3. Findings
We completed eight interviews with floodplain managers (hereinafter participants) who represented decision-makers at the city, county, and state levels (Fig. 2). We interviewed participants in two main regions: the Oklahoma City metropolitan area and eastern Oklahoma, both of which experience more precipitation and flooding than western Oklahoma. All interviewees were current floodplain managers, though specific job titles varied across the sample. Five participants had the title “floodplain manager,” one had the title “consultant,” one had the title “emergency manager,” and one’s title was “engineer.”
At the county level as far as climate change goes, that is probably a newer topic. You know, it’s not something we’ve really addressed as of yet. I think it’s something that the more and more it’s talked about, it’s something that we’re going to have to address at some point. I don’t know how as the county we can do that just because the county we’re a lot more, I’d say financially constrained than cities.
Our main analysis that addressed our research questions resulted in five themes: 1) types of decisions: types of decisions that floodplain managers make; 2) abundance of information: types and sources of information used by floodplain managers; 3) social network: people involved in the decision-making process; 4) outside factors: additional factors that influence decisions; and 5) informational needs: desire to use forecast information that may not be available yet. Table 2 summarizes the themes and codes.
Summary of themes, their descriptions, and associated codes from thematic analysis of participant interviews.
a. Theme 1: Types of decisions
Throughout the interview process, each person described different types of decisions for which they are responsible. The following section and the first theme from our analysis describe these types of decisions and the time scales on which they occur. These results are described below and appear prior to results regarding specific research questions, to situate the reader and contextualize later results.
Decisions occurred across a variety of time scales, ranging from responding to events in real time to long-term planning for a jurisdiction. In the short term, decisions often were being made as a response to the onset of precipitation or the expectation of a precipitation event within the next two to three days. Most frequently, participants were responsible for planning construction or road-paving activities that could be impacted by any type of precipitation. For example, they might have to decide how far to pave a road on one day such that they could complete both sides of the road if precipitation was expected the next day. Additionally, weather forecasts could be used to prepare job sites ahead of precipitation, ensuring topsoil was protected or clearing storm drains. Weather forecasts also were used to prepare within a community, by staging barricades for closing roads or notifying residents they may need to evacuate depending on the intensity of impacts.
Participants were less likely to make decisions on the subseasonal to seasonal time scale. Many were not making those types of decisions at all, rather only focusing on much longer time scales, such as seasonal predictions. When participants were making decisions within these longer time scales, they often did not consult different predictions or projections for their jurisdictions. Instead, their decisions were shaped and guided by experiential knowledge they had about rainy or dry seasons in their area. For those who did make subseasonal to seasonal decisions, they mentioned scheduling more staff during the time most impacts were expected or preparing resources to respond to flooding. Others described how they may send crews to clear ditches and remove debris around creeks or storm drains to alleviate any potential flooding. These activities depended on the availability of employees and what other projects were in progress. On the seasonal time scale, floodplain managers mentioned planning construction projects or other large-scale developments around the spring and fall rains that they commonly experienced each year.
Floodplain managers also frequently made decisions for their communities related to long-term planning and development. These decisions included preparing for facility maintenance or monitoring and maintaining canals within a jurisdiction, making repairs as needed. Other decisions were related to addressing flooding issues in localized areas, with some floodplain managers seeking funding from the federal government to relocate neighborhoods that flooded often. Others addressed persistent roadway erosion due to rivers and streams in their jurisdictions and made plans to move or redesign roads as needed. Some floodplain managers designed and planned the construction of new infrastructure and bridges throughout the state. Finally, floodplain managers also worked with others in their jurisdictions to develop multiple types of long-term plans related to community development. These documents included master drainage plans, rezoning and development plans, stormwater plans, and hazard mitigation plans.
Overall, floodplain managers were responsible for decisions across multiple time scales in their jurisdictions. These time scales could range from responding to precipitation events as they are happening or preparing for the long-term impacts of extreme precipitation events. Decisions, particularly those on longer time scales, were influenced by multiple factors, including types of weather and climate information, shared information within a floodplain manager’s social network, or other factors such as a community’s budget, all of which will be discussed in depth below.
1) Research question 1: What forecast information do floodplain managers use to plan for extreme precipitation events and where do they obtain this information from?
(i) Theme 2: Abundance of information
Floodplain managers used a variety of weather and climate information from multiple sources to make decisions in their jobs regarding all types of hazards, not only extreme precipitation events. Information ranged from real-time observations and short-term weather forecasts to seasonal forecasts. These observations and forecasts covered multiple atmospheric variables, such as temperature, wind, and precipitation. Specific information included precipitation and temperature forecasts, radar and satellite products, atmospheric observations from weather balloons, and even historical precipitation observations and stream gauge data.
Floodplain managers received weather and climate information from multiple sources. Information came from the Oklahoma Mesonetwork (“Mesonet”; a network of surface observation stations across the state); broadcast meteorologists; email blasts from federal and state agencies, such as the Federal Emergency Management Agency (FEMA) or the U.S. Geological Survey (USGS). Another popular source was communication with NWS forecasters via NWSChat (a one-to-one or one-to-many messaging tool that is used and mediated by the NWS to communicate during high-impact weather events). Participants with emergency management experience were more likely to use more technical information (e.g., satellite data or atmospheric conditions) than those without emergency management experience.
(ii) Theme 3: Social network
He’s [emergency manager in a nearby city] pretty much a weather geek. So, he’s a really great resource because he stays on top of all of this, and I like to just get his opinion on what he thinks is going to happen. I’ll make my own decision, but I’ll certainly seek input from those people whose judgement I’ve learned to trust.
In this instance, both decision-makers had access to the same forecast information, but their discussions could reinforce how each person was interpreting the forecast and preparations could be made accordingly.
NWS forecasters were also crucial sources of information for floodplain managers, especially in the few days leading to an event. Some floodplain managers had close relationships with specific forecasters at their local weather service office and would call the office ahead of an event to discuss forecast details. Others accessed forecasts on their own from the NWS website. Overall, it was clear that the NWS was a key source of trusted forecast information that floodplain managers utilized.
(iii) Theme 4: Outside factors
Prior experience with a flood also acted as a source of information. Every floodplain manager who was interviewed had experience with flooding in their community that either informed later decisions to change policies or affirmed that previous changes to policy were beneficial. For example, one manager described how a flood in 2016 had led their community to develop a canal-wall mitigation plan to monitor whether canals needed maintenance. When another flood impacted the community in 2019, this plan had helped identify possible failure points in the canals and maintenance was completed ahead of time so the community had no canal failures in the second flood. A different floodplain manager described how a series of floods proved that home buyouts in part of their jurisdiction were helpful because homes that previously had flooded routinely were moved out of the floodplain. By taking previous experiences and incorporating them into new policies or practices, floodplain managers were able to be better prepared for the next flood that impacted them.
2) Research question 2: How do different forecasts influence the decision-making timeline of a given floodplain manager?
(i) Theme 2: Abundance of information
Floodplain managers often used information in real time to respond to impacts their community was experiencing. For example, one floodplain manager stated, “I’m looking and seeing what the radar predictions are from the news channels and the Weather Channel and then I’m watching real-time radar as it’s happening. I’m using real-time data.” During an event, real-time observations helped guide emergency response and where additional resources would be needed.
Besides radar and satellite products, another source of these real-time observations was the Oklahoma Mesonet (McPherson et al. 2007), which was frequently mentioned by interview participants. Floodplain managers used the Mesonet to assess impacts in neighboring communities and then applied their localized knowledge to determine what impacts their jurisdiction could face. One participant stated, “I think the Mesonet is great too, because even on flooding events, if I look at the sites that are in the South Canadian Basin, I can get a pretty good idea of if we’re going to have a flood here on the South Canadian River.” Mesonet data also were used after precipitation events occurred to develop reports and quantify impacts, or even to incorporate into regression formulas used for planning and bridge design.
The rain, it influenced us by determining our paving schedule. It determined how far we went one day so that we could at least finish the other side of the road in case the rain came that afternoon. If the rain did come, it halted our work, but we had planned and changed the length of our half of the road paving based on the predicted rainfall.
At longer time scales, floodplain managers discussed seasonal forecasts that gave them an idea of conditions for a few months from now, but they also mentioned these forecasts lacked the detail required to act on them. Instead, they were used “as a guide to help us pay attention to what might be coming.” Floodplain managers also knew typical precipitation patterns associated with the seasons. One stated, “I try to look out by the seasons. We’ve got the spring rains, we’ve got the fall rains, that dictates our construction season.” This climatological knowledge about when precipitation was more likely did influence when projects were initialized. In general, weather forecasts were used more often to make specific decisions, and seasonal forecasts were used as an alert for possible future conditions.
Some floodplain managers were dealing with decision-making and planning on much longer time scales, handling recurring flooding problems that lasted decades or longer in their communities. While these floodplain managers did not use specific climate projections to assess impacts their jurisdictions might face in the future, they knew which areas of their jurisdiction were susceptible to routine flooding and thus took steps to alleviate the impacts. These actions included seeking FEMA funding to help residents move out of floodplains, redesigning infrastructure to handle more intense precipitation events, or developing new stormwater or drainage plans. For example, one floodplain manager noted, “Another thing that the city did, we developed a master drainage plan. I believe it took 6 years and it identified several hundred stormwater projects within the city and came up with a project list that was able to prioritize them.” These long-term projects helped determine priority areas for floodplain managers and often aimed to make communities more resilient to flooding impacts.
However, while weather and climate information is a key component of decision-making, it is only one factor that shapes how floodplain managers make decisions. Social networks and other outside factors, such as a jurisdiction’s budget, the responsibilities and authority a given floodplain manager has, the desires of county commissioners, and public awareness about flooding issues, also influence how floodplain managers are making decisions in their communities.
(ii) Theme 3: Social network
Multiple people in each jurisdiction have influence over issues related to floodplain management. Floodplain managers frequently discussed collaborating with their social network, including stormwater managers, county or city engineers, the public works department, emergency managers, or the transportation department. As one floodplain manager said, “When I’m seeing things that need to be done, I talk to the mayor, the director of public works, our city planner, our code inspector, maybe our emergency manager, and our street superintendent. Every one of those people are involved in some way with stormwater or floodplain management from an administrative or technical point of view.” Floodplain managers are not acting in isolation when it comes to floodplain issues. Instead, they work with colleagues and other professionals to develop plans and solutions to problems that arise.
And that’s the problem with floodplain management and stormwater management. When there’s an event, everybody says, right away, ‘let’s fix this stormwater problem.’ But two weeks later, all they care about is the street in front of them because it’s got a pothole. So, my business and everybody that’s in my business, we’re always struggling to help educate the people on why we are doing this.
Keeping public attention on flooding for a long period of time can be difficult, but many floodplain managers noted that recent floods enable progress with new facilities or projects because members of the public are reminded of the impacts of flooding. Despite this, it can still be difficult to sustain passion for change, especially when residents realize the cost of fixing stormwater or floodplain issues.
(iii) Theme 4: Outside factors
The problem is that it all takes money. Everything comes back to money. And they [the county commissioners] can say ‘We don’t have the money to do that. We understand what you’re saying, we know you can come up with a technical solution. But that solution costs money, and where is that money going to come from?’
Identifying funding sources and applying for external grants were large obstacles that participants must overcome.
I’m very limited on my authority. Not only in the field, but when it comes to city personnel too. I cannot dictate and tell crews what to do. I can make suggestions and run it up the chain of command. And for the most part, everyone is on board because no one wants to see flood damage.
The added layer of complexity associated with running decisions by others with more authority was a common point of frustration for participants, especially when they felt capable of making those decisions themselves.
At the end of the day, it’s up to the commissioners. Pretty much everything we do has to go to the Board of County Commissioners. So, it’s public meetings, and they vote on topics. At the end of the day, it’s their decision of either making it happen or not.
Overall, even if a floodplain manager was able to use weather or climate information and their professional network to come to a decision, whether that decision moved forward was reliant on commissioners’ ultimate approval.
Other challenges were mentioned as well, though less frequently than those related to budgets or authority. For example, one participant had multiple responsibilities outside of floodplain management, such as hiring new personnel, dealing with insurance claims, or overseeing training for employees. Others mentioned how hard it could be to navigate the rules and regulations surrounding development within their communities and at the state level. Local politics associated with funding projects and resource allocation were also challenges, especially for those working at the county level. Some interview participants mentioned that in their communities, stormwater and floodplain management are afterthoughts to their superiors in the decision-making hierarchy, so it could be difficult to make progress on projects.
Overall, floodplain managers were not solely relying on weather or climate information to make decisions. Instead, they were working within complex social networks with multiple colleagues and practitioners involved in developing and seeing a project through. County commissioners could dictate which projects are funded—if money were available at all for floodplain-related issues. Even if the best weather and climate information were available, floodplain managers may not be able to use it to develop solutions to problems they face, because of these other outside factors.
3) Research question 3: What types of forecast information do floodplain managers wish they had when making decisions in their jurisdictions?
Theme 5: Informational needs
While there are multiple types of forecast information currently available with which to make decisions, floodplain managers also desired new kinds of forecasts to guide decisions in their jobs. However, floodplain managers desired some forecast products that the meteorology and climatology communities cannot physically predict, given the current state of forecasting, observations, and historical data. Floodplain managers want forecasts at precise locations, stating “Of course, it would be great for them to pinpoint precisely what square mile where the rain is going to hit” or “Getting more localized is always a good thing.” Beyond a few hours ahead of time, it can be challenging to pinpoint precise precipitation amounts at the county level or more localized spatial scales.
Floodplain managers also wanted more localized, long-term forecasts as well. These desired forecasts could be used to guide bridge design or stormwater-plan developments for specific infrastructure or projects. The floodplain manager who had the title of engineer even explained that it would be helpful to have forecasts for specific rivers or watersheds far into the future: “It would be cool if there was some sort of prediction you could count on. And if that prediction could say ‘well, we need this bridge bottom beam elevation to be here because in 20 years, they’re predicting an increase in water surface elevation and it’s 99% accurate.’ That’d be great.” Currently, these types of predictions with such a long lead time and spatial precision are not possible (Dessai et al. 2009).
Floodplain managers identified additional types of information that would be helpful to them in their positions. One explained that having contact information for other decision-makers in their jurisdictions (e.g., other professionals in floodplain management) would be helpful for growing their professional network and discussing challenges. Better real-time data were also highly requested, with multiple participants desiring more Oklahoma Mesonet stations or USGS river gauges so they could have higher spatial resolution of observations. Additionally, participants wanted easier access to historical rainfall data, with some noting that the data were available, just not always in a format that was easy to use. Much of this desired information is difficult for the atmospheric science community to provide under current funding and priorities, but learning about the informational needs of different practitioners helps the weather, research, and operational communities understand what kinds of information would be useful for decision-makers and better support ongoing preparation and planning.
4. Discussion and conclusions
Floodplain managers use a variety of weather and climate information from multiple sources when they are making decisions within their jurisdictions. Besides this information, outside factors influence how decisions are made, such as budget shortfalls, desires of county commissioners, and limited authority of the floodplain manager. Often, these outside factors are more influential than a weather forecast or climate prediction. Additionally, even if the best forecasts were available, floodplain managers may be unable to act on them. Finally, when making decisions, floodplain managers desire forecast information that the atmospheric science community may be unable to predict given the current state of forecasting, indicating a disconnect between the people producing forecasts and those who are using the forecasts to make decisions.
All five themes from the interviews align with one or more components of the PADM. Table 3 displays the overlap between elements of the PADM and our themes. Abundance of information, which describes the types of weather and climate information floodplain managers use and where they obtain this information, relates to the environmental cues and information sources described in the PADM. The social network theme (i.e., the people floodplain managers work with to make decisions) relates to types of social cues in the model. Informational needs are associated with PADM’s information sources, with floodplain managers identifying new types of forecast information they would like to receive. Outside factors are related to the situational impediments that Lindell and Perry (2012) describe in the behavioral-response stage of the model. However, the types of outside factors that floodplain managers deal with are at a larger scale than the PADM originally described (i.e., the individual level). Therefore, the types of situational impediments that a floodplain manager may have to contend with are likely structural, societal issues such as budget constraints or authority in decision-making.
PADM elements and overlap with themes.
In addition, types of decisions are related to a different portion of the PADM—the decision process and information search—where forecast recipients seek out additional information about a hazard. Although long-term forecasts are available and can provide substantial advanced notice, floodplain managers were likely to continue waiting until an event neared before making decisions. However, for recurring problems like routine neighborhood flooding, some floodplain managers did act. These decisions were not necessarily influenced by different forecasts or predictions. Instead, they were influenced by public pressure to limit flood damage or by the knowledge that flooding would continue to be a problem if not addressed.
When asked about types of information related to extreme precipitation that they would like to have but currently do not, multiple floodplain managers asked for specific forecasts at localized levels (i.e., within a given square mile), with long lead times (multiple years ahead of time). While this level of specificity is not possible to provide, it is important for the atmospheric science community to understand the needs of various stakeholders when producing new forecast information.
One way to strengthen the content of warning messages in the PADM is by working with stakeholders to coproduce new forecast products. Knowledge coproduction has multiple benefits, including an increased likelihood that final products will be used and considered trustworthy (Cash et al. 2003). Additionally, coproduced knowledge is often more easily understood by final users because it is written and communicated in familiar language (Jasanoff and Wynne 1998). Coproduced forecast information is easier to integrate into existing decision-making processes (Lemos et al. 2012). By working with floodplain managers to use and interpret different types of forecast products, the atmospheric science community can explain components of the forecast. Then, combining forecast information with their localized expertise on flooding, floodplain managers can apply knowledge to make more informed decisions for their communities.
Even if floodplain managers had perfect weather and climate information, other factors often stand in the way of applying it. Because budgets are usually limited, managers are sometimes unable to implement technical solutions to floodplain management problems, even if they know what actions need to be taken. In addition, county commissioners determine project priorities, leaving floodplain managers to implement them, even if those projects are not the most pressing issues. Overall, these constraints mean that climate-informed floodplain management often does not happen, and communities are not taking advantage of the wide variety of forecast information available, leaving them vulnerable to changing conditions of future climates.
a. Study limitations and future directions
This study highlighted how floodplain managers in Oklahoma use weather and climate information to make decisions for their jurisdictions; however, the study has some limitations. First, the participant pool did not capture decision-makers from the entire precipitation gradient across Oklahoma. Floodplain managers in the wetter, eastern portion of the state may use different management strategies than floodplain managers in the drier, western portion of Oklahoma. Additionally, parts of Oklahoma will face different climate change impacts, requiring floodplain managers to adapt in a variety of ways. Second, the study only included participants from one U.S. state. Oklahoma has its own guidelines and regulations for floodplain management; these vary by state and influence how decisions are being made. Third, no floodplain managers who oversee tribal jurisdictions were interviewed, yet the region encompasses lands of 39 tribes, each with their own government and set of policies. Large tracts of land where tribes manage their own natural resources also will face the impacts of climate change, often with fewer resources than are available on nontribal land. Therefore, future research should expand this work to issues affecting tribal land. Future work also could investigate how floodplain managers might use climate projections in their area for long-term planning or increase communication between floodplain managers and those making predictions to provide contextualized, local information.
b. Conclusions
As the climate continues to become warmer and, in some locations, wetter, Oklahoma will face new issues related to floodplain management. Precipitation events are projected to become less frequent but more intense, meaning more precipitation will fall in a given event (Kloesel et al. 2018). Increased intensity of precipitation events may exacerbate existing flooding issues. For example, outdated stormwater infrastructure could be overwhelmed by a larger volume of water in the system. Areas that did not previously flood may begin to experience flooding more routinely when more precipitation falls, and neighborhoods may need to be relocated, as one interviewee noted. Existing infrastructure (e.g., roads and bridges) may not withstand increasing precipitation amounts. These impacts affect the decisions that floodplain managers make, so it is imperative these practitioners are engaging in climate-informed, long-term planning for their communities.
By improving floodplain management measures now, communities can be better prepared for the future and reduce the impacts of flooding. There appear to be gaps between what research suggests are best practices for floodplain management and what floodplain managers actually do. First, in some conversations with practitioners (VanBuskirk et al. 2021), managers have mentioned working with their professional networks to establish stronger connections between departments in their jurisdictions, thereby implementing a cross-government approach. For example, floodplain managers who regularly interface and collaborate with public works or stormwater management could use those existing relationships to explore long-term solutions for dealing with water and flooding. Second, city councils and other elected bodies could review the distribution of power to ensure that floodplain managers have the authority necessary to move forward with climate-informed decision-making.
Floodplain managers are not the only people who are impacted by communities not using climate information for long-term planning; residents also face impacts. Climate change will disproportionately impact those that are already most vulnerable, leaving them susceptible to the impacts of more routine flooding. However, floodplain managers and climate-informed floodplain management could help reduce vulnerability. For example, managers could assist vulnerable residents with moving out of floodplains or other flood-prone areas. Additionally, integrating floodplain management with other jurisdictional practices and processes could lead to better and more holistic decision-making. Climate change is one of the biggest issues facing communities across the world, and ensuring practitioners are making more-informed decisions for their jurisdictions enhances resilience and leaves them better able to absorb the shocks of a warmer world.
Acknowledgments.
This work was funded by the National Science Foundation ICER 1663840. The authors thank interview participants for their time and Drs. Mark Shafer and Travis Gliedt for reviewing earlier versions of this paper.
Data availability statement.
Because of the privacy and ethical concerns noted by the University of Oklahoma’s Institutional Review Board, neither the data nor the source of the data can be made available.
REFERENCES
Arnal, L., M.-H. Ramos, E. Coughlan de Perez, H. L. Cloke, E. Stephens, F. Wetterhall, S. J. van Andel, and F. Pappenberger, 2016: Willingness-to-pay for a probabilistic flood forecast: A risk-based decision-making game. Hydrol. Earth Syst. Sci., 20, 3109–3128, https://doi.org/10.5194/hess-20-3109-2016.
Arndt, D. S., J. B. Basara, R. A. McPherson, B. G. Illston, G. D. McManus, and D. B. Demko, 2009: Observations of the overland reintensification of Tropical Storm Erin (2007). Bull. Amer. Meteor. Soc., 90, 1079–1094, https://doi.org/10.1175/2009BAMS2644.1.
Association of State Floodplain Managers, 2010: Sample position descriptions. Accessed 16 March 2022, https://www.floods.org/career-center/careers-in-floodplain-management/sample-position-descriptions/.
Boddy, C. R., 2016: Sample size for qualitative research. Qual. Mark. Res., 19, 426–432, https://doi.org/10.1108/QMR-06-2016-0053.
Bolson, J., C. Martinez, N. Breuer, P. Srivastava, and P. Knox, 2013: Climate information use among southeast US water managers: Beyond barriers and toward opportunities. Reg. Environ. Change, 13, 141–151, https://doi.org/10.1007/s10113-013-0463-1.
Braun, V., and V. Clarke, 2006: Using thematic analysis in psychology. Qual. Res. Psychol., 3, 77–101, https://doi.org/10.1191/1478088706qp063oa.
Bruno Soares, M., and S. Dessai, 2016: Barriers and enablers to the use of seasonal climate forecasts amongst organisations in Europe. Climatic Change, 137, 89–103, https://doi.org/10.1007/s10584-016-1671-8.
Carbone, G. J., and K. Dow, 2005: Water resource management and drought forecasts in South Carolina. J. Amer. Water Resour. Assoc., 41, 145–155, https://doi.org/10.1111/j.1752-1688.2005.tb03724.x.
Cash, D. W., W. C. Clark, F. Alcock, N. M. Dickson, N. Eckley, D. H. Guston, J. Jäger, and R. B. Mitchell, 2003: Knowledge systems for sustainable development. Proc. Natl. Acad. Sci. USA, 100, 8086–8091, https://doi.org/10.1073/pnas.1231332100.
Coelho, C. A., and S. M. Costa, 2010: Challenges for integrating seasonal climate forecasts in user applications. Curr. Opin. Environ. Sustainability, 2, 317–325, https://doi.org/10.1016/j.cosust.2010.09.002.
Companion, M., and M. S. Chaiken, 2017: Responses to Disasters and Climate Change: Understanding Vulnerability and Fostering Resilience. CRC Press, 298 pp.
Craig, C. A., and S. Feng, 2018: A temporal and spatial analysis of climate change, weather events, and tourism businesses. Tour. Manage., 67, 351–361, https://doi.org/10.1016/j.tourman.2018.02.013.
Creswell, J. W., and C. N. Poth, 2017: Qualitative Inquiry and Research Design: Choosing Among Five Approaches. 4th ed. SAGE Publications Inc., 488 pp.
Crochemore, L., C. Cantone, I. G. Pechlivanidis, and C. S. Photiadou, 2021: How does seasonal forecast performance influence decision-making? Insights from a serious game. Bull. Amer. Meteor. Soc., 102, E1682–E1699, https://doi.org/10.1175/BAMS-D-20-0169.1.
Dahl, N., and M. Xue, 2016: Prediction of the 14 June 2010 Oklahoma City extreme precipitation and flooding event in a multiphysics multi-initial-conditions storm-scale Ensemble Forecasting System. Wea. Forecasting, 31, 1215–1246, https://doi.org/10.1175/WAF-D-15-0116.1.
Dessai, S., M. Hulme, R. Lempert, and R. Pielke, 2009: Climate prediction: A limit to adaptation? Adapting to Climate Change: Thresholds, Values, Governance, I. Lorenzoni, K. L. O’Brien, and W. N. Adger, Eds., Cambridge University Press, 64–78.
Elliott, R., C. T. Fischer, and D. L. Rennie, 1999: Evolving guidelines for publication of qualitative research studies in psychology and related fields. Br. J. Clin. Psychol., 38, 215–229, https://doi.org/10.1348/014466599162782.
Exum, N. G., E. Betanzo, K. J. Schwab, T. Y. J. Chen, S. Guikema, and D. E. Harvey, 2018: Extreme precipitation, public health emergencies, and safe drinking water in the USA. Curr. Environ. Health Rep., 5, 305–315, https://doi.org/10.1007/s40572-018-0200-5.
Goodman, L. A., 1961: Snowball sampling. Ann. Math. Stat., 32, 148–170, https://doi.org/10.1214/aoms/1177705148.
Hilton, R. W., 1981: The determinants of information value: Synthesizing some general results. Manage. Sci., 27, 57–64, https://doi.org/10.1287/mnsc.27.1.57.
Jasanoff, S., and B. Wynne, 1998: Science and decisionmaking. The Societal Framework, S. Rayner and E. L. Malone, Eds., Vol. 1, Human Choice and Climate Change, Battelle Press, 1–88.
Kloesel, K., and Coauthors, 2018: Southern Great Plains. Impacts, Risks, and Adaptation in the United States: Fourth National Climate Assessment, Vol. II, D. R. Reidmiller et al., Eds., U.S. Global Change Research Program Rep., 978–1026, https://doi.org/10.7930/NCA4.2018.CH23.
Kuligowski, E., 2021: Evacuation decision-making and behavior in wildfires: Past research, current challenges and a future research agenda. Fire Saf. J., 120, 103129, https://doi.org/10.1016/j.firesaf.2020.103129.
Lemos, M. C., C. J. Kirchhoff, and V. Ramprasad, 2012: Narrowing the climate information usability gap. Nat. Climate Change, 2, 789–794, https://doi.org/10.1038/nclimate1614.
Lindell, M. K., and R. W. Perry, 2012: The protective action decision model: Theoretical modifications and additional evidence. Risk Anal., 32, 616–632, https://doi.org/10.1111/j.1539-6924.2011.01647.x.
Liu, Y., Z. Ouyang, and P. Cheng, 2019: Predicting consumers’ adoption of electric vehicles during the city smog crisis: An application of the protective action decision model. J. Environ. Psychol., 64, 30–38, https://doi.org/10.1016/j.jenvp.2019.04.013.
Maxwell, J. A., 2012: Qualitative Research Design: An Interactive Approach. 3rd ed. SAGE Publications Inc., 232 pp.
McPherson, R. A., and Coauthors, 2007: Statewide monitoring of the mesoscale environment: A technical update on the Oklahoma Mesonet. J. Atmos. Oceanic Technol., 24, 301–321, https://doi.org/10.1175/JTECH1976.1.
Meseguer-Ruiz, O., J. A. Lopez-Bustins, L. Arbiol-Roca, J. Martin-Vide, J. Miró, and M. J. Estrela, 2021: Temporal changes in extreme precipitation and exposure of tourism in eastern and south-eastern Spain. Theor. Appl. Climatol., 144, 379–390, https://doi.org/10.1007/s00704-021-03548-6.
NCEI, 2020: U.S. Billion-Dollar Weather and Climate Disasters. NOAA, https://www.ncei.noaa.gov/access/monitoring/billions/.
Nelson, R. R., and S. G. Winter Jr., 1964: A case study in the economics of information and coordination the Weather Forecasting System. Quart. J. Econ., 78, 420–441, https://doi.org/10.2307/1879475.
Oklahoma Climate Survey, 2021: Precipitation history—Annual, statewide. Accessed 7 December 2022, https://climate.ok.gov/index.php/climate/climate_trends/precipitation_history_annual_statewide/CD00/prcp/Annual/oklahoma_climate.
Oklahoma Climatological Survey, 2011: Climate of Oklahoma overview. Accessed 13 April 2022, https://climate.ok.gov/index.php/site/page/climate_of_oklahoma.
Pagano, T. C., H. C. Hartmann, and S. Sorooshian, 2001: Using climate forecasts for water management: Arizona and the 1997–1998 El Niño. J. Amer. Water Resour. Assoc., 37, 1139–1153, https://doi.org/10.1111/j.1752-1688.2001.tb03628.x.
Ramos, M. H., S. J. van Andel, and F. Pappenberger, 2013: Do probabilistic forecasts lead to better decisions? Hydrol. Earth Syst. Sci., 17, 2219–2232, https://doi.org/10.5194/hess-17-2219-2013.
Rasmussen, L. V., C. J. Kirchhoff, and M. C. Lemos, 2017: Adaptation by stealth: Climate information use in the Great Lakes region across scales. Climatic Change, 140, 451–465, https://doi.org/10.1007/s10584-016-1857-0.
Rayner, S., D. Lach, and H. Ingram, 2005: Weather forecasts are for wimps: Why water resource managers do not use climate forecasts. Climatic Change, 69, 197–227, https://doi.org/10.1007/s10584-005-3148-z.
Ripberger, J. T., M. J. Krocak, W. W. Wehde, J. N. Allan, C. Silva, and H. Jenkins-Smith, 2019: Measuring tornado warning reception, comprehension, and response in the United States. Wea. Climate Soc., 11, 863–880, https://doi.org/10.1175/WCAS-D-19-0015.1.
Rosenzweig, C., F. N. Tubiello, R. Goldberg, E. Mills, and J. Bloomfield, 2002: Increased crop damage in the US from excess precipitation under climate change. Global Environ. Change, 12, 197–202, https://doi.org/10.1016/S0959-3780(02)00008-0.
Saldaña, J., 2009: The Coding Manual for Qualitative Researchers. SAGE Publications Inc., 329 pp.
Santana, F. N., D. J. X. Gonzalez, and G. Wong-Parodi, 2021: Psychological factors and social processes influencing wildfire smoke protective behavior: Insights from a case study in Northern California. Climate Risk Manage., 34, 100351, https://doi.org/10.1016/j.crm.2021.100351.
Schaefer, G. L., and R. F. Paetzold, 2001: SNOTEL (SNOwpack Telemetry) and SCAN (Soil Climate Analysis Network). Automated Weather Stations for Applications in Agriculture and Water Resources Management: Current Perspectives, K. G. Hubbard and M. V. K. Sivakuman, Eds., WMO/TD-1074, 187–194.
Shafiee-Jood, M., T. Deryugina, and X. Cai, 2021: Modeling users’ trust in drought forecasts. Wea. Climate Soc., 13, 649–664, https://doi.org/10.1175/WCAS-D-20-0081.1.
Sherman-Morris, K., T. Pechacek, D. J. Griffin, and J. Senkbeil, 2020: Tornado warning awareness, information needs and the barriers to protective action of individuals who are blind. Int. J. Disaster Risk Reduct., 50, 101709, https://doi.org/10.1016/j.ijdrr.2020.101709.
Siebeneck, L., E. Zavar, and R. Manandhar, 2022: An overview of local floodplain administrators and training needs in FEMA region 6. J. Homeland Secur. Emerg. Manage., 19, 149–173, https://doi.org/10.1515/jhsem-2021-0016.
South Central CASC, 2022: Precipitation projections. Accessed 7 December 2022, https://southcentralclimate.org/resources/climate-projections/precipitation-projections/.
Steinemann, A. C., 2006: Using climate forecasts for drought management. J. Appl. Meteor. Climatol., 45, 1353–1361, https://doi.org/10.1175/JAM2401.1.
Suarez, P., W. Anderson, V. Mahal, and T. Lakshmanan, 2005: Impacts of flooding and climate change on urban transportation: A systemwide performance assessment of the Boston Metro Area. Transp. Res., 10D, 231–244, https://doi.org/10.1016/j.trd.2005.04.007.
Taylor, A. L., S. Dessai, and W. B. de Bruin, 2015: Communicating uncertainty in seasonal and interannual climate forecasts in Europe. Philos. Trans. Roy. Soc., A373, 20140454, https://doi.org/10.1098/rsta.2014.0454.
Terpstra, T., and M. K. Lindell, 2013: Citizens’ perceptions of flood hazard adjustments: An application of the protective action decision model. Environ. Behav., 45, 993–1018, https://doi.org/10.1177/0013916512452427.
Tongco, M., 2007: Purposive sampling as a tool for informant selection. Ethnobot. Res. Appl., 5, 147–158, https://doi.org/10.17348/era.5.0.147-158.
Tyler, J., A.-A. Sadiq, D. S. Noonan, and R. M. Entress, 2021: Decision making for managing community flood risks: Perspectives of United States floodplain managers. Int. J. Disaster Risk Sci., 12, 649–660, https://doi.org/10.1007/s13753-021-00367-1.
VanBuskirk, O., P. Ćwik, R. A. McPherson, H. Lazrus, E. Martin, C. Kuster, and E. Mullens, 2021: Listening to stakeholders: Initiating research on subseasonal-to-seasonal heavy precipitation events in the contiguous United States by first understanding what stakeholders need. Bull. Amer. Meteor. Soc., 102, E1972–E1986, https://doi.org/10.1175/BAMS-D-20-0313.1.
Vasileiou, K., J. Barnett, S. Thorpe, and T. Young, 2018: Characterising and justifying sample size sufficiency in interview-based studies: Systematic analysis of qualitative health research over a 15-year period. BMC Med. Res. Methodol., 18, 148, https://doi.org/10.1186/s12874-018-0594-7.
Wang, S. S.-Y., W.-R. Huang, H.-H. Hsu, and R. R. Gillies, 2015: Role of the strengthened El Niño teleconnection in the May 2015 floods over the southern Great Plains. Geophys. Res. Lett., 42, 8140–8146, https://doi.org/10.1002/2015GL065211.
White, C. J., and Coauthors, 2022: Advances in the application and utility of subseasonal-to-seasonal predictions. Bull. Amer. Meteor. Soc., 103, E1448–E1472, https://doi.org/10.1175/BAMS-D-20-0224.1.
Yussouf, N., J. S. Kain, and A. J. Clark, 2016: Short-term probabilistic forecasts of the 31 May 2013 Oklahoma tornado and flash flood event using a continuous-update-cycle storm-scale ensemble system. Wea. Forecasting, 31, 957–983, https://doi.org/10.1175/WAF-D-15-0160.1.
