During my fieldwork campaign, I was not able to answer these questions, but numerous observations provided mixed indications of the nature and importance of ethnic networks in the region. Therefore, simply using language or membership in an ethnic group in the region might be an inadequate indicator, and such networks among farmers need to be analyzed further. Reducing vulnerability is a challenge for decision makers seeking to reduce social vulnerabilities in adapting to change. These efforts can take two approaches based on their timing: proactive or reactive. Proactive approaches, also known as anticipatory, are those actions taken in anticipation of the expected changes . Reactive approaches, also called responsive, are those action taken once damage materializes . The optimal approach involves including both approaches; however, many efforts are still only reactive and focused on short-term solutions. In this section, I present the efforts I found in the region and identify which are only reactive, only proactive, or a combination of both. Reducing weather-related impacts by anticipating an emergency requires the availability of emergency personnel, disaster risk plans, and maps, among other things. The regional government created a summary in 2016 of the resources available per municipality in various disaster risk management categories. A municipality requires a disaster management plan to reduce its exclusive use of reactionary approaches when such events occurs; however, 38% of Puno’s municipalities do not possess such plan . Some municipalities possess only partial plans for emergency management. During my fieldwork campaign, decision makers commented on the possibility of suspending mayors from a number of months if they don’t provide such plans by a certain deadline every year. Encouraging local decision-makers to create emergency management plans is imperative,vertical farming but the effects of such enforcements on other aspects of the community must be thought out.
Another resource to reduce vulnerability and manage a disaster is to make maps of such events available. However, 48% of the municipalities lack such maps for decision making. Figure 21 presents the locations of municipalities with complete or incomplete sets of maps. Other resources relating to disaster management and prevention involve the availability of working groups, civil defense platforms, and emergency operation centers. Almost half the municipalities do not possess an emergency operation center . Furthermore, 27% of municipalities do not have a working group dedicated to disaster risk management. Figure 21 shows that most of these municipalities are in the northwestern and southeastern corners of the region. Many countries possess organizations with government support that aim to support populations living in vulnerable areas as they try to cope with disasters. In Peru, these are called defensa civil or civil defense. In Puno, 17% of the municipalities are without a civil defense platform, as shown in Figure 21 , and many are in the same locations that lack working groups. The Ministry of Economy allocates a budget for expenditures related to reducing vulnerability and attending to emergency disasters. This budget is aimed at reducing the vulnerability of the population in the face of hazard threats. Numerous government agencies can make use of the budget for their interventions, including the Ministry of Agriculture; Ministry of Housing, Construction and Sanitation; Ministry of Transportation; Ministry of Health; Ministry of Education; INDECI; and regional and local governments.This type of budget for the department of Puno starts with 7,371,945 nuevos soles , but it changes every year according to national budget allocations. However, approaches are typically reactive when you look at how provinces use their hazard budget . During 2013, the province of Azangaro had the largest budget and used the highest percentage of its assigned budget , while Huancane used the lowest . During 2014, the province of Moho used its entire budget while the province of El Collao did not use any of its budget.
During 2015, higher percentages were used; the province of Lampa used the highest percentage of the money allocated. Putina used the lowest percentage that year. For unknown reasons, the 2016 budget information is available only until September of that year. The use of these budgets suggests that a reactionary approach dominates the region. Plans for managing risk and reducing vulnerability are increasingly common over the years; each year, reports about management of frost-related impacts involve a significant number of humanitarian aid and communication campaigns to educate the population. Education campaigns are useful for those families that are not necessarily wealthy but possess the resources to follow what the campaigns suggest. However, many farmers do not possess the resources to take the precautionary methods suggested by the campaign. Furthermore, the struggle to create precautionary approaches to weather hazards is higher at certain levels of management. For example, as mentioned above, the regional authorities were thinking about suspending mayors that would not complete a disaster risk management plan. This approach would incentivize some mayors to comply but at the same time would hurt other administrative responsibilities in the municipality. While the region is improving its plans to prevent weather-related impacts, however, precautionary approaches—even inside the plans— heavily rely on reactionary approaches. Why is there such a heavy balance toward reacting, and what will Puno need to avoid the perennial return of frost hazards and decreasing water availability in the region? Operationalizing vulnerability is considered a priority for supporting climate risk decision-making . However, understanding vulnerability as a comparative metric diminishes without a visualization methodology . The heterogeneity of variables and of goals in interpreting vulnerability have led to interest in visualizing vulnerability through mapping techniques. However, literature explicitly discussing vulnerability mapping and its influence on public policy is relatively scant. Vulnerability maps potentially allow us to identify and locate vulnerable populations . The technique’s objective is to communicate ‘vulnerability of place,’ which aids spatial planning and helps educate the public about climate interactions with human/environment systems .
Maps provide a common ground for discussion and communication among stakeholders . Measuring and mapping vulnerability is challenging. An array of concerns commonly emerges, mainly related to limited data availability and various methodological issues . There is an intrinsic trade-off between the wealth of information reflecting real-world complexity and the need to be able to communicate and use that information for policy-making and informing the public . In the context of climate variability and change, vulnerability mapping has proliferated since the release of the IPCC AR4 . This 4th assessment used the outcome vulnerability approach that is most popular for assessing vulnerability in the climate change community. However, more robust mapping techniques to assess social vulnerability emerged from the contextual vulnerability interpretation used in the IPCC AR5. Therefore,flood tray vulnerability maps increasingly focus on the spatial patterns of social-science constructs, such as the capitals I use in this research. In recent decades, there has been a well-documented revival of interest in spatial social research . Spatial social science recognizes the crucial role that spatiality plays in human society and encourages the understanding of spatial patterns and processes . In a sense, many social scientists have discovered geography. For geographers, scale is a key concept that aids in understanding spatial patterns and processes. The scale concept involves the spatial units at which we observe and characterize patterns, entities, and processes. Literature, particularly in geography, discusses the importance of scale in both lay and scientific representations of the world. The concept of scale includes three domains: thematic, temporal, and spatial. Here, I focus on spatial scale. Problems with scale and resolution are well-known in the geographic literature. The Modifiable Areal Unit Problem is a typical expression of this problem. MAUP is the “geographic manifestation of the ecological fallacy in which conclusions based on data aggregated to a particular set of districts may change if one aggregates the same underlying data to a different set of districts” . Scientific concern with MAUP can be traced back to the mid-1930s with a study by Gehlke and Biehl. This study of male juvenile delinquency in Cleveland, Ohio, showed that the correlation coefficient varied with the scale of aggregation. However, the MAUP term was not coined until 1979 when Openshaw and Taylor worked with Iowa’s electoral data. They coined the term MAUP to label the inconsistencies they found in their results with different spatial configurations. MAUP entails two types of problems: scaling and zoning. The scaling problem, also called the aggregation problem, is that using data aggregated at different spatial scales can result in different spatial patterns of a variable. Spatial scale affects statistical analyses insofar as data aggregated more tend to inflate correlations as compared with less aggregated levels .
The zoning or grouping problem is more difficult to understand than the scaling problem. The zoning problem concerns the effect of zone shape and location on spatial patterns of the data. The two MAUP components may present errors affecting the validity of the results. There is no generally straightforward solution to MAUP problems; they may not even be problems so much as they are an expression of the fact that geographic reality varies with different spatial units. It is imperative to recognize the threat of scaling and zoning problems when assessing and mapping vulnerability. The highest level of generalization in the study happens at the municipality level, which possesses the highest level of data aggregation in the study. Figure 23 presents the composite index for social vulnerability with data aggregated at the municipality level. The northwestern area, as well as various municipalities surrounding Lake Titicaca, appear to have the highest levels of social vulnerability . In developing countries, the municipality level is the most common spatial scale for mapping vulnerability due to data availability. However, while information at the municipality level might help national authorities, it is not optimal for decision-makers at more local levels. This level of aggregation does not allow decision-makers to find the right areas for intervention. Assessments at the census unit and farm levels could aid in finding such sensitive populations. The index aggregated at the agricultural census unit level presents a different image for social vulnerability. The southeastern area that borders with Bolivia and the end of the Altiplano include some of the worst cases for social vulnerability. But most of the highly vulnerable cases are in the northern areas above Lake Titicaca. Social vulnerability at the census unit level appears to be spatially associated with topography at the border of the Altiplano.Figure 25 presents a comparison with the topography of the region. The end of the Altiplano is not the area with the highest elevation, and fluctuation in altitude is not found throughout the entire vulnerable area. Since only one of the 22 indicators involves topography in this study, the clear delineation of the Altiplano border raises a question. What conditions cause socio-economic indicators to reflect the topography of the region? We cannot answer this question in this study, but we can identify it as needing more investigation. Further research should explore the reason behind such spatial association. If the census unit level provided more information than the municipality level, can the farm level provide even more information? Visualizing vulnerability at the farm levels is difficult since the coordinates overlap with each other in many cases. Therefore, I visualize the farm analysis with maps representing the percentage of farmers that are most vulnerable. Figure 26 presents the percentage of farmers inside the administrative region that are part of the worst quartile . Some census units show up to 100% of their farmers in this category. However, 80% is the highest proportion of most vulnerable farmers inside an administrative division at the municipality level. Spatial statistics aid in further understanding what is happening at the farm level. Spatial analysis at the farm level indicates that the locations at the end of the Altiplano possess statistically significant clusters of highly vulnerable farms. The analysis reveals that 60%, and in many cases above 80%, of the farmers inside those census units are part of the vulnerable clusters. Contrary to other levels, the farm level index did not present visible patterns visible without performing spatial statistics analysis. Furthermore, the outliers in the data are minimal and dispersed through the region.