*2.1.1. Sondeos*

A sondeo (Spanish for sounding) is a semi-structured discussion in which a two- to threeperson multi-disciplinary team engages one or two people from the target audience in conversation8. Most of the sondeos conducted for the SECC have been part of a graduate course in field research methods. The students and their instructors meet to discuss the problem of interest and to agree on a general set of questions to ask. These questions are a guide to conversation, rather than a formal questionnaire. The course members divide into small multi-disciplinary teams to conduct their conversations, typically at the residence or place of work for the people of target audience. For the agricultural community, extension agents have been vital in helping identify people from diverse target audiences, where the targets have included small and large farms, vegetable growers at farmers markets, farmers with and without irrigation, and others.

An important benefit to a conversational approach is that it often elicits key issues that the researcher could not have anticipated, issues that would likely have been missed with an interview or survey that has a list of pre-established questions. In keeping with its conversational nature, the researchers do not take notes during the conversation. Rather, when the conversation is completed, each researcher writes their own synthesis of the discussion. In one day, a single team can usually complete three or four discussions, which typically last about one hour, followed by note writing.

At the end of each day, all teams assemble to discuss their findings and to identify new questions that will guide conversations on the following day. Some questions may be retained throughout the sondeo in order to provide continuity, but the discussions evolve day-by-day as teams engage in conversations, learn, share their learning, and modify their conversation guide questions. After one week of field work, the course identifies a leader to write the sondeo report. Examples of SECC sondeo reports can be downloaded from http://SEClimate.org/pubs.php.

#### *2.1.2. Climate working groups*

482 Risk Management – Current Issues and Challenges

engagement.

*2.1.1. Sondeos* 

are further described:

with and without irrigation, and others.

http://SEClimate.org/pubs.php.

typically last about one hour, followed by note writing.

By providing a range of engagement methods, individuals can self select how they will engage in the community, depending on their level of interest, availability of time, and willingness to commit to an activity. This ability of community members to select the engagement activities in which they will participate applies to nearly all members of the community, including end-users, extension agents, and researchers. The exception to the self selection clause is a core team of three or more individuals who are fully committed to the community. The SECC strives to have a team that includes at least one social scientist (anthropologist or rural sociologist), climate scientist, and agricultural scientist. The ability of this committed team to work together will be the most critical factor in the success of the

Two of the engagement methods noted above – sondeos and working groups (both powerful and less commonly used in other reported participatory research approaches) –

A sondeo (Spanish for sounding) is a semi-structured discussion in which a two- to threeperson multi-disciplinary team engages one or two people from the target audience in conversation8. Most of the sondeos conducted for the SECC have been part of a graduate course in field research methods. The students and their instructors meet to discuss the problem of interest and to agree on a general set of questions to ask. These questions are a guide to conversation, rather than a formal questionnaire. The course members divide into small multi-disciplinary teams to conduct their conversations, typically at the residence or place of work for the people of target audience. For the agricultural community, extension agents have been vital in helping identify people from diverse target audiences, where the targets have included small and large farms, vegetable growers at farmers markets, farmers

An important benefit to a conversational approach is that it often elicits key issues that the researcher could not have anticipated, issues that would likely have been missed with an interview or survey that has a list of pre-established questions. In keeping with its conversational nature, the researchers do not take notes during the conversation. Rather, when the conversation is completed, each researcher writes their own synthesis of the discussion. In one day, a single team can usually complete three or four discussions, which

At the end of each day, all teams assemble to discuss their findings and to identify new questions that will guide conversations on the following day. Some questions may be retained throughout the sondeo in order to provide continuity, but the discussions evolve day-by-day as teams engage in conversations, learn, share their learning, and modify their conversation guide questions. After one week of field work, the course identifies a leader to write the sondeo report. Examples of SECC sondeo reports can be downloaded from

A working group includes members from science and the broader community who meet regularly, typically 3 or 4 times per year, to engage in dialogue on the new findings from science, information and technology needs of the broader community9. The steps for building and nurturing a climate working group are outlined in Table 1. The SECC has successfully established working groups for agricultural and water supply utility communities. Both have about 25 to 30 members and both are highly productive, yet each has distinct features that reflect the differences among the communities.

A key element to the success of a climate working group is commitment, both on the part of the individual members of the groups and the institutions that they represent, if any. For example, most members of the water supply utility climate working group represent a particular water provider, city, or agency. If a member of any particular institution is not able to attend a meeting, the participating institutions identify an alternate who can attend. This policy provides both continuity and assures that the institutions have also committed to the working group.


**Table 1.** Phases and activities for the building and nurturing a climate working group. Source: Bartels et al. (2011)9.

By far, these climate working groups demand the greatest level of commitment from the learning community for any of the engagement methods that have been tested, but it is precisely this commitment that helps them advance science. The climate working groups help researchers build collaborative relationships with different stakeholder groups for ongoing learning, both by the scientists and the stakeholders. They link research with realworld decision needs to help improve resource management strategies of stakeholders as well as improving the research and education programs of the science community. Most importantly, climate working groups engage members from diverse stakeholder groups that might not otherwise interact and promote the legitimacy of the science community as a source of information and technology that is relevant to solving the wicked climate problems that society faces.

Improving Climate Risk Management at Local Level –

Techniques, Case Studies, Good Practices and Guidelines for World Meteorological Organization Members 485

Climate risk assessments typically include statistical analyses of historical climate indicator records and assessment of information on climate-sensitive impacts, together with understanding of the climate mechanisms and the cascade of processes leading to these impacts. Geographical information and mapping may also be used to assess the zones where impacts are recurrent and are associated with human losses and/or infrastructure damages. Temporal changes of impacts and their related climate hazard characteristics are also often a key part of climate risk assessment, and may be directly linked with social, economical or environmental variables which may change exposure and resilience. In an ideal world, there would be millions of meteorological stations contributing to development of an accurate idea of the historical evolution of climate variables. The reality is, however, that there are not enough meteorological stations, the available stations are not evenly distributed in time and space, available data are not always digitized or shared, and there can be problems in some cases in the quality, completeness or homogeneity of the available data. Notwithstanding such issues, individual climate records measured at specific places integrate the history of the complex interactions between land, air, sea, ecosystems, and community in those locales. The final result is expressed in those climate records and consequently the history of this whole interaction process is reflected in time series of the measured values, or of their

Analysis derived from climate indices/indexes, such as that provided by the CCl/CLIVAR/JCOMM Expert Team (ET) on Climate Change Detection and Indices (ETCCDI) (see http://www.clivar.org/organization/etccdi/etccdi.php) is a powerful tool that can be applied at local level. In places where high quality climate records (preferably long period, minimal gaps, and homogeneous) are available, the information that can be delivered through such climate analysis is absolutely useful, in conjunction with social and economic and other information for that locale, in DRR, adaptation and CRM processes. Such information is more accurate and in most cases more appropriate than that generated by downscaled models, but in zones where there are no available stations, information

Climate risk assessment may also include a future element, utilising climate change projections and/or 'what-if' scenarios to explore the potential impacts of future scenarios of change. Practical obstacles to using information about future conditions are diverse, ranging from limitations in modeling climate system complexities (e.g. projections having coarse spatial and temporal resolution, limited predictability of some relevant variables, at scales that matter for decision making and forecast skill characterization), to procedural, institutional, and cognitive barriers in receiving or understanding climatic information, and the capacity and willingness of decision-makers to modify actions20,21. In addition, functional, structural, and social factors inhibit joint problem identification and collaborative knowledge production between providers and users. These include divergent objectives, needs, scope, and priorities; different institutional settings and standards, as well as

departures from a chosen reference period.

generated by downscaling is the next best alternative.

differing cultural values, understanding, and mistrust22,23.

#### **2.2. Climate risk assessment**

As with the definition of climate risk, or CRM, there are many definitions of risk assessment. A common theme across most is the requirement for a process and/or technique that provides information with which to assess the key risk or risks. For example, the Society for Risk Analysis proposes that: "Risk assessment is the process of establishing information regarding acceptable levels of a risk and/or levels of risk for an individual, group, society, or the environment" (see: http://www.sra.org/resources\_glossary\_p-r.php).

Risk assessments involve analysis techniques, methodologies and tools that have the key quality of assessing uncertainty (a common quality of risk), either quantitatively or qualitatively, and representing this as some measure of likelihood and/or probability. Climate risk assessment is used to help decision makers optimize resources for responding to climate-related disasters and reducing risks and impacts associated with current and future-projected climate variablity and change. It is one of the first stages of CRM, and involves identification and synthesis of hazard and vulnerability information/data that is relevant to the specific climate-related risks identified through the 'User and provider engagement and collaboration' step. One very important consideration in all climate risk assessments is the balance between the quantification of climate hazards (intensity, frequency and/or duration) and the approach to estimate the main elements of vulnerability on the ground i.e. level of exposure, poverty, exclusion, education, organizational capacity, infrastructure among others. Both hazard and vulnerability estimations may be validated using historical information of climate events and changes in socio-economic vulnerabilities and associated impacts. However, such assessment may encounter problems, for example, in some cases quality of data may be poor, data may not be available, skill of forecasts at different scales can be low. As well, even if the climate information is complete and correct, the user may not access it, or may not understand or know how to apply it.

Indicators of climate-related risks (impacts, hazards and vulnerabilities) are often used to focus a risk assessment on the specific areas of interest for the decision maker. Indicators are values that can be monitored (and/or modelled) to assess changes in the state of a system, and are important tools for simplifying complex processes, with potentially multiple drivers and feedbacks, into useful and accessible information. Defining which indicators are appropriate for decision makers, as well as climate monitoring or projection purposes can be a complex process, and many different approaches have been adopted13,14,15,16. One of the more common approaches used for indicator-based studies uses the driving force-pressurestate-impact-response (DPSIR), pressure-state-response (PSR) or driving force-stateresponse (DSR) which organize indicators in the context of a causal chain17,18,19.

problems that society faces.

**2.2. Climate risk assessment** 

well as improving the research and education programs of the science community. Most importantly, climate working groups engage members from diverse stakeholder groups that might not otherwise interact and promote the legitimacy of the science community as a source of information and technology that is relevant to solving the wicked climate

As with the definition of climate risk, or CRM, there are many definitions of risk assessment. A common theme across most is the requirement for a process and/or technique that provides information with which to assess the key risk or risks. For example, the Society for Risk Analysis proposes that: "Risk assessment is the process of establishing information regarding acceptable levels of a risk and/or levels of risk for an individual, group, society, or

Risk assessments involve analysis techniques, methodologies and tools that have the key quality of assessing uncertainty (a common quality of risk), either quantitatively or qualitatively, and representing this as some measure of likelihood and/or probability. Climate risk assessment is used to help decision makers optimize resources for responding to climate-related disasters and reducing risks and impacts associated with current and future-projected climate variablity and change. It is one of the first stages of CRM, and involves identification and synthesis of hazard and vulnerability information/data that is relevant to the specific climate-related risks identified through the 'User and provider engagement and collaboration' step. One very important consideration in all climate risk assessments is the balance between the quantification of climate hazards (intensity, frequency and/or duration) and the approach to estimate the main elements of vulnerability on the ground i.e. level of exposure, poverty, exclusion, education, organizational capacity, infrastructure among others. Both hazard and vulnerability estimations may be validated using historical information of climate events and changes in socio-economic vulnerabilities and associated impacts. However, such assessment may encounter problems, for example, in some cases quality of data may be poor, data may not be available, skill of forecasts at different scales can be low. As well, even if the climate information is complete and correct,

the environment" (see: http://www.sra.org/resources\_glossary\_p-r.php).

the user may not access it, or may not understand or know how to apply it.

response (DSR) which organize indicators in the context of a causal chain17,18,19.

Indicators of climate-related risks (impacts, hazards and vulnerabilities) are often used to focus a risk assessment on the specific areas of interest for the decision maker. Indicators are values that can be monitored (and/or modelled) to assess changes in the state of a system, and are important tools for simplifying complex processes, with potentially multiple drivers and feedbacks, into useful and accessible information. Defining which indicators are appropriate for decision makers, as well as climate monitoring or projection purposes can be a complex process, and many different approaches have been adopted13,14,15,16. One of the more common approaches used for indicator-based studies uses the driving force-pressurestate-impact-response (DPSIR), pressure-state-response (PSR) or driving force-stateClimate risk assessments typically include statistical analyses of historical climate indicator records and assessment of information on climate-sensitive impacts, together with understanding of the climate mechanisms and the cascade of processes leading to these impacts. Geographical information and mapping may also be used to assess the zones where impacts are recurrent and are associated with human losses and/or infrastructure damages. Temporal changes of impacts and their related climate hazard characteristics are also often a key part of climate risk assessment, and may be directly linked with social, economical or environmental variables which may change exposure and resilience. In an ideal world, there would be millions of meteorological stations contributing to development of an accurate idea of the historical evolution of climate variables. The reality is, however, that there are not enough meteorological stations, the available stations are not evenly distributed in time and space, available data are not always digitized or shared, and there can be problems in some cases in the quality, completeness or homogeneity of the available data. Notwithstanding such issues, individual climate records measured at specific places integrate the history of the complex interactions between land, air, sea, ecosystems, and community in those locales. The final result is expressed in those climate records and consequently the history of this whole interaction process is reflected in time series of the measured values, or of their departures from a chosen reference period.

Analysis derived from climate indices/indexes, such as that provided by the CCl/CLIVAR/JCOMM Expert Team (ET) on Climate Change Detection and Indices (ETCCDI) (see http://www.clivar.org/organization/etccdi/etccdi.php) is a powerful tool that can be applied at local level. In places where high quality climate records (preferably long period, minimal gaps, and homogeneous) are available, the information that can be delivered through such climate analysis is absolutely useful, in conjunction with social and economic and other information for that locale, in DRR, adaptation and CRM processes. Such information is more accurate and in most cases more appropriate than that generated by downscaled models, but in zones where there are no available stations, information generated by downscaling is the next best alternative.

Climate risk assessment may also include a future element, utilising climate change projections and/or 'what-if' scenarios to explore the potential impacts of future scenarios of change. Practical obstacles to using information about future conditions are diverse, ranging from limitations in modeling climate system complexities (e.g. projections having coarse spatial and temporal resolution, limited predictability of some relevant variables, at scales that matter for decision making and forecast skill characterization), to procedural, institutional, and cognitive barriers in receiving or understanding climatic information, and the capacity and willingness of decision-makers to modify actions20,21. In addition, functional, structural, and social factors inhibit joint problem identification and collaborative knowledge production between providers and users. These include divergent objectives, needs, scope, and priorities; different institutional settings and standards, as well as differing cultural values, understanding, and mistrust22,23.

Improving Climate Risk Management at Local Level –

Techniques, Case Studies, Good Practices and Guidelines for World Meteorological Organization Members 487

exposure and susceptibility both increase vulnerability, increases in coping capacity reduce it. This approach mixes physical exposure (i.e. the presence (location) of people, livelihoods, environmental services and resources, infrastructure, or economic, social, or cultural assets in places that could be adversely affected by physical events and which, thereby, are subject to potential future harm, loss, or damage) with the social determinants of vulnerability. According to the most recent IPCC-SREX Report (2012)22, the IPCC describes vulnerability as the propensity or predisposition to be adversely affected. Every location on the planet has its own vulnerability profile and a specific evolution pattern. Historically this vulnerability pattern can be approximated with some social or economical indicators, statistics of disaster or land use multi-temporal comparison. The central focus of climate risk assessment is to understand the relevant climate hazards and their evolution over time, together with the vulnerabilities and how these have evolved, and a likely to change in the future, in a particular area. It is not possible to implement CRM or adaptation actions only with climate scenarios. This information must be complemented with the estimation of current

No single, consistent approach for conducting risk assessments has emerged, instead a range of different techniques have been used24. The choice of a particular technique is influenced by several factors, including: the goal of the assessment, the exposure units to be studied (an exposure unit is defined as the sector, location or activity being assessed), availability of data, choice of models suitable for the projection of future outcomes, and the time frame involved. A major challenge for future climate change assessments is the uncertainty associated with future projections and the propagation of this uncertainty throughout an impact assessment3. One approach has been to give a range of uncertainty bounded by low and high scenarios of climate change. However the outcomes of such

An example of a successful climate risk assessment for the agricultural areas in the highlands region of Ecuador was developed in 2011 by the International Research Center on El Niño (CIIFEN), Ecuador. This was requested by the Ministry of Environment as part of the National Plan for Adaptation. For the assessment, agricultural areas were identified based on up-to-date satellite information, and specific field verification. Information and indicators for agriculture aptitude, erosion, hydrological deficit, level of access to water for irrigation, type of soil, were considered, and social and economical indicators were selected. All information was analyzed spatially at parish level and combined to produce a vulnerability map covering the Ecuadorian highland region. This was further combined with historical climate hazard maps of "dry consecutive days" and "high temperature indexes", as reported in the Second National Communication of Ecuador to the UNFCCC, 200925. The resultant map of the climate risks for the agricultural sector in the highlands of Ecuador (Fig. 1) is currently used by national and local authorities to assign priorities, allocate resources and address the key elements involved in the vulnerability of the agriculture sector to cope with the potential climate hazards

vulnerability and potential future evolution.

based in the historical trends.

analyses may be too broad for planning effective adaptation.

**Figure 1.** Climate-related risk map for agriculture in the highlands of Ecuador. Color scale: Red: high risk; Light green: low risk; Grey: no data. Source: MAE (2012) www.ambiente.gob.ec

A fundamental part of risk assessment is related to vulnerability. However, it is very difficult to provide a unique formulation or set of indicators for vulnerability, as these will vary across sectors, geographically and in reponse to socio-economic conditions. A typical view of vulnerability considers the combination of several elements: the level of exposure (of an element which must be specified, e.g population, livelihood, infrastructure, etc.), the level of susceptibility which is a degree of how much the natural hazard can affect it, less or divided by the coping capacity of the exposed element which includes all the factors of resilience in the community, livelihood, infrastructure, etc.). While increases in the level of

**Figure 1.** Climate-related risk map for agriculture in the highlands of Ecuador. Color scale: Red: high

A fundamental part of risk assessment is related to vulnerability. However, it is very difficult to provide a unique formulation or set of indicators for vulnerability, as these will vary across sectors, geographically and in reponse to socio-economic conditions. A typical view of vulnerability considers the combination of several elements: the level of exposure (of an element which must be specified, e.g population, livelihood, infrastructure, etc.), the level of susceptibility which is a degree of how much the natural hazard can affect it, less or divided by the coping capacity of the exposed element which includes all the factors of resilience in the community, livelihood, infrastructure, etc.). While increases in the level of

risk; Light green: low risk; Grey: no data. Source: MAE (2012) www.ambiente.gob.ec

exposure and susceptibility both increase vulnerability, increases in coping capacity reduce it. This approach mixes physical exposure (i.e. the presence (location) of people, livelihoods, environmental services and resources, infrastructure, or economic, social, or cultural assets in places that could be adversely affected by physical events and which, thereby, are subject to potential future harm, loss, or damage) with the social determinants of vulnerability. According to the most recent IPCC-SREX Report (2012)22, the IPCC describes vulnerability as the propensity or predisposition to be adversely affected. Every location on the planet has its own vulnerability profile and a specific evolution pattern. Historically this vulnerability pattern can be approximated with some social or economical indicators, statistics of disaster or land use multi-temporal comparison. The central focus of climate risk assessment is to understand the relevant climate hazards and their evolution over time, together with the vulnerabilities and how these have evolved, and a likely to change in the future, in a particular area. It is not possible to implement CRM or adaptation actions only with climate scenarios. This information must be complemented with the estimation of current vulnerability and potential future evolution.

No single, consistent approach for conducting risk assessments has emerged, instead a range of different techniques have been used24. The choice of a particular technique is influenced by several factors, including: the goal of the assessment, the exposure units to be studied (an exposure unit is defined as the sector, location or activity being assessed), availability of data, choice of models suitable for the projection of future outcomes, and the time frame involved. A major challenge for future climate change assessments is the uncertainty associated with future projections and the propagation of this uncertainty throughout an impact assessment3. One approach has been to give a range of uncertainty bounded by low and high scenarios of climate change. However the outcomes of such analyses may be too broad for planning effective adaptation.

An example of a successful climate risk assessment for the agricultural areas in the highlands region of Ecuador was developed in 2011 by the International Research Center on El Niño (CIIFEN), Ecuador. This was requested by the Ministry of Environment as part of the National Plan for Adaptation. For the assessment, agricultural areas were identified based on up-to-date satellite information, and specific field verification. Information and indicators for agriculture aptitude, erosion, hydrological deficit, level of access to water for irrigation, type of soil, were considered, and social and economical indicators were selected. All information was analyzed spatially at parish level and combined to produce a vulnerability map covering the Ecuadorian highland region. This was further combined with historical climate hazard maps of "dry consecutive days" and "high temperature indexes", as reported in the Second National Communication of Ecuador to the UNFCCC, 200925. The resultant map of the climate risks for the agricultural sector in the highlands of Ecuador (Fig. 1) is currently used by national and local authorities to assign priorities, allocate resources and address the key elements involved in the vulnerability of the agriculture sector to cope with the potential climate hazards based in the historical trends.
