Abstract

Seasonal climate forecasting skill has improved over the past decades, accompanied by expectations that these forecasts, along with other climate information, will be increasingly used by water managers in certain regions of the United States. Most research efforts focus on why adoption does not occur; however, the important question of why adoption does occur has received little attention. Barriers to the use of climate information by this sector frequently identified include risk aversion, institutional constraints, and low forecast reliability. Relatively fewer researchers have focused on the identification and analysis of cases of adoption of climate information in the water management sector. Relying upon the results from observations and semistructured interviews conducted between 2006 and 2010 in South Florida, this research identifies the characteristics that enabled the early adoption of climate information by the South Florida Water Management District, one of the largest water management organizations in the United States. The findings herein are analyzed in relation to existing theories on technology transfer and innovation diffusion. Lessons from this specific case are situated in the context of the broader U.S. water management landscape. The research finds that the existence of in-house climate expertise, innovative agency culture, social networks linking water and climate science researchers, and serendipitous policy windows were critical factors enabling adoption. Additionally, models and information, including a long-range hydrologic model and a national government–issued seasonal climate forecast were readily available and could be incorporated into preexisting and trusted decision-support tools. Implications for climate services in the U.S. water sector are discussed.

1. Introduction

At a monthly Governing Board meeting in October of 2007 at the South Florida Water Management District (SFWMD)1 in West Palm Beach, Florida, a staff scientist addressed a large audience to discuss regional water resource conditions. The audience included local government employees, environmentalists, Everglades National Park researchers and advocates, local concerned citizens, agricultural, recreational, and tribal representatives, reporters, and SFWMD staff members. The SFWMD staff scientist discussed the state of El Niño–Southern Oscillation (ENSO) and described how it could potentially affect water levels in regional reservoirs, such as Lake Okeechobee, in the following months. She was providing this information to inform stakeholders about the considerations that would enter into upcoming decisions regarding reservoir releases. She gave her presentation on the potential seasonal climate impacts in a routine manner, one that seemed to interest the audience but did not generate an extraordinary response. From the response and the relaxed environment in the room, it seemed that this was not the first time this type of information had been presented in this setting. And in fact, seasonal climate information has been a routine part of the SFWMD planning process since 2000.

However routine this practice may be at the SFWMD, this is not the case for water management agencies across the United States. In fact, although seasonal climate information and forecast improvement have been the focus of considerable research efforts over the past two decades (Zebiak and Cane 1987; Anderson et al. 1998; Saha et al. 2006; Wang et al. 2009; Stefanova et al. 2012; Smith et al. 2012), their actual integration into water management decision making in the United States remains limited (USGCRP 1990; Glantz 1994, 1996; Callahan et al. 1999; O'Connor et al. 2005; Pagano et al. 2002; Carbone and Dow 2005; Rayner et al. 2005; Miles et al. 2006; Lemos 2008; Beller-Simms et al. 2008). The unique case of the SFWMD invites exploration of the dynamics that enabled the adoption of climate forecasts. This research addresses the following questions: 1) What conditions within and external to the SFWMD influenced the “successful” transfer of technology? 2) How do these characteristics compare to models of technology transfer and traits of early adopters? 3) Are the characteristics that encouraged innovation easily amenable to replication among other water management agencies?

In section 1, background on the integration of climate information into water resource management in the United States and theoretical bases for our framework of analysis are presented along with a description of the SFWMD case study site. Section 2 presents results from participatory research with regional stakeholders that provide insight into the different factors that encouraged innovation in the SFWMD. Section 3 discusses the theoretical relevance of our findings. We conclude with a discussion of the broader implications of our findings for current U.S. climate service initiatives related to water resource management.

Seasonal climate forecast use within U.S. water agencies

Water management challenges that exist during average conditions are exacerbated in years that are wetter or drier than normal. Rationally, strategies to reduce risk and improve water management outcomes during these years are sought after. One strategy, the implementation of seasonal climate forecasts (SCFs), has the potential to help decision makers avoid some of the negative consequences associated with reactive versus proactive responses to above or below normal rainfall conditions. Seasonal climate forecasts, which are driven largely by ENSO states, can provide information regarding climate conditions that are likely to occur in certain regions of the world with 1- to 3-month lead times (Ropelewski and Halpert 1987; Smith et al. 2012; see also the climate news available online at http://www.cpc.ncep.noaa.gov/). Methods for forecasting ENSO events have improved since the advent of dynamical seasonal climate prediction (Cane et al. 1986; Wang et al. 2009). Because of both increased ocean monitoring and observation [associated with the Tropical Ocean and Global Atmosphere (TOGA) program] and to improvements in coupled ocean–atmosphere climate modeling, SCF skill has improved since around the end of the twentieth century (Wang et al. 2009; Smith et al. 2012). However, prediction skill still varies geographically and seasonally and according to the variable being forecast (Wang et al. 2009). In general, with longer lead times, SCF skill decreases; however, precipitation forecasts for the southeastern United States with 3-month lead times show skill (Saha et al. 2006). In fact, Stefanova et al. (2012) found that the largest potential predictability of precipitation for the United States is in the southeast in spring and winter seasons. Stefanova et al. (2012) report results, which are based upon predictability, hindcasts, and Brier and Gerrity skill scores, showing that precipitation forecasts for the southeastern United States are particularly skillful.2 Although SCFs have skill for spring and winter seasons for the region, the remaining uncertainty due to atmospheric chaos and model parameterizations represents inherent limitations, which still present a challenge for many water managers who are among the potential users of these forecasts.

Using SCFs in water management would mean adapting decision making according to the information presented by forecasts to better prepare for extremes in flooding or drought. However, despite improvements in SCFs, U.S. water managers have resisted change (Rayner et al. 2005; O'Connor et al. 2005; Lemos 2008). As a result of the limited adoption, numerous barriers to the integration of SCFs into water resource management in the United States have been identified by researchers. These works have focused on both aspects of the forecasts themselves, as well as aspects of the decision environment and institutional contexts that constrain use (Jacobs et al. 2005; Ingram et al. 2008; Dilling and Lemos 2011). For example, factors including accessibility, communication, interpretation, and reliability of climate forecasts have been identified as barriers to their use (Stern and Easterling 1999; Barnston et al. 1999; Vedwan et al. 2008). Characteristics of the decision environment including both formal and informal institutions, the risk-averse nature of water managers, and public accountability of decision makers have also been shown to constrain the use of SCFs (Callahan et al. 1999; Broad and Agrawala 2000; Pagano et al. 2001; Pulwarty and Melis 2001; Pagano et al. 2002; Changnon and Vonnahme 2003; Carbone and Dow 2005; Rayner et al. 2005; Lemos 2008). For a comprehensive review of these works, see Dilling and Lemos (2011).

However, among U.S. water managers, there exist a small number of agencies that have been able to adopt forecast use despite the existence of many of the barriers that are discussed in the literature (Lemos 2008). While barriers have been well studied, with only a few exceptions, factors that enable this type of innovation among water managers have not been analyzed (Steinemann 2006; Ziervogel and Downing 2004; Lemos 2008; Feldman and Ingram 2009; Lowrey et al. 2009; Lemos et al. 2010). In the studies that focus on factors enabling forecast use in the United States, findings show that adoption is linked to extensive interactions between climate scientists and potential forecast users (Lowrey et al. 2009), iterative processes between scientists and decision makers (Dilling and Lemos 2011), and translation of materials into usable forms (Steinemann 2006). Additionally, Lemos (2008) finds that among U.S. water managers, factors related to “individual flexibility, discretion, and accountability also affect the ability of managers to use climate information in water management” (Lemos 2008, p. 1388).

Case studies focusing on factors that have enabled innovation—where innovation, in this regard, is described as the adoption of SCFs in decision making—can in theory illuminate possible directions for improving the types of forecasts, their optimal forms for communication, and structural policy and informational barriers that can be addressed to maximize potential use of forecasts. This research builds upon prior studies in hopes of contributing toward a middle-range theory3 by providing findings from a case study of an early adopter of SCFs, the South Florida Water Management District—a regional governmental agency that oversees the water resources from Orlando to the Florida Keys and serves a population of 7.7 million residents. While the SFWMD demonstrates many of the traits identified as barriers in literature, including, for example, high levels of visibility, accountability, and risk-adverse decision makers, the agency has been able to overcome these obstacles to incorporate SCFs into operational decision making. This paper analyzes why.

2. Theoretical framework: Technology transfer and innovation diffusion

We draw on the technology transfer literature for a framework for our analysis. Technology transfer studies cover a broad array of subjects, with relevance for diverse organizations and disciplinary communities. Because of its broad reach, technology transfer is defined and perceived very differently by separate audiences and researchers (Zhao and Reisman 1992; Bozeman 2000). We use a definition of technology transfer, which is, in its most basic form, the application of information into use (Rogers 1995). Although often thought of as a tangible entity or product, technology also encompasses information, including skills and knowledge (Sahal 1981; Grubler 1998). When viewed in this light, SCFs and climate information in general can be thought of as a type of technology that can be transferred (Agrawala and Broad 2002). However, despite prior identification of SCFs as a transferrable technology, literature that explicitly links technology transfer with seasonal climate forecasts remains limited, especially in the public sector (Garfin et al. 2008). In the climate domain, the majority of technology transfer literature concentrates on the international transfer of clean energy and adaptation technologies, likely due to the design of international climate change policy mechanisms.

A general typology of early adopters of technology also exists in the literature on innovation diffusion, defined as a process by which an innovation is communicated through certain channels over time among the members of a social system (Rogers 2003). Early adopters are typically those who perceive the forecasts or innovation to have advantages, exhibit compatibility between technology and existing system, and engage in interpersonal communication between different individuals (Table 1). Further, innovators tend to be active information seekers, with a high exposure to mass media and far-reaching interpersonal networks (Rogers 2003). This aspect of innovation diffusion theory has been applied to study SCF adoption, or the lack thereof (Hayman et al. 2007). Prior findings on characteristics of early adopters along with the technology transfer framework inform our analysis of actions by the SFWMD.

Table 1.

Some well-established generalizations of indicators for innovation (Rogers 2003).

Some well-established generalizations of indicators for innovation (Rogers 2003).
Some well-established generalizations of indicators for innovation (Rogers 2003).

Among the limited applications of the concept is work that compares how well different models of technology transfer (including the “appropriability model,” the “dissemination model,” the “knowledge utilization model,” and the “contextual adaptation model”) describe case studies where SCFs have been applied. Findings show that the contextual adaptation model, which suggests that the utility of a technology is shaped by how the technology fits into the decision-making context, best described cases of successful SCF transfer (Johnson et al. 1997; Agrawala and Broad 2002; Ziervogel and Downing 2004). Working within the contextual adaptation model, there have been efforts to connect potential SCF users with scientists to improve fit and to understand the dynamics that effect interactions between them. These efforts have been largely studied under the heading of coproduction, defined as the production of information or technology through collaboration of scientists and decision makers or stakeholders who incorporate values and criteria from both communities (Cash et al. 2006). Research into the coproduction of climate science and water decision making contributes to our understanding of barriers and needs for the creation of usable knowledge or technology (Lemos and Morehouse 2005; Jacobs et al. 2005; Cash et al. 2006).

More information is still needed to support the development of a technology transfer framework suitable for climate science applications. Ideally, this would include all stages of technology transfer, from dissemination through adoption, and could guide organizations working within the boundaries of climate science and water resource management. In an attempt to reduce this knowledge gap, we analyze the SFWMD case study of successful SCF transfer within a framework of determinants of technology transfer identified by researchers in the behavioral sciences. Our goal in doing so is to recognize possible determinants of technology transfer specifically related to SCF use in U.S. water resource management.

Another line of research on technology transfer can be traced to the field of behavioral sciences (Backer et al. 1995; Bozeman 2000; Rogers 2003). From this research, both empirical and analytic perspectives on determinants of effective technology transfer have been identified (Glaser et al. 1983; Backer 1991; Dunn and Holzner 1988; Backer et al. 1995; Rogers 2003). As described by Backer et al. (1995), who studied substance abuse programs, four fundamental conditions are necessary for successful technology transfer: 1) dissemination (awareness of and access to the new knowledge), 2) evaluation (credible evidence that adoption leads to improved practice without excessive costs or undesirable side effects (e.g., assessment of the benefit of using the technology), 3) resources (money, materials, and personnel needed to implement the new practice), and 4) human dynamics of change (interventions to overcome resistances to change among the people who will need to implement the innovation) (Backer 1991). Six strategies for successful technology transfer are also identified in this body of literature (Fig. 1) (Backer et al. 1995). To the determinants in our framework, we added forecast skill as the relatively high precipitation forecast skill for the southeastern United States represents a unique component to adoption in our case study. Our analysis uses this framework to see if and how adoption of forecast use at the SFWMD was contingent upon these determinants and strategies.

Fig. 1.

Determinants of and strategies for successful technology transfer identified by Backer et al. (1995).

Fig. 1.

Determinants of and strategies for successful technology transfer identified by Backer et al. (1995).

3. Case study: The South Florida Water Management District

The SFWMD is responsible for an area covering all or parts of 16 counties in Southern Florida and 16 164 square miles (Fig. 2). Home to 7.7 million residents (43% of the state population), the SFWMD is the largest of Florida's five regional water management districts (USGS 2008). In 2010, the district employed 1800 people and had a fiscal year budget of $1.5 billion (SFWMD 2010a). These factors combine to give the agency substantial political power and visibility within the state.

Fig. 2.

Map of the SFWMD (from SFWMD 2011).

Fig. 2.

Map of the SFWMD (from SFWMD 2011).

Rainfall is the primary source for replenishing water supplies and sustaining regional ecology (SFWMD 2010a). In normal years, 70 percent of the average (1970 to 2000) annual precipitation of around 60 inches falls in the summer months, June through September (Southeast Regional Climate Center 2008). Rainfall in southern Florida is affected by both ENSO and the Atlantic multidecadal oscillation (AMO) (Mestas-Nuñez and Enfield 2003), with the strongest signal due to ENSO (Mantua et al. 1997; O'Brien et al. 1999; Obeysekera et al. 2006; Stefanova et al. 2012). During El Niño years, greater than normal winter precipitation and below normal temperatures are more likely (Table 2). La Niña years bring generally drier, warmer winter conditions, increasing water shortage risks in spring and summer months (Table 2).

Table 2.

ENSO impacts in Florida (adapted from SECC AgroClimate: http://agroclimate.org). One asterisk (*) indicates data from Jagtap et al. (2002), and two (**) from Zierden (2008).

ENSO impacts in Florida (adapted from SECC AgroClimate: http://agroclimate.org). One asterisk (*) indicates data from Jagtap et al. (2002), and two (**) from Zierden (2008).
ENSO impacts in Florida (adapted from SECC AgroClimate: http://agroclimate.org). One asterisk (*) indicates data from Jagtap et al. (2002), and two (**) from Zierden (2008).

a. SFWMD

The SFWMD began in 1948 as the Central and Southern Florida Flood Control Project (C&SF FCD), a massive flood control project extending from Orlando to the Everglades. Lake Okeechobee, which covers almost 700 square miles, is known as the heart of the regional water system, with the Kissimmee River to the north and the sawgrass Everglades to the south, eventually connecting to Florida Bay (SFWMD 2008). Lake Okeechobee supplies drinking water for populations around the lake, agricultural irrigation, back-up supplies for east coast communities, and freshwater for estuaries and the Everglades. To maintain regional water levels and prevent saltwater intrusion, the district operates approximately 2300 miles of canals and levees, 61 pumping stations, and 2200 water control structures (SFWMD 2008).

The principal source of water for Lake Okeechobee is the Kissimmee River and the main outlets are the artificial waterways that cut through the peninsula toward the Gulf of Mexico and Atlantic Ocean. When the land was in its natural state, the lake would seasonally overflow its banks and find channels to the Atlantic through the St. Lucie, New, and Miami Rivers and to the Gulf through the Caloosahatchee River (Blake 1980). However, the canals now connecting the lake to estuaries have created long-term stresses on the ecosystems that are subject to impacts of lake release schedules (Fling et al. 2004).

Lake Okeechobee management objectives include flood control, agricultural and urban water supply, fulfilling Seminole Tribe water rights, navigation, recreation, and ecological habitat preservation. These objectives are managed within the constraints of the lake regulation schedule by protocols decided upon by the United States Army Corps of Engineers (USACE), with input from the SFWMD. If possible, when rain is scarce, water levels in coastal canals are maintained higher to prevent saltwater intrusion and to protect water supply. In anticipation of heavy rainfall, floodgates are opened to lower water levels in canals and reservoirs.

b. SFWMD stakeholders

In the SFWMD, water releases and other water management decisions are highly visible. Stakeholder and public input into decision making have increased over the past decade and participation is now part of the institutional framework. Through regular meetings and workshops with stakeholder groups, interested parties participate in decision making to provide input into highly contentious water release and allocation decisions (Fig. 3). During times of water shortage, tensions exist between environmentalists looking to protect the Everglades and nearby estuaries, agriculturalists that want water for sugar production and other economically fruitful crops, and municipal water suppliers who want to grow their customer base and prefer to avoid imposing water restrictions.

Fig. 3.

Schematic of information flow, stakeholder interaction, and where climate fits in at the SFWMD.

Fig. 3.

Schematic of information flow, stakeholder interaction, and where climate fits in at the SFWMD.

4. Methods

This paper is the result of a multidisciplinary research project carried out over a 4-yr period between 2006 and 2010 in southeast Florida. A mixed-methods approach that relied upon qualitative and quantitative analyses was employed. The qualitative component of the work consisted of semistructured ethnographic interviews (Spradley 1979), snowball sampling, and participant observation (Bernard 1988). Snowball sampling is a network sampling method where a few key informants are identified and asked to 1) list other members of their population and 2) recommend individuals from this list who might be interviewed (Bernard 1988). To build a better understanding of the decision-making environment, research was conducted with decision makers at state, regional, and local government agencies, with water utilities employees, and agricultural, environmental, and tribal representatives. The sampling and selection for interviews was nonrandom, or purposive (Glaser and Strauss 1967; Agar 1980; Bernard 1988). In addition, through a fellowship with the Governor's Office, one member of the research team (Bolson) worked in the Florida Department of Environmental Protection Office of Water Policy for seven months and thus gained insight and access to documents and persons involved with water management. The focused interactions afforded by this experience were invaluable to this research. Over the course of this research, we spoke with 68 stakeholders, attended 32 meetings, participated in hundreds of informal conversations, carried out an intensive literature review, and conducted content analysis.

a. Stakeholder identification and selection

Early stakeholder identification was achieved by attending and observing monthly SFWMD Governing Board meetings. In monthly meetings SFWMD staff scientists present their research to influence the recommendations that the Governing Board provides to the Army Corps of Engineers who are ultimately responsible for reservoir release decisions (Fig. 3). Potential participants were contacted, either in person at meetings or by phone, to organize initial interviews. The stakeholder representatives who agreed to participate represent a diverse group of regional interests (Table 3).

Table 3.

Agencies represented in semistructured stakeholder interviews and numbers of individuals participating from each agency.

Agencies represented in semistructured stakeholder interviews and numbers of individuals participating from each agency.
Agencies represented in semistructured stakeholder interviews and numbers of individuals participating from each agency.

b. Semistructured interviews

Early interview interactions focused on learning about the water system, the context for decision making, and assessing needs for and uses of climate information. Interviews with 68 people consisted of in-person meetings or phone interviews when meeting face to face was not feasible. Through snowball sampling methods, we identified seven key informants who are (or formerly were) actively engaged in climate issues at the district. The sample included staff and management level engineers, senior-level operations staff, senior-level planners (past and current), and two department directors, all of whom worked at the SFWMD during the time when SCF use was initiated at the SFWMD. Interviews with these key informants were organized around another set of questions (11) designed to characterize the factors contributing to the adoption of SCFs in decision making. These interviews were conducted by telephone and lasted, on average, for one hour. Interview questions were divided into several sections, each with a central question that was followed by additional inquiries designed to clarify and/or probe deeper into the discussion. We asked interviewees to describe their role in the organization; how they learned about forecasts and became interested in using them at the SFWMD; how the adoption process occurred and the role of different individuals in that process; specific factors that enabled the SFWMD to incorporate forecasts into operational decision making; aspects of the decision environment and management culture, including support for research and innovation; barriers encountered during the adoption process and how these were overcome; and the role of external scientists and networks in the adoption of SCF use. Although we used a questionnaire to guide interviews, they were conducted informally and interviewees were encouraged to discuss the issues they thought were important.

The results from the key informant interviews were analyzed to reconstruct the history of forecast use in the region and how the SFWMD was able to formalize the use of SCFs in decision making. Responses and interactions were coded to identify the factors that allowed for innovation and evaluated within the framework of technology transfer. Findings are organized from general contextual descriptions to more specific findings.

5. Results and discussion

a. The use of seasonal climate forecasts at the SFWMD

From the early 1900s through 2000, the lake was operated using calendar-based regulation schedules. However, during the 1990s, together, the SFWMD and the USACE conducted a comprehensive study to develop a new regulation schedule for Lake Okeechobee to allow for more flexibility in decision making. Around the same time, a meteorologist employed by the SFWMD recognized the potential benefit SCFs offered to water managers. Along with a group of collaborators, from both within and outside the agency, this individual researched various methods for predicting inflows to Lake Okeechobee (Zhang and Trimble 1996; Trimble et al. 1998, 2006).

Ultimately SFWMD scientists proposed a suite of tools to use in operational planning, which included position analysis (Hirsch 1978), precipitation predictions from the Climate Prediction Center (CPC) (1- and 3-month climate outlooks), subsampling from past years, and hydrologic modeling (Cadavid et al. 1999). Position analysis methods estimate risk associated with a system's operational plan over future months given current conditions (Cadavid et al. 1999). When the CPC issues an El Niño or La Niña forecast, indicating higher probability of wet or dry conditions, the SFWMD uses conditional position analysis (Cadavid et al. 1999). The objective of conditional position analysis is to estimate the future responses of the system given not only the current state but also a climate forecast. Together with simulations run using the South Florida Water Management Model (SFWMM), these tools allow SFWMD scientists to estimate hydrologic net inflow to Lake Okeechobee for future months based upon current conditions, historical events, and SCFs4 (Fig. 4). By July 2000, the USACE had adopted the Water Supply and Environment (WSE) Lake Okeechobee regulation schedule (SFWMD 2010b).

Fig. 4.

SFWMD operational guidelines decision tree (SFWMD 2007).

Fig. 4.

SFWMD operational guidelines decision tree (SFWMD 2007).

The formal integration of climate information into decision making has persisted through subsequent revisions of the regulation schedule. In 2005, the USACE and SFWMD identified high risk of structural failure of the Herbert Hoover Dike. Although they had already planned to devise a new lake regulation schedule, the newly realized danger resulted in an expedited schedule. The Lake Okeechobee Regulation Schedule (LORS) Study identified and evaluated alternative Lake Okeechobee regulation schedules. The process resulted in the adoption of an interim regulation schedule for the lake (the 2008 LORS), which regulates lake levels while repairs to the dike are completed. Even the 2008 LORS incorporates the use of climate forecasts, as it specifies the use of CPC 1- and 3-month precipitation outlooks and the Palmer Drought Severity Index (USACE 2008; SFWMD 2010b). The USACE guidance document also calls for the presentation of climate outlooks at public workshops through simplified figures (Table 4).

Table 4.

Example of how USACE reports the water supply performance measures to the public (from SFWMD 2010b). For the scoring scheme, high, moderate, and low refer to high, moderate, and low probability of adverse impacts. WCA refers to Watershed Critical Area.

Example of how USACE reports the water supply performance measures to the public (from SFWMD 2010b). For the scoring scheme, high, moderate, and low refer to high, moderate, and low probability of adverse impacts. WCA refers to Watershed Critical Area.
Example of how USACE reports the water supply performance measures to the public (from SFWMD 2010b). For the scoring scheme, high, moderate, and low refer to high, moderate, and low probability of adverse impacts. WCA refers to Watershed Critical Area.

b. Factors enabling technology transfer

Although the SFMWD ultimately succeeded in changing the Lake Okeechobee regulation schedule, the process was described by a senior-level engineer, involved in the process, as challenging, long, and arduous. To learn how they were able to overcome challenges to forecast adoption at the SFWMD, we asked the question, “What factors were most important in allowing you to begin using seasonal climate forecasts in decision-making?” Responses to this question were coded into five categories: 1) embedded capacity, 2) networks with climate science researchers, 3) innovative agency culture, 4) tools to represent and communicate information, and 5) policy window/timing. Subtopics were used to further organize responses (Table 5). We then organized these categories according to the determinants of the technology transfer framework, resources, evaluation, networks, or human dynamics of change that they best matched.

Table 5.

Response categories and rates for factors that enabled seasonal climate forecast use adoption and the corresponding determinants of technology transfer.

Response categories and rates for factors that enabled seasonal climate forecast use adoption and the corresponding determinants of technology transfer.
Response categories and rates for factors that enabled seasonal climate forecast use adoption and the corresponding determinants of technology transfer.

1) Resources

Staffing was identified as being vital to the adoption of forecasts by a majority (57%) of respondents. One senior-level SFWMD planner, a management-level engineer, and both the Modeling and Operations Department Directors credited a small group of scientists, who acted as opinion leaders or champions on climate issues, as being critical to adoption. One individual, in particular, who was hired at the SFWMD to fill an engineering position despite his background in meteorology, was identified by these respondents as a major agent of change. In speaking with this individual, who is still an SFWMD engineer, we learned that it was not only his expertise that enabled the process to evolve, but also the leadership at the time. He informed us that the leadership encouraged folks to pursue applied research that interested them, and allowed them creativity and flexibility in their work. We confirmed the existence of strong leadership support for applied research by interviewing a retired SFWMD planner who oversaw the climate research activities at the time of adoption.

With the support of his superiors, the staff engineer was able to spend time and resources developing methods to convert forecasts from the terciles presented in precipitation outlooks into hydrologic outlooks that could be used to drive SFWMD system models. The conditional position analysis methods and SFWMD decision trees resulted from this work (Fig. 4). According to two management-level engineers and the SFWMD Modeling Department Director, this step was critical to establishing sustained forecast use as it facilitated compatibility between forecast products and already existing systems of decision making. From these accounts, we learned that in-house experts with capacity to conduct the required research and understanding of organizational operations, and leadership that allowed the expenditure of resources on translating SCFs to fit with operational protocols, enabled the linkage between SCFs and SFWMD decision making.

In-house experts continue to be essential to forecast use. This is because forecasts are not quantitatively plugged into a model. Instead, the use of forecasts is a more subjective process, where an individual staff member interprets each forecast in the context of other pieces of information, including the availability of analog years and additional climate information, and decides on whether to use it in the conditional position analysis. The same staff member who originally argued for the use of forecasts is still the lead on this task. Continued reliance upon this staff expert for guidance on forecast use raises questions about the long-term sustainability of SCF use at the SFWMD. This issue was raised by a staff engineer during our interview:

I am wondering if I leave what the next guy will do, I have to try to get other people involved and I am not sure who is ready to do this (personal communication, 21 April 2011).

Even though the consideration of forecasts is written into formal operational guidelines, the actual use of forecasts still depends on in-house expertise. As the process involves a certain amount of subjectivity, it is possible that institutional memory could be lost with staff turnover and, along with it, the capacity to use forecasts.

2) Resources/evaluation

(i) Long-term hydrologic modeling tool

Among 86% (the greatest agreement among participants on any question) of the participants interviewed, the most important factor to the adoption of SCFs identified was the existence of a long-term hydrologic model. The model interviewees referred to was the South Florida Water Management Model, which allowed them to run long-term simulations to demonstrate how past decisions could have been improved with the use of forecasts. According to both staff and management-level engineers, the Operational and the Modeling Department Directors, and one of the planners with whom we spoke, the ability to show impacts, to run “what if” scenarios, and to look at all the different performance measures that matter to SFWMD stakeholders was a major factor in helping to get everyone on board with forecast use.

A chief engineer informed us that the long-term simulations allowed their group to show stakeholders and the Governing Board how meeting management objectives might have been improved by use of SCFs. He conveyed that model output bolstered stakeholder support for SCFs, another critical factor in their adoption. The retired SFWMD planner, with whom we spoke, demonstrated agreement: “Everything we do the public scrutinizes, the public bought into the model runs” (personal communication, 21 April 2011). The Chief Engineer of Operations relayed his and others' early fears that including SCF use would further complicate the legal environment. He described how building willingness of managers to consider the scenarios shown by model simulations helped them overcome this barrier.

Other responses (43%) pointed to the importance of model simulations for building specific stakeholder group's support. A staff engineer informed us that model output encouraged buy-in from environmental stakeholders who recognized the potential to improve water storage/releases and avoid negative ecological impacts like those in the past. The SFWMD model simulations using SCFs may have also bridged diverse stakeholder interests. For example, one planner commented,

Ultimately, there were two sides, one side wanted to raise the lake for water supply, the other side, the people who cared about the littoral zone of the lake, wanted to keep the lake low. The use of climate forecasts was used as a meeting ground between these two (personal communication, 21 April 2011).

We found that SFWMD stakeholders and community members viewed the model output as credible evidence that adoption of SCF use could lead to improved practice. Because of the diverse expectations and visibility of regional water resource management decisions, stakeholder support of SCF use was essential to adoption. The SFWMM and the capacity to run simulations generated the necessary support by providing sufficient evaluation of the potential benefits of SCF use.

(ii) Federal Government––issued forecast (CPC)

In the Lake Okeechobee regulation schedule National Oceanic and Atmospheric Administration (NOAA) CPC outlooks are formally identified as the source for SCFs. When we discussed this legality with a SFWMD chief engineer, he informed us that using CPC forecasts provided increased security over other products because of the forecast source. As he said,

When we negotiated the change to operations, the USACE felt that they had to say they were using the federal agency's predictions. It carries a lot of weight to use the NOAA CPC forecasts (personal communication, 22 April 2011).

Another respondent, the Director of Operations, substantiated this claim by reporting that because of the economic risk associated with forecast use, the SFWMD and USACE decided it best to use a federally issued product. The decision to use these forecasts seems to be related less to forecast skill than to trust in the product's provider. Researchers have identified tools that could potentially be an improvement over the CPC outlooks, yet still the SFWMD continues to rely on their use as outlined in the operational guidelines (Miralles-Wilhelm et al. 2005). Thus, the availability of a federally issued forecast was a contributing factor in SCF adoption.

3) Dissemination

In addition to the critical roles of in-house climate experts, systems models, and NOAA's CPC Outlooks, the Director of Operations described the valuable contributions of other scientists with whom SFWMD scientists and engineers collaborated. Judging from observations and accounts of interactions between SFWMD engineers and outside scientists, SFWMD scientists appear to be and to have been part of an established network that includes climate science researchers from outside the agency. District engineers and modelers worked and continue to work with research scientists from universities, NOAA's Atlantic Oceanographic and Meteorological Laboratory, and the National Hurricane Center, all of which are physically located nearby. As a meteorologist by training, one staff engineer already knew about ENSO and the potential of SCFs when he was hired by the SFWMD. Therefore SFWMD did not depend upon external networks to learn about the existence of climate forecasts; however, they did look to those networks for support. A staff engineer and the Director of Operations informed us that networks provided data, expertise, and consultation throughout the translation process, which were key to operational tool development.

4) Human dynamics of change

(i) Agency culture

Given their role in helping stakeholders and operational staff overcome resistance to change, model simulations, CPC forecast availability, and in-house experts who pursued SCF use can all be viewed as agents of change. Additionally, interview responses indicated that the culture of the SFWMD agency supported change or innovation in a number of ways. These include characteristics of leadership (mentioned by 43%), staffing decisions (mentioned by 29%), flexibility in research and work (mentioned by 57%), and communication (mentioned by 71%).

(ii) Leadership and staffing

From interview responses and observations, we learned that the SFWMD, while still being operationally conservative and risk averse, makes changes through what one senior-level engineer called “incremental changes over the long term” (personal communication, 22 April 2011). In general, the individuals we spoke with had been employed at the agency for many years (12–20 years) and described low turnover among scientific staff and leadership. In talking with a management-level engineer, we learned that within this structure, innovation occurred when a newcomer entered the organization, helping to change the management culture. Further, a SFWMD staff engineer reported that the staff was encouraged to think outside the box and to seek outside information. These responses reveal certain aspects of an innovative agency culture where individuals with diverse backgrounds were hired and senior-level management were supportive of new ideas being brought forth.

(iii) Flexibility

We also learned from respondents about the adaptive culture of the SFWMD and how this culture reflects in the agency staff's and management's willingness to change. The Director of Operations conveyed how the culture has grown over time, potentially in response to the agency's history of engineering and experimentation resulting from projects like the Central and Southern Florida Project and the Comprehensive Everglades Restoration Project (CERP). He described it as follows:

The natural system has been highly altered and damaged. It is our intention to do the best with what we can. You have to embrace science and technology to do work in our environment. We are learning all the time. The mistakes of the past were not of engineering but of getting locked in to certain positions. Flexibility is very important. It is important to be adaptable (personal communication, 21 April 2011).

This response, in particular, indicates that an adaptive, experimental precedent has been set at the agency and that flexibility, which encourages innovation, has resulted from the unique history of water management in the region.

(iv) Communication

Responses from a majority (71%) of the interviewees indicated that communication was also crucial to SCF adoption. The Director of Modeling detailed weekly meetings with USACE and other federal employees held throughout the year, before and after the WSE was accepted. We learned from the Operations Department Director that the success of the integration process depended upon regular sustained interactions among SFWMD staff, USACE staff, and stakeholders. Staff and management-level engineers also referenced a great deal of communication among staff members, with upper-level management, and with stakeholders. A chief engineer told us how early communications were facilitated by framing the use of SCFs as a cost-free management option. This view was reiterated by another staff engineer who said that not having to spend money on this technology provided an incentive to test it out.

Communication was not only critical to getting forecasts adopted; it remains a priority and continues through public dissemination of outlooks, stakeholder workshops, and outreach. The district communicates potential benefits of SCF use and about how decisions using SCFs are made to the public. The Director of Modeling cautioned that because people see moderate success of one forecast use, they expect the same type of success for every event and that concerns decision makers. For example, in 2000, high water levels caused adverse effects to the Lake Okeechobee ecosystem. Concerns about the ecosystem and a climate forecast that indicated an increased probability of an end to dry conditions led decision makers to release freshwater from the lake. The releases sent freshwater pulses to the coast, which damaged the Caloosahatchee and St. Lucie Estuaries and, combined with persistent drought conditions, resulted in water supply shortages (Steinemann et al. 2002). The events resulted in backlash from stakeholders and public outcry. In situations like this, the SFWMD has used the events as opportunities to communicate about forecasts, to establish dialogue, and to encourage continued buy-in from stakeholders. Communications strategies have allowed the SFWMD to overcome some of the natural resistance to organizational change, but acceptance of the probabilistic nature of SCFs remains a challenge.

(v) Policy window opening

In interviews, both prior and ongoing weather events were mentioned as providing the opportunity to adopt forecasts at the agency. For example, the Director of Operations and a management-level engineer identified the 1998 ENSO event as a signal event that motivated the agency to start looking at climate. According to their accounts, there was interest in applying data from both SFWMD staff engineers and from interested stakeholders. The management engineer elaborated what happened:

In 1998, NOAA started publishing data from a series of buoys in the Pacific Ocean. We all knew what El Niño was but we did not see it until it manifested itself and we had anomalous conditions, the lake went up to 15 feet in the dry season. People saw buoys, and asked about why they were not being used (personal communication, 18 February 2009).

The confluence of events including, the 1998 ENSO, the hiring of a meteorologist on staff, the increase in public and staff interest in SCFs, the availability of a CPC SCF, and a policy window opening (created by the opportunity to formally adapt the lake regulation schedule) fostered the formal integration of new information (Hart and Victor 1993). Another engineer described these events simply: “all the stars had to line up” (personal communication, 18 February 2009).

5) Determinants of Technology Transfer

What was characterized by one interviewee as serendipity or as all of the stars aligning might be described by the presence of determinants and/or strategies of successful tech transfer described by other cases. First, dissemination was accomplished in hiring an individual with knowledge of and interest in SCFs, who also acted as a champion, and through interactions with a network of scientists. The individuals who were involved in original decisions to use SCFs also fit well within the typology of early adopters described by Rogers (2003) (Table 1). Second, interviewees described resources including in-house expertise, staff time, and tools including the SFWMM and the CPC forecasts, as critical to adoption. Third, evaluation of SCF potential was made possible through SFWMM simulations, another factor identified as critical to adoption. Finally, we discovered many factors and strategies applied by SFWMD scientists and staff, which allowed them to overcome human resistance to change. These included intensive communication and interactions, modeling simulation exercises, research/professional flexibility, dynamic leadership, and formally institutionalizing SCF use in regulation schedules. However, there were also some factors revealed by interviewees that might be a bit more difficult to categorize as determinants including the opening of a policy window and occurrence of a signal event. The relative degree to which these factors influenced adoption was difficult to assess; however, according to four key informants, timing seems to have been a critical factor in adoption. From analysis of discussions with these informants, it becomes apparent that the SFWMD was well prepared with expert staff, tools, leadership, and capacity at the right time. The determinants and strategies of technology transfer were in place as was an opportunity.

c. Agency versus forecast specific characteristics surrounding tech transfer

Integrating climate into decision making met the classic resistance at the SFWMD; however, successful technology transfer was achieved. The results of this research indicate that agency characteristics were more responsible for overcoming those barriers than forecasts themselves. In fact, trust in forecast provider, not the skill of forecasts, appeared to be the driving factor in selection of CPC forecasts. Because of the risk of repercussions from suboptimal outcomes of forecast application, the SFWMD ultimately decided that the federal government–issued product, although not necessarily the most skillful product on the market (Miralles-Wilhelm et al. 2005), would be the least controversial. However, it is worth reemphasizing that seasonal precipitation forecasts for the region during spring and winter are widely considered more skillful than for any other region in the United States (Stefanova et al. 2012).

d. Replicability

The replicability of the sustained application of SCF beyond South Florida is certainly feasible. As a chief engineer said,

It is always a big effort to set a technology up and get it going, but others can copy what we have done. They can't use our models but they can retool some models. You have to have a model with your operational protocols (personal communication, 21 April 2011).

Tools like the SFWMM are part and parcel of water management culture, but other factors that were identified as enablers of innovation by this case study might be more difficult to replicate. These include the factors identified as the human agents of change, the existence of exceptional regional forecast skill, and the opening of a policy window. Additionally, a large factor in innovation was the “in-house” research group that had freedom and good relations with/overlap with the operational group. For those pursuing work in the climate services arena, identification of these characteristics may provide insight into the potential for integration of climate information, helping them make more efficient decisions on where and how to focus resources.

6. Conclusions

This article set out to understand the factors contributing to the adoption of seasonal climate forecasts in a case study of the SFWMD, an agency that has been touted as an example of operational seasonal climate forecast use. Findings reveal a complex set of factors, including serendipity that led to adoption. Without connections to outside climate researchers, a meteorologist with climate expertise on staff, significant climate forecast skill, sophisticated models, hindcasting proof of concept, or “the stars lining up,” adoption might not have occurred. From a broader analytical perspective, it is evident that the theoretical framework of technology transfer provides a suitable lens to evaluate the dominant factors driving SCF adoption. However, additional factors including forecast skill, the ability to demonstrate skill, and the opening of a policy window should also be included in that framework. Which of these factors and/or combinations of factors are necessary and which are not necessitates additional research involving a comparative case study approach.

Unlike many other technologies that prove themselves through survival in the open market, SCFs are often part of a larger decision-making process involving public and private sectors, and thus their value may be harder to discern. Despite seeming proud of their accomplishments and committed to the process, interview participants could not offer information about the value added from using forecasts. While hindcasting exercises described in previous sections demonstrated better reservoir release decisions when including ENSO information into the SFWMM, there have not been efforts to translate “better” decisions into economic, ecological, or other metrics. While the added value of SCF inclusion may be self-evident to the current SFWMD administration, some have noted that future managers or Governing Board members may not be sympathetic to innovations, or in times of budget crunches, may be simply unable to justify continued support of the salary of a climatologist within the agency, especially in the case of a negative decision outcome tied to a “bad” forecast. This has implications for additional skill sets such as economics that may be useful to add to the SFWMD team,5 wherein the ability to demonstrate the “value added” of SCF application at the SFWMD may help to maintain its use.

Our findings do not suggest a formula for widespread adoption of SCFs among U.S. water managers, especially at smaller agencies that do not have the resources and expertise that the SFWMD and other similar-sized agencies have. In fact, there is question as to whether SCF should play an increased role in regions where forecast skill is weak. With these caveats, our findings may have implications for climate service design within the water management sector (e.g., see the Climate Services Description at http://www.climate.gov/#climateWatch). Because both the theoretical determinants of tech transfer and the traits of early adopters can be used to explain SFWMD success, they might be used to guide other adaptive management efforts beyond the use of SCFs. The SFWMD case suggests that focus should be turned toward creating internal capacity by increasing skill sets of employees and fostering a culture of openness to innovation—even if initially confined to the research realm—to deal with a range of climate issues at water management institutions. A proactive approach by U.S. water managers to develop in-house expertise and increase communication among networks of water managers, climate scientists, and relevant social science researchers can begin to address the complexity of organizational decision making and human behavior in the context of environmental fluctuations. Such collaboration can allow water management agencies at a range of scales to more rationally evaluate the potential for adoption of SCFs and other nontraditional informational inputs.

REFERENCES

REFERENCES
Agar
,
M.
,
1980
: The Professional Stranger: An Informal Introduction to Ethnography. Academic Press, 276 pp.
Agrawala
,
S.
, and
K.
Broad
,
2002
: Technology transfer perspective on climate forecast applications. Research in Science and Technology Studies: Knowledge and Technology Transfer, M. De Laet, Ed., Knowledge and Society Series, Vol. 13, JAI Press, 45–69.
Anderson
,
D. L. T.
,
E. S.
Sarachik
, and
P. B.
Webster
, Eds.,
1998
:
The TOGA decade—Reviewing the progress of El Niño research and predictions
.
J. Geophys. Res.
,
103
,
14 167
14 510
.
Backer
,
T. E.
,
1991
: Drug abuse technology transfer. National Institute on Drug Abuse Rep., 85 pp.
Backer
,
T. E.
,
S. L.
David
, and
G.
Saucy
,
1995
: Introduction. Reviewing the Behavioral Science Knowledge Base on Technology Transfer, T. E. Backer, S. L. David, and G. Saucy, Eds., National Institute on Drug Abuse Research Monograph Series, Vol. 155, United States Department of Health and Human Services, 1–21.
Barnston
,
A. G.
,
M. H.
Glantz
, and
Y.
He
,
1999
:
Predictive skill of statistical and dynamical climate models in SST forecasts during the 1997–98 El Niño episode and the 1998 La Niña onset
.
Bull. Amer. Meteor. Soc.
,
80
,
217
243
.
Beller-Simms
,
N.
,
H.
Ingram
,
D.
Feldman
,
N.
Mantua
,
K. L.
Jacobs
, and
A.
Waple
, Eds.,
2008
: Decision-support experiments and evaluations using seasonal-to-interannual forecasts and observational data: A focus on water resources. Climate Change Science Program and Subcommittee on Global Change Research Rep., National Climatic Data Center, 190 pp. [Available online at http://www.climatescience.gov/Library/sap/sap5-3/final-report/sap5-3-final-all.pdf.]
Bernard
,
R. H.
,
1988
: Research Methods in Cultural Anthropology. Sage Publishing, 824 pp.
Blake
,
N. M.
,
1980
: Land into Water—Water into Land: A History of Water Management in Florida. University Presses of Florida, 344 pp.
Bozeman
,
B.
,
2000
:
Technology transfer and public policy: A review of research and theory
.
Res. Policy
,
29
,
627
655
.
Broad
,
K.
, and
S.
Agrawala
,
2000
:
The Ethiopia food crisis: Uses and limits of climate forecasts
.
Science
,
289
,
1693
1694
.
Cadavid
,
L. G.
,
R. V.
Zee
,
C.
White
,
P.
Trimble
, and
J.
Obeysekera
,
1999
: Operational hydrology in South Florida using climate forecast. Proc. 19th Annual Hydrology Days, Washington, DC, Amer. Geophys. Union, 1–17.
Callahan
,
B.
,
E.
Miles
, and
D.
Fluharty
,
1999
:
Policy implications of climate forecasts for water resources management in the Pacific Northwest
.
Policy Sci.
,
32
,
269
293
.
Cane
,
M. A.
,
S. E.
Zebiak
, and
S. C.
Dolan
,
1986
:
Experimental forecasts of El Niño
.
Nature
,
321
,
827
832
.
Carbone
,
G. J.
, and
K.
Dow
,
2005
:
Water resource management and drought forecasts in South Carolina
.
J. Amer. Water Resour. Assoc.
,
41
,
145
155
.
Cash
,
D. W.
,
J. C.
Borck
,
A. G.
Patt
, and
Coauthors
,
2006
:
Countering the loading-dock approach to linking science and decision making: Comparative analysis of El Niño/Southern Oscillation (ENSO) forecasting systems
.
Sci. Technol. Hum. Values
,
31
,
465
494
.
Changnon
,
S. A.
, and
D. R.
Vonnahme
,
2003
: Impact of spring 2000 drought forecasts on Midwestern water management. J. Water Res. Plann. Manage.,129, 18–25.
CPC
, cited
2011
: National Weather Service: Climate Prediction Center: Climate News. [Available online at http://www.cpc.ncep.noaa.gov/.]
Dilling
,
L.
, and
M. C.
Lemos
,
2011
:
Creating usable science: Opportunities and constraints for climate knowledge use and their implications for science policy
.
Global Environ. Change
,
21
,
680
689
.
Dunn
,
W. N.
, and
B.
Holzner
,
1988
:
Knowledge in society: Anatomy of an emergent field
.
Knowl. Soc.
,
1
,
3
26
.
Feldman
,
D. L.
, and
H. M.
Ingram
,
2009
:
Making science useful to decision makers: Climate forecasts, water management, and knowledge networks
.
Wea. Climate Soc.
,
1
,
9
21
.
Fling
,
H.
,
N.
Aumen
,
T.
Armentano
, and
F.
Mazzotti
,
2004
: The role of flow in the Everglades landscape. University of Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, Rep. CIR1452, 7 pp.
Garfin
,
G. M.
, and
Coauthors
,
2008
: Regional knowledge transfers: Coping with drought through collaboration. Extended Abstracts, 17th Conf. on Applied Climatology, Whistler, BC, Canada, Amer. Meteor. Soc., 7.3. [Available online at https://ams.confex.com/ams/13MontMet17AP/techprogram/paper_141284.htm.]
Gerrity
,
J. P.
,
1992
:
A note on Gandin and Murphy's equitable skill score
.
Mon. Wea. Rev.
,
120
,
2709
2712
.
Glantz
,
M. H.
,
1994
: Usable Science: Food Security, Early Warning, and El Niño: Proceedings of the Workshop on ENSO/FEWS, Budapest, Hungary, October 1993. UNEP, 250 pp.
Glantz
,
M. H.
,
1996
: Currents of Change: El Niño's Impact on Climate and Society. Cambridge University Press, 208 pp.
Glaser
,
B.
, and
A.
Strauss
,
1967
: The Discovery of Grounded Theory: Strategies for Qualitative Research. Aldine Publishing Company, 271 pp.
Glaser
,
E. M.
,
H.
Abelson
, and
K. N.
Garrison
,
1983
: Putting Knowledge to Use. Jossey-Bass, 636 pp.
Grubler
,
A.
,
1998
: Technology and Global Change. Cambridge University Press, 464 pp.
Hart
,
D. M.
, and
D. G.
Victor
,
1993
:
Scientific elites and the making of U.S. policy for climate change research, 1957–74
.
Soc. Stud. Sci.
,
23
,
643
680
.
Hayman
,
P.
,
J.
Crean
,
J.
Mullen
, and
K.
Parton
,
2007
:
How do probabilistic seasonal climate forecasts compare with other innovations that Australian farmers are encouraged to adopt?
Aust. J. Agric. Res.
,
58
,
975
984
.
Hilton
,
R. W.
,
1981
:
The determinants of information value: Synthesizing some general results
.
Manage. Sci.
,
27
,
57
64
.
Hine
,
C.
,
2007
:
Multi-sited ethnography as a middle range methodology for contemporary STS
.
Sci. Technol. Hum. Values
,
32
,
652
671
.
Hirsch
,
R. M.
,
1978
:
Risk analyses for a water supply system: Occoquan Reservoir, Fairfax, and Prince William Counties, Virginia, USA
.
Hydrol. Sci. Bull.
,
23
,
476
505
.
Ingram
,
H.
,
D.
Feldman
,
N.
Mantua
,
K. L.
Jacobs
,
D.
Fort
,
N.
Beller-Simms
, and
A. M.
Waple
,
2008
: The changing context. Decision-support experiments and evaluations using seasonal-to-interannual forecasts and observational data: A focus on water resources. N. Beller-Simms et al., Eds., Climate Change Science Program and Subcommittee on Global Change Research Rep., National Climatic Data Center, 7–28. [Available online at http://downloads.climatescience.gov/sap/sap5-3/sap5-3-final-ch1.pdf.]
Jacobs
,
K.
,
G.
Garfin
, and
M.
Lenart
,
2005
:
More than just talk: Connecting science and decision making
.
Environment
,
47
,
6
21
.
Jagtap
,
S. S.
,
J. W.
Jones
,
P.
Hildebrand
,
D.
Letson
,
J. J.
O'Brien
,
G.
Podestá
,
D.
Zierden
, and
F.
Zazueta
,
2002
:
Responding to stakeholders' demands for climate information: From research to applications in Florida
.
Agric. Syst.
,
74
,
415
430
.
Johnson
,
S. D.
,
E. F.
Gatz
, and
D.
Hicks
,
1997
:
Expanding the content base of technology education: Technology transfer as a topic of study
.
J. Technol. Educ.
,
8
,
35
49
.
Lemos
,
M. C.
,
2008
:
What influences innovation adoption by water managers? Climate information use in Brazil and the U.S
.
J. Amer. Water Resour. Assoc.
,
44
,
1388
1396
.
Lemos
,
M. C.
, and
B. J.
Morehouse
,
2005
:
The co-production of science and policy in integrated climate assessments
.
Global Environ. Change
,
15
,
57
68
.
Lemos
,
M. C.
,
A.
Bell
,
N. E.
Engle
,
R. M.
Formiga-Johnsson
, and
D. R.
Nelson
,
2010
:
Technical knowledge and water resources management: A comparative study of river basin councils, Brazil
.
Water Resour. Res.
,
46
,
W06523
,
doi:10.1029/2009WR007949
.
Letson
,
D.
,
G.
Podestá
,
C.
Messina
, and
A.
Ferreyra
,
2005
:
The uncertain value of perfect ENSO phase forecasts: Stochastic agricultural prices and intra-phase climatic variations
.
Climatic Change
,
69
,
163
196
.
Lowrey
,
J.
,
A.
Ray
, and
R.
Webb
,
2009
:
Factors influencing the use of climate information by Colorado municipal water managers
.
Climate Res.
,
40
,
103
119
.
Mantua
,
N. J.
,
S. R.
Hare
,
Y.
Zhang
,
J. M.
Wallace
, and
R. C.
Francis
,
1997
:
A Pacific decadal climate oscillation with impacts on salmon
.
Bull. Amer. Meteor. Soc.
,
78
,
1069
1079
.
Merton
,
R.
,
1968
: Social Theory and Social Structure. The Free Press, 702 pp.
Mestas-Nuñez
,
A. M.
, and
D. B.
Enfield
,
2003
: Investigation of intra-seasonal to multi-decadal variability in South Florida rainfall. NOAA/OAR Tech. Memo., 79 pp. [Available online at http://sunburn.aoml.noaa.gov/phod/docs/Mestas_Enfield_FR_2003.pdf.]
Meza
,
F. J.
,
J. W.
Hansen
, and
D.
Osgood
,
2008
:
Economic value of seasonal climate forecasts for agriculture: Review of ex-ante assessments and recommendations for future research
.
J. Appl. Meteor. Climatol.
,
47
,
1269
1286
.
Miles
,
E.
,
A.
Snover
,
L. W.
Binder
,
E.
Sarachik
,
P.
Mote
, and
N.
Mantua
,
2006
:
An approach to designing a national climate service
.
Proc. Natl. Acad. Sci. USA
,
26
,
19 616
19 623
.
Miralles-Wilhelm
,
F.
,
P. J.
Trimble
,
G.
Podesta
,
D.
Letson
, and
K.
Broad
,
2005
: Climate-based estimation of hydrologic inflow into Lake Okeechobee, Florida. J. Water Res. Plann. Manage.,131, 394–401.
Murphy
,
A. H.
,
1973
:
A new vector partition of the probability score
.
J. Appl. Meteor.
,
12
,
595
600
.
Obeysekera
,
J.
,
P.
Trimble
,
C.
Neidrauer
,
C.
Pathak
,
J.
VanArman
,
T.
Strowd
, and
C.
Hall
,
2006
: Consideration of long-term climatic variability in regional modeling for SFWMD planning & operations. 2007 South Florida Environmental Rep., Appendix 2-3, 51 pp. [Available online at http://www.sfwmd.gov/portal/page/portal/pg_grp_sfwmd_sfer/portlet_prevreport/volume1/appendices/v1_app_2-3.pdf.]
O'Brien
,
J. J.
,
D. F.
Zierden
,
D.
Legler
,
J. W.
Hansen
,
J. W.
Jones
,
A. G.
Smajstrla
,
G.
Podestá
, and
D.
Letson
,
1999
: El Niño, La Niña and Florida's climate: Effects on agriculture and forestry. Florida Consortium Staff Paper, 32 pp.
O'Connor
,
R. E.
,
B.
Yarnal
,
K.
Dow
,
C. L.
Jocoy
, and
G. J.
Carbone
,
2005
:
Feeling at risk matters: Water managers and the decision to use forecasts
.
Risk Anal.
,
5
,
1265
1275
.
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
.
Pagano
,
T. C.
,
H. C.
Hartmann
, and
S.
Sorooshian
,
2002
:
Factors affecting seasonal forecast use in Arizona water management: A case study of the 1997–98 El Niño
.
Climate Res.
,
21
,
259
269
.
Pulwarty
,
R.
, and
T. S.
Melis
,
2001
:
Climate extremes and adaptive management on the Colorado River: Lessons from the 1997–1998 ENSO event
.
J. Environ. Manage.
,
63
,
307
324
.
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
.
Rogers
,
E. M.
,
1995
: Diffusion of Innovations. 4th ed. Free Press, 518 pp.
Rogers
,
E. M.
,
2003
: Diffusion of Innovations. 5th ed. Free Press, 512 pp.
Ropelewski
,
C. F.
, and
M. S.
Halpert
,
1987
:
Global and regional scale precipitation patterns associated with the El Niño/Southern Oscillation
.
Mon. Wea. Rev.
,
115
,
1606
1626
.
Saha
,
S.
, and
Coauthors
,
2006
:
The NCEP Climate Forecast System
.
J. Climate
,
19
,
3483
3517
.
Sahal
,
D.
,
1981
:
Alternative conceptions of technology
.
Res. Policy
,
10
,
2
24
.
SFWMD
,
2007
: South Florida Water Management District Water supply/environmental regulation schedule (WSE) frequently asked questions. Preparations for the Water Summit, West Palm Beach, Florida, South Florida Water Management District, 2 pp.
SFWMD
,
2008
: South Florida Environmental Report Executive Summary: A joint report of the South Florida Water Management District and the Florida Department of Environmental Protection. 55 pp.
SFWMD
, cited
2010a
: The South Florida Water Management District information page. [Available online at http://www.sfwmd.gov.]
SFWMD
,
2010b
: Final adaptive protocols for Lake Okeechobee operations. A Joint Report of the South Florida Water Management District, the U.S. Army Corps of Engineers, Jacksonville District, and the Florida Department of Environmental Protection, 271 pp.
SFWMD
, cited
2011
: South Florida Water Management District: About us. [Available online at http://www.sfwmd.gov.]
Smith
,
D. M.
,
A. A.
Scaife
, and
B. P.
Kirtman
,
2012
:
What is the current state of scientific knowledge with regard to seasonal and decadal forecasting?
Environ. Res. Lett.
,
7
,
015602
,
doi:10.1088/1748-9326/7/1/015602
.
Southeast Regional Climate Center
, cited
2008
: Historical regional climate summaries for Florida. [Available online at http://www.sercc.com.]
Spradley
,
J.
,
1979
: The Ethnographic Interview. Holt, Rinehart, and Winston, 247 pp.
Stefanova
,
L.
,
V.
Misra
,
J. J.
O'Brien
,
E. P.
Chassignet
, and
S.
Hameed
,
2012
:
Hindcast skill and predictability for precipitation and two-meter air temperature anomalies in global circulation models over the southeast United States
.
Climate Dyn.
,
38
,
161
173
.
Steinemann
,
A. C.
,
2006
:
Using climate forecasts for drought management
.
J. Appl. Meteor. Climatol.
,
45
,
1353
1361
.
Steinemann
,
A. C.
,
K.
Havens
, and
L.
Hornung
,
2002
:
The managed recession of Lake Okeechobee, Florida: Integrating science and natural resource management
.
Conserv. Ecol.
,
6
,
17
.
Stern
,
P. C.
, and
W. E.
Easterling
,
1999
: Making Climate Forecasts Matter. National Academy Press, 175 pp.
Trimble
,
P.
,
E. R.
Santee
, and
C. J.
Neidrauer
,
1998
:
Preliminary estimate of impacts of sea level rise on the regional water resources of southeastern Florida
.
J. Coast. Res.
,
26
,
252
255
.
Trimble
,
P.
,
J.
Obeysekera
,
D.
Cadavid
, and
E. R.
Sante
,
2006
: Application of climate outlooks for water management in south Florida. Climate Variations, Climate Change, and Water Resources Engineering, J. D. Garbrecht and T. C. Peichota, Eds., American Society of Civil Engineers, 65–85.
USACE
, cited
2008
: U.S. Army Corp of Engineers Everglades Division, Herbert Hoover Dike Branch: Herbert Hoover Dike project description. [Available online at http://www.saj.usace.army.mil/Portals/44/docs/FactSheets/HHD_FS_Rehab_Spring2013_508.pdf.]
USGCRP
,
1990
: Our changing planet: Fiscal Year 1990 Research Plan, Executive Summary. A Report by the United States Global Change Research Program and the Subcommittee on Global Change Research, 50 pp. [Available online at http://library.globalchange.gov/products/annualreports/our-changing-planet-the-u-s-global-change-research-program-for-fiscal-year-1990.]
USGS
,
2008
: Water use in Florida 2005 and trends 1950–2008. U.S. Geological Survey Fact Sheet 2008-3080, 4 pp. [Available online at http://pubs.usgs.gov/fs/2008/3080/.]
Vedwan
,
N.
,
S.
Ahmad
,
F.
Miralles-Wilhelm
,
K.
Broad
,
D.
Letson
, and
G.
Podesta
,
2008
:
Institutional evolution in Lake Okeechobee Management in Florida: Characteristics, impacts, and limitations
.
Water Resour. Manage.
,
22
,
699
718
.
Wang
,
B.
, and
Coauthors
,
2009
:
Advance and prospectus of seasonal prediction: Assessment of the APCC/CliPAS 14-model ensemble retrospective seasonal prediction (1980–2004)
.
Climate Dyn.
,
33
,
93
117
.
Zebiak
,
S.
, and
M. A.
Cane
,
1987
:
A model El Niño–Southern Oscillation
.
Mon. Wea. Rev.
,
115
,
2262
2278
.
Zhang
,
E.
, and
P. J.
Trimble
,
1996
:
Predicting effects of climate fluctuations for water management by applying neural networks
.
World Resour. Rev.
,
8
,
334
348
.
Zhao
,
L. M.
, and
A.
Reisman
,
1992
:
Toward meta research on technology-transfer
.
IEEE Trans. Eng. Manage.
,
39
,
13
21
.
Zierden
,
D. F.
,
2008
: Florida's climate: From El Niño to climate change. Proc. Water Resource Institute Symp., Gainesville, FL, University of Florida. [Available online at http://waterinstitute.ufl.edu/symposium/downloads/presentations/zierden.pdf.]
Ziervogel
,
G.
, and
T. E.
Downing
,
2004
:
Stakeholder networks: Improving seasonal climate forecasts
.
Climatic Change
,
65
,
73
101
.

Footnotes

1

The SFWMD is the largest of the five regional water management agencies in Florida. The SFWMD is responsible for responsible for water supply, water quality, flood control, and environmental restoration in central and southern Florida.

2

The Brier skill score (BSS) (Murphy 1973) measures the probabilistic hindcast skill compared to a climatological hindcast, or the magnitude of the probability of forecast error. The Gerrity skill score (GSS) is used for assessment of categorical forecast skill (Gerrity 1992). The GSS is a weighted sum of elements from the contingency table of possible combinations of the forecast and observed categories, where the weights favor forecasts closer to the observed categories. The BSS (GSS) values are between zero (negative infinity) and one, where values over zero indicate the forecast is more skillful than climatology and a value of one indicates a perfect forecast. The BSS for winter [December–February (DJF)] precipitation anomaly was shown to be 0.22 for >0 mm day−1 and 0.23 for >0.5 mm day−1 over all land points in the southeast United States (Stefanova et al. 2012). The GSS for precipitation for the multimodel ensemble for DJF over the region was shown to be 0.37 (Stefanova et al. 2012). Stefanova et al. (2012) found that the Southeast is the only U.S. region with a predictability ratio exceeding 0.5, in both spring [March–May (MAM)] and winter (DJF). Only the wintertime hindcasts show strong anomaly correlations, exceeding 70%, while springtime forecasts show much lower anomaly correlations of 10% to 30% (Stefanova et al. 2012).

3

Merton (1968) called for theories of the middle range as “a way of avoiding over-ambitious and premature attempts to develop unified theories with little obvious connection to observable social experience; and a tendency to produce descriptive data focused on specific situations without providing enough conceptualization to guide future study or generalize to other situations” (Hine 2007).

4

These methods are incorporated into operational guidelines detailed by the SFWMD decision tree (Fig. 4) and are described in detail in Cadavid et al. (1999) and Miralles-Wilhelm et al. (2005) (SFWMD 2010b).

5

There are many examples that quantify the economic value of climate information, including SCFs, largely through agricultural applications (Hilton 1981; Letson et al. 2005). For example, innovative methods demonstrated in prior agro-climate research, including incorporation of elicited management responses into integrated economic modeling and measurement of environmental impacts in addition to monetary outcomes of forecast response, may be applicable in the water sector as well (Meza et al. 2008).