Geomatics, Natural Hazards and Risk
ISSN: 1947-5705 (Print) 1947-5713 (Online) Journal homepage: http://www.tandfonline.com/loi/tgnh20
Exposure to climate and climate change in Mexico Alejandro Monterroso & Cecilia Conde To cite this article: Alejandro Monterroso & Cecilia Conde (2015) Exposure to climate and climate change in Mexico, Geomatics, Natural Hazards and Risk, 6:4, 272-288, DOI: 10.1080/19475705.2013.847867 To link to this article: http://dx.doi.org/10.1080/19475705.2013.847867
© 2013 Taylor & Francis
Published online: 31 Oct 2013.
Submit your article to this journal
Article views: 318
View related articles
View Crossmark data
Citing articles: 1 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=tgnh20 Download by: [37.44.207.110]
Date: 20 January 2017, At: 22:13
Geomatics, Natural Hazards and Risk, 2015 Vol. 6, No. 4, 272–288, http://dx.doi.org/10.1080/19475705.2013.847867
Exposure to climate and climate change in Mexico ALEJANDRO MONTERROSO*y and CECILIA CONDEz yDepartamento de Suelos, Universidad Autonoma Chapingo, Km 38.5 carretera Mexico Texcoco, Chapingo, Texcoco 56230, Mexico zCentro de Ciencias de la Atmosfera, Universidad Nacional Autonoma de Mexico, Av. Universidad No. 3000 Col. Universidad Nacional Aut onoma de Mexico C.U. Delegaci on Coyoac an, D.F. Mexico, Distrito Federal 04510, Mexico (Received 20 July 2013; accepted 18 September 2013) An index with the potential to integrate different climate hazards into a single parameter is required to guide preventive decision making. We integrated in a single index the degree of exposure to climate that the nation’s municipalities have. We selected this spatial scale because the municipality is the basic unit of administrative and economic planning; consequently, this is the scale at which policies of adaptation to climate change must be fostered. We conceptualized exposure as the sum of historic extreme events, the degree of ecosystem conservation and current climate and its future scenarios. This approach allowed us to create a climate hazard exposure index at the municipality scale integrating past and present. Maps of this index can be constructed to serve as a medium of risk communication and to aid policy design. We used information from eighteen variables to statistically standardize and compute the hazard exposure index by applying empirical formulae. We found that actually, out of ten Mexicans, three live in flood-prone zones, three may suffer the passage of tropical cyclones, five reside in drought zones and two live in extreme drought regions. Additionally, hailstorms affect five out of ten Mexicans, while eight out of ten are affected by frosts. Incorporating climate change, in the future more municipalities and a higher population will live in high exposure. Because understanding exposure is a necessary prerequisite to understanding vulnerability, knowledge of the spatial distribution of exposure should be useful for reducing the identified climate hazard exposure and vulnerability to climate change.
1. Introduction Understanding exposure to climate change is a fundamental prerequisite to understanding vulnerability to climate change. Vulnerability to climate change is “the degree to which a system is susceptible to, and unable to cope with, adverse effects of climate change, including climate variability and extremes” (IPCC 2007). The elements that define a system’s vulnerability are its exposure, its sensitivity and its adaptive capacity, including both an internal and external dimension (McCarthy et al. 2001; Parry et al. 2007). The internal dimension of vulnerability is represented by the sensitivity and adaptive capacity of the system. The internal dimension is related to the defence and security capacity – the capacity to anticipate, confront, resist and
*Corresponding author. Email:
[email protected] Ó 2013 Taylor & Francis
Exposure to climate and climate change in Mexico
273
recover from a certain impact or damage. It has been argued that the internal or occupancy processes of system characteristics, such as land use, social, economic, cultural, political or environmental factors, may be the determinants of vulnerability (Smit 2006). The external dimension of vulnerability refers to the exposure and risk to a certain phenomenon or stressor. The definition proposed by the Intergovernmental Panel on Climate Change (IPCC) for exposure states that it is “the nature and degree to which a system is exposed to significant climate variations.” Exposure is related to the climate stress level of a certain analysis unit or system (O’Brien et al. 2004), the danger (stressor) and the nature and extent of change in climate variables. Exposure may be represented as changes in long-term climate conditions or by changes in climate variability, including the magnitude and frequency of extreme events. It has been proposed that evaluating exposure, the external dimension of vulnerability, is possible by using biophysical indicators or interest variables (Luers et al. 2003). The method of interest variables may be applied by measuring a group of variables previously defined for a determined stressor, while the exposure will be defined in terms of changes that may occur in these variables. A variant of this method is the probability that a variable surpasses a certain threshold (Mansury & Healy 2002). The method that uses indicators has been the most widely used to quantify the components of vulnerability; it has been applied to understand the physical and socioeconomic factors that contribute to vulnerability (Hebb & Mortsch 2007; Monterroso et al. 2013). The evaluation by indicators may be applied on any scale (e.g., home, municipality, region and country), although its main limitation is the inability to capture the temporal and social complexity of the systems evaluated. Additionally, the application of indicators is limited by the subjectivity associated with their selection and assigned weights, availability of information and difficulty in proving or validating results (Luers et al. 2003). However, the method is valuable because of its capacity to monitor tendencies and explore conceptual models. Biophysical evaluations are also known as impact evaluations (dependent on the sensitivity) and are mainly focused on the physical impacts of climate variability or climate change, such as extreme events or future changes in climate variables, on different attributes (F€ ussel & Klein 2006). Examples of the use of global-scale indicators are the Human Development Index from the United Nations (UNDP 1990), the Global Food Security Index (Downing et al. 2001) and the Index of Human Insecurity (Lonergan 1998). Examples of regional-scale indicators are the vulnerability maps from The Energy Research Institute (TERI 2003), vulnerability due to globalization (O’Brien et al. 2004) and indicators proposed by Luers et al. (2003). Heltberg et al. (2009) used indicators classified as direct and indirect indicators of risk from climate change and the high frequency of extreme events. Other similar studies are those by Cuttler (1996), Cuttler et al. (2000), Brenkert and Malone (2005), and Aymone and Ringler (2009), and the study for Mexico by Monterroso et al. (2013). However, there is currently no systematic methodology to operationalize the evaluation of exposure and vulnerability to climate change, especially within the context of multiple stressors (O’Brien et al. 2004). Previous research provides two approaches for applying indicators methods. The first approach assumes that all indicators have the same importance and the same
274
A. Monterroso and C. Conde
weights are assigned to each of them. The second approach assigns different weights to the indicators to avoid the uncertainty inherent in assigning all weights equal values, especially given the diversity of indicators used. There are many different forms of the second approach, including methods such as expert judgment (Kaly et al. 1999; Kaly & Pratt 2000), principal component analysis (Easter 1999), correlation with past events (Brooks et al. 2005) and fuzzy logic (Eakin & Tapia 2008). Climate is a multidimensional phenomenon that is manifested at different spatiotemporal scales; therefore, Burton (1997) proposed three distinct climatic levels: climate variables (temperature), climate events (a storm) and long-term climate processes (anthropogenic climate change). F€ ussel and Klein (2006) indicate that to evaluate exposure the aforementioned factors must be considered as well as the sitespecific circumstances. In addition, they note that most studies have focused on longterm climate processes (e.g. mean annual temperature, mean annual precipitation and sea level) because data for these processes are more easily accessible and the results of these studies are rapidly incorporated into models of potential impacts. Few studies have evaluated climate exposure from the perspective of climate change in Mexico. For this reason, the objective of this study is to provide analyses of past and future climate hazard exposures using the municipalities in Mexico as the study area. To achieve this objective, we evaluated the past exposure of the country’s municipalities and studied future exposure by considering climate change scenarios. We performed this study by using indicators that allowed us to identify regions of Mexico that have been most exposed to climate variability. The level of climate stress was evaluated by considering the frequency with which certain extreme phenomena (related to precipitation and temperature) have occurred in recent years over a particular geographic area (in this case at the municipality level of Mexico). We also considered that a stressor can be evaluated by using possible future states of the variables (temperature and precipitation) under climate change scenarios. We propose that, to estimate climate hazard exposure, both the past occurrence and future scenarios of the relevant variables must be considered.
2. Data and methods We looked for historic indicators with which to characterize the climate stress that municipalities in Mexico have experienced in recent years. We grouped eighteen variables into three classes: extreme events, socioeconomic and climate change. Information for each municipality in Mexico (a total of 2455 municipalities in the country) was obtained from the National Center for Disasters Prevention (CENAPRED 2010), the municipal observatory (SEDESOL 2010), the Inventory System of the Effects of Disasters (DESINVENTAR 2012) and the Atmospheric Sciences Center of the National Autonomous University of Mexico (UNAM – UNIATMOS; Ò Fernandez et al. 2009). Geographical analyses were performed with ArcGIS (verÒ sion 10). Statistical analyses were performed with PASW Statistics (version 18). The methodology is described in the following sections.
2.1. Variables Table 1 shows information concerning the variables used in this study. The eight indicators of extreme events were selected using the hypothesis that the
Exposure to climate and climate change in Mexico
275
Table 1. Names, abbreviations and definitions of the variables. Variable
Definition Extreme events (1985–2011) Total waterlogging and floods in a municipality Total hailstorms (Hail is frozen water that falls violently from clouds as more or less hard and thick pellets.) Total frosts, such as a lowering of temperature with harmful effects on people, crops, goods and services Total intense, persistent or torrential precipitation events in a specific region Total extreme temperature increases in a region with effects on people, crops, goods and services Total droughts, such as a dry season without rainfall or with a rainfall deficit for prolonged periods Total atmospheric static discharges (lightning) with effects on people, animals, goods or infrastructure Total mass movements of the land surface, displacement, detachment and slope collapses
E1
Floods
E2
Hail
E3
Frost
E4
Heavy rains
E5
Heat waves
E6
Drought
E7
Thunderstorms
E8
Landslides
E9
Environmental problems, 2005
E10 E11 E12
Socioeconomic Total of five problems (illegal logging, fires, pests, biodiversity loss and water pollution) Current surface without use, Municipality surface area without 2005 vegetation or use (%) Agricultural losses, 2007 Agricultural surface area in a municipality (%) with damage caused by climate events Population change Estimated population change (%) by 2030 compared to 2000
E13 E14
Temperature, 1950–2000 Temperature, HadGEM1
E15
Temperature, ECHAM5
E16 E17
Precipitation, 1950–2000 Precipitation, HadGEM1
E18
Precipitation, ECHAM5
Climate change Mean annual temperature ( C) Future temperature change (DT), Hadley Centre Global Environmental Model (HadGEM1) for scenario A2 by 2030 Future temperature change (DT), MPI ECHAM5 for scenario A2 by 2030 Mean annual precipitation (mm) Future precipitation change (%), HadGEM1 for scenario A2 by 2030 Future precipitation change (%), MPI ECHAM5 for scenario A2 by 2030
municipalities that have presented one or more of these phenomena in recent years show more exposure to the frequency of extreme events. Municipalities with a high frequency of intense precipitation, floods, frosts and hailstorms are more exposed and are most vulnerable.
276
A. Monterroso and C. Conde
The socioeconomic variables account for the presence of environmental and socioeconomics problems, such as illegal logging, wildfires, pest and disease outbreaks, biodiversity loss, water contamination, the surface of the municipality without vegetation, agricultural economic losses related to climate and population percent change predicted by 2030. These variables attempt to integrate the nonclimate problems that cause a municipality to be more exposed to changes in its normal climate conditions. The second group of indicators represent baseline and future climate conditions projected to 2030 under the IPCC A2 emissions scenario (Nakicenovic et al. 2000). Climate change scenario outputs from two models were included to illustrate the uncertainty associated with exposure to climate change. Although the information represents annual behaviour, we used it as a proxy for what climate variability may occur. We hypothesize that it is possible to spatially identify municipalities that show differences with respect to the climate variables observed, making it possible to designate them as the municipalities most exposed to climate change. In our study, we did not include the evaluation of economic or human losses due to climate events because these factors are outside the scope of our objectives. However, the human losses and damage to infrastructure that have been observed due to these phenomena are substantial. Another limitation of this study is that we did not consider a municipality’s elevation above sea level despite its importance to the subject of exposure and climate change. 2.2. Formulation of the hazard exposure index In this study, we assessed the exposure of a system to three factors: event frequency, current environmental and socioeconomic conditions and degree of future impacts. We designed the formula so that the maximum value for each variable reflected the maximum severity in exposure. The variables were standardized but were not weighted: Z ¼ ðXi X Þ=DS;
ð1Þ
where Z represents the standardized value, Xi is the observed value, X is the mean of the set of values i and DS is the standard deviation of the set of values i. We calculated a value for each dimension we studied: extreme events (EE), socioeconomic factors (SE) and climate (CC). Next, the three dimensions were integrated in an empirical hazard exposure index: Hazard exposure indexðEIÞ ¼ ðEE þ SE þ CCÞ:
ð2Þ
The climate change scenarios were constructed by replacing the climate values (CC) with those obtained from the MPI ECHAM5 model and the HadGEM1 model. Therefore, we obtained one base scenario and two climate change scenarios. Each scenario was normalized to values between 0 and 100 using the following formula: N ¼ ðXi Xm Þ=ðXM Xm Þ;
ð3Þ
Exposure to climate and climate change in Mexico
277
where N is the normalized value between 0 and 100, Xi is the observed value, Xm is the minimum value observed and XM is the maximum value observed in the set of data i.
2.3. Mapping the hazard exposure indexes Once the current and future hazard exposure indexes were constructed, each municipality in Mexico was assigned a calculated value, which allowed prioritization of municipalities by exposure degree. The final range of values was classified into five groups according to a geometric distribution of the frequencies, and a qualitative indicator of exposure severity was assigned to each group: very low (0–20), low (21– 40), medium (41–60), high (61–80) and very high (81–100). The results were integrated in a database of 2455 municipalities in Mexico, and it was possible to map the hazard exposure index and severity at the municipal level.
3. Results and discussion 3.1. Climate risks Given its geographic location, Mexico is highly susceptible to hydrometeorological events. Perhaps because of this susceptibility, the highest observed exposure is to precipitation variables as opposed to exposure related to temperature variables (table 2). Additionally, statistics show an increase in recent years in the tendency for hydrometeorological events to occur (WorldBank 2012). The Gulf of Mexico and Caribbean regions of Mexico have been most impacted by floods, although in recent years it has been observed that these also occur in central Mexico. Tropical cyclones are the main cause of floods and occur during the Table 2. Statistics of the variables used in this study (all data from 2455 municipalities in Mexico). Variable Floods Hail Frost Heavy rains Drought Heat waves Thunderstorm Landslides Environmental problems Actual land without use Agricultural losses Population change Mean temperature ( C) Mean precipitation (mm) Temperature HadGEM ( C) Precipitation HadGEM (mm) Temperature ECHAM ( C) Precipitation ECHAM (mm)
Minimum
Maximum
Mean
Standard deviation
0 0 0 0 0 0 0 0 0 0 0 68.6 10.8 2 12.0 2 11.6 1
82 11 54 25 18 39 5 36 5 61.3 98.8 425.5 30.6 656 32.3 609 31.9 607
3.04 0.21 0.86 0.90 1.22 0.33 0.07 0.60 2.33 0.91 43.83 2.99 22.18 154.60 23.66 150.38 23.34 135.22
6.85 0.74 2.47 1.92 2.19 1.62 0.33 2.09 1.53 2.53 23.03 42.37 4.49 80.51 4.48 80.65 4.46 75.97
278
A. Monterroso and C. Conde
rainy season (i.e., from June to November). Hailstorms also occur during the same months of the rainy season and are more frequent in the central and northern portions of Mexico. In contrast, the northern portion of Mexico is the region most affected by droughts; however, drought reports were also found for the central and southern portions of the country, where in the past droughts were not observed. Frosts are more frequent in the central and northern portions of Mexico. Damage from frost is manifest mainly as agricultural losses. Although heat waves are not common, reports in recent years indicate events have occurred in some municipalities. These factors show that the documented hazards are due to the high climate variability that historically has included diverse floods, droughts, frosts and hailstorms. The variables that we considered do not differ from those used by other researchers. For example, in the World Risk Report (Beck & Shepard 2012), the authors considered earthquakes, cyclones, floods, droughts and sea level rise. In the study, Mexico is categorized with a medium climate hazard exposure, occupying the thirtieth place out of a total of 173 countries. Additionally, the Disaster Risk Index (Peduzzi et al. 2009) used the same climate risks but also considered the possible human losses. In the study, Mexico is categorized in class five out of seven for mortality risks.
3.2. Climate hazard exposure index As indicated above, we integrated the fourteen variables according to equation (2) and obtained a climate hazard exposure index for the municipalities of Mexico. The index included all variables without those related to climate change scenarios. We obtained positive correlations between the index and the dimension of frequency of extreme events (0.57), socioeconomics factors (0.74) and the current climate (0.63) (figure 1). The correlation coefficient shows a measure of the strength of linear dependence between two variables. For the case of the first dimension, we attributed the low correlation value to the fact that the indicator accounts for only the event occurrence frequency and not its severity. A large number of municipalities in central Mexico reported zero impact from this type of phenomenon, but we know that there have been recent reports of higher frequencies. Given that, for the three dimensions, a higher frequency indicates a higher climate hazard exposure, thus the index was considered valid. Figure 2(a) shows the spatial distribution map of climate hazard exposure in Mexico. The figure shows that it is possible to map the climate hazard exposure in a country by integrating a few basic variables. A map displaying the frequency of extreme climate events can provide a country with information that allows its policymakers to more effectively locate and provide assistance to affected populations. For example, with our index, we observed that 1506 municipalities in Mexico have very low climate hazard exposure, while one municipality has a very high exposure and twelve more have high climate hazard exposure. The states with municipalities that present the highest values of climate hazard exposure are Quintana Roo, Chiapas, Guerrero and Mexico City. The frequency of the events studied indicates that nearly the entire country has been exposed to their occurrence. However, the northern states are affected by a higher recurrence of such phenomena, mainly droughts and frosts. The Southeast is
Figure 1. Dispersion data per exposition group: (a) extreme events, (b) socioeconomic factors and (c) climatology, related to the hazard exposure index.
Exposure to climate and climate change in Mexico 279
280
A. Monterroso and C. Conde
Figure 2. Climate hazard exposure classes for (a) the base scenario, (b) the HadGEM1 climate change scenario and (c) the ECHAM5 climate change scenario. The climate change scenarios are for the year 2030 with scenario A2.
Exposure to climate and climate change in Mexico
281
the region where heavy rains and even floods occur. The central and western states have a lower recurrence of two or more extreme events, although the severity of past events has been significant. Figure 2(b) and 2(c) show the spatial distribution map obtained from the climate change scenarios: it is possible to observe the increase of climate hazard exposure due to future changes. Regions exposed to a variety of simultaneous climate phenomena were grouped in the hazard exposure index. Mapping and spatial overlay of climate events are useful tools for identifying the location of the most exposed zones. By using these techniques, we identified the extent that each variable contributes to the final classes of climate hazard exposure. For example, we found that the variables that contribute less to the extreme events dimension were hail and thunderstorms. In contrast, the variables that contributed the most to a higher exposure were floods, frosts, droughts, surface without apparent vegetation and the estimated population change by the year 2030 (table 3). We used temperature and mean precipitation for the municipalities in Mexico for the period 1950–2000. Municipalities with higher temperatures and precipitation were considered to have higher climate severity, although this does not necessarily refer to negative impacts. Our point of view is that climate severity is a way of measuring events associated with these variables. Therefore, we only considered the months from June to September, when most of the studied climate events occurred. For example, we assumed that as the mean temperature increases the degree of exposure increases. Consequently, we also assumed that as the mean temperature increases the maximum temperature increases. There is an approximate difference of 16 C between the very low and very high exposure classes. For precipitation, the difference is almost 500 mm in the study period. We found that the municipalities with mean temperatures greater than 27 C were exposed to some alteration in their productive systems. This alteration also occurred for those municipalities with precipitation lower than 200 mm during the months in the study period. To estimate the challenge policymakers face of providing assistance to the most exposed municipalities, we quantified how many of these are in each class. For the exposure classes very low and low, there are 1506 and 877 municipalities, respectively. For the classes of medium to very high, there are seventy-two municipalities, representing 3% of the total municipalities (figure 3). There were twelve municipalities in the high exposure class and one in the very high exposure class. The states containing municipalities with the highest exposure values are Quintana Roo, Oaxaca, Chiapas and Mexico City. We also included the municipalities’ total population for the year 2005. A total of 3,644,601 people have very high exposure, 6,719,103 people have high exposure and 21,962,429 people have medium exposure. For the two lowest classes, there are 49,628,638 people with low exposure and 20,037,227 people with very low exposure. According to our results, more than ten million people (10.2%) in Mexico have high and very high climate hazard exposure. The geographic distribution of climate hazard exposure can help in identifying people whose livelihoods are threatened by natural phenomena. This information allows for an understanding of the exposed population. Our results are similar to those found by Anzaldo (2008). We found that thirtyeight million people have this flood risk. The total population living in drought zones
Floods
38 19 4 2 1
Actual land without use
18 6 3 1 1
Exposition class
Very high High Medium Low Very low
Exposition class
Very high High Medium Low Very low
425 88 73 6 -13
Population change (%)
2 1 1 0 0
Hail
27 14 13 12 11
Mean temperature ( C)
12 4 2 1 0
Frost 7 5 3 2 1
Drought 7 3 1 0 0
Heat waves
126 335 386 522 656
29 16 14 13 12
141 333 353 609 588
HadGEM model Mean precipitation Temperature Precipitation (mm) ( C) (mm)
9 3 3 1 1
Heavy rains 14 3 1 1 0
Landslides
28 16 14 13 12
Temperature ( C)
86 311 313 607 549
Precipitation (mm)
ECHAM model
1 0 0 0 0
Thunderstorm
Table 3. Mean contribution of variables to the hazard exposure index.
4 3 2 2 2
59 45 42 35 33
Environmental Agricultural problems losses
282 A. Monterroso and C. Conde
Exposure to climate and climate change in Mexico
283
Figure 3. Number of municipalities by exposure class and scenario. The red arrows show increases with respect to the base scenario and the blue arrows show decreases.
identified in 2008 by Anzaldo et al. was forty-two million; we found this population to be fifty million. Within this exposed group, more than twelve million people are in extreme drought zones in arid and semiarid regions of Mexico. Hailstorms affect more than fifty million people in at least 400 Mexican municipalities. We found seventy-five fewer municipalities in frost zones, but these zones represent three-quarters of the national population. This population is exposed to the impacts of this phenomenon on their agricultural systems and respiratory diseases. The challenge of exposure is large given that, in just the fifteen municipalities with the highest exposure, the population in 2005 was almost five million people. Knowledge of the spatial distribution and the total number of people at risk should foster efforts towards reducing the climate hazard exposure identified.
3.3. Future scenarios To produce the future scenarios, the HadGEM1 and MPI ECHAM 5 climate change model outputs were applied to the temperature and precipitation components. We applied the model outputs to only the temperature and precipitation components within the climate dimension of our study. We kept the extreme events and socioeconomic dimensions unchanged. The ECHAM5 model estimates for mean temperature an increase of 2 C to 8 C for the analysed months and in general for Mexico. For precipitation, the model suggests a reduction of 4% to 15%. The HadGEM1 model estimates that temperatures will increase from 2 C to 5 C, while precipitation will decrease from 1% to 18%. By introducing these changes to the hazard exposure index, we obtained two future indexes (one for each model). Although we did not model the future behaviour of extreme phenomena, we think that the current results advance our understanding of
284
A. Monterroso and C. Conde
the possible future behaviour of extremes, and it may be possible to include these events with future results from the IPCC (2012). The correlations for the socioeconomic and extreme events dimensions with the hazard exposure indexes remained basically unchanged for both models. The climate dimension correlations showed the only major change. For the ECHAM5 model, the correlation changed from 0.63 for the base scenario to 0.80; for the HadGEM1 model, the correlation changed to 0.82. Although, both models indicate higher future climate hazard exposure, our indexes were calculated using information for annual climate change. It is possible that a higher severity may be obtained by using monthly data. Of the two models applied to our index, the HadGEM1 model suggests a higher future climate hazard exposure. For the base scenario, thirteen municipalities were in the exposure classes of high to very high. The HadGEM1 model shows forty municipalities in the high class and seven in the very high class. This result shows three times more exposed municipalities than the present day scenario. The ECHAM5 model suggests that twenty-two municipalities will experience the higher exposures, almost doubling the current amount. It is important to highlight that the states that show high values of the hazard exposure index are those along coasts, with those within the basins of the Gulf of Mexico and Pacific standing out. The number of municipalities in Mexico with very low exposure will decrease from the current value of 1506 to 843 and 240 according to the ECHAM5 and HadGEM1 models. This class is the only exposure class that will decrease according to the climate change models. The total number of municipalities within all other classes will increase. The municipalities that currently have high hazard exposure will most likely remain in this category in the future. To estimate the challenge of future exposure in Mexico, we also included the total population at risk. The estimated changes in population for each municipality and model are presented in tables 4 and 5. The number of municipalities that will remain unchanged in their exposure class is 860 (35% of the national total) according to the HadGEM1 scenario and 1608 (65.5%) according to the ECHAM5 scenario. The number of municipalities that will have an increase in exposure is 1585 (64.5%) according the HadGEM1 model and 814 (33%) according to the ECHAM5 model. The number of municipalities that will have a decrease in climate hazard exposure is ten (0.5%) according the HadGEM1 model and thirty-three (1.5%) according to the ECHAM5 model. Of the municipalities that will experience an increase in exposure, more than seventy million people (70% of the national total) will be affected according to the HadGEM1 model and thirty million (30%) according to the ECHAM5 model. The municipalities that will experience a decrease in their exposure represents approximately four (two) million people according to the HadGEM1 (ECHAM5) model. The HadGEM1 model produced the most alarming changes. There are thirteen million people in the municipalities that are currently identified to have medium exposure. In the future scenario, that class will change to high exposure. Less than half a million people are in municipalities that change from medium to very high exposure. Almost three million people who are currently in the high exposure class are at risk of moving to the very high exposure class. In our study, the variables and the indexes we obtained help us to understand the causes and risk distribution, attempts to control and reduce risk and improve society’s adaptive capability to climate hazard exposure.
Very low
Low
Medium
High
Very high
(1) 712,057 [100%] (4) 2,899,950 [46.5%] (2) 333,451 [2.4%] (7) 6,592,990 [51.9%] (22) 13,682,719 [34.4%] (11) 1,072,524 [1.3%]
High
(1) 2,725,286 [1.5%] (33) 9,695,474 [59.7%] (273) 26,964,651 [32.8%] (81) 2,598,379 [4.7%]
Medium
(2) 1,282,379 [3.3%] (586) 21,648,774 [64.7%] (1192) 26,812,757 [77.9%]
Low
(7) 259,601 [1.0%] (233) 2,485,855 [17.3%]
Very low
Very low
Low
Medium
High
Very high
(1) 712,057 [100%]
(1) 117,828 [0.1%]
(10) 10,854,836 [83.4%] (10) 5,831,853 [20.8%]
High
(2) 1,363,390 [16.5%] (43) 18,692,283 [69.4%] (115) 10,145,456 [14.3%] (34) 478,965 [2.7%]
Medium
(6) 469,887 [9.6%] (736) 39,374,029 [82.6%] (654) 13,657,669 [46.6%]
Low
(25) 308,237 [2.8%] (818) 17,760,357 [50.6%]
Very low
Notes: The number within the parenthesis indicates the municipality total, and the number within the brackets indicates the percentage change or no change. The municipalities that increase in exposure are indicated with italic. The municipalities that decrease in exposure are indicated with bold. The main diagonal indicates municipalities that do not change exposure.
Base conditions
Very high
Table 5. Population total by exposure class according to the base scenario and the ECHAM climate change model.
Notes: The number within the parenthesis indicates the municipality total, and the number within the brackets indicates the percentage change or no change. The municipalities that increase in exposure are indicated with italic. The municipalities that decrease in exposure are indicated with bold. The main diagonal indicates municipalities that do not change exposure.
Base conditions
Very high
Table 4. Population total by exposure class according to the base scenario and the HadGEM1 climate change model.
Exposure to climate and climate change in Mexico 285
286
A. Monterroso and C. Conde
4. Conclusions With the creation of the hazard exposure index, we showed the correlation between the degree of climate hazard exposure of municipalities and the frequency and occurrence of natural phenomena. Additionally, the study shows an analysis of climate hazard exposure based on the spatial distribution of phenomena in Mexico. In this work, we combined variables of physical exposure with socioeconomic and climate change variables to calculate current and future climate hazard exposure. The variables integrated different hazard into one value to determine the spatial distribution and frequency of hazards. With this one value we showed the spatial distribution and frequency of hazards. In the beginning of this study, we set an objective to evaluate the historic exposure of Mexican municipalities and how that exposure will change in the future by considering climate change scenarios. Our results demonstrate that this index may be used as a risk communication medium and for the design of policies meant to reduce the exposures of the most affected populations. The index can be used as an instrument of communication to help design strategies to provide effective assistance to populations through risk reduction policies. We conceptualized exposure in the context of three dimensions: extreme events, socioeconomic problems and climate. This approach allowed us to understand exposure as a result of what has happened in the past (extreme historic events) plus the degree of current ecosystem conservation (prevailing environmental problems) plus the climate and its future scenarios (climate change). This approach allowed us to create an index that integrates the past, the present and the future. The resulting maps make it possible to identify those areas that are more or less exposed to climate change. The first step to effective policy design to reduce climate hazard exposure is to have a reliable inventory of natural phenomena and exposed elements. With this information, exposure can be better evaluated in terms of spatial and temporal distributions. We mapped exposure of an empirical index that shows the spatial distribution of diverse climate phenomena. However, we acknowledge this study’s limitations. We did not take into account other important climate change variables, such as sea level increase. The different scales and comparability of variables open new questions for future research. The magnitude of and the damage caused by the occurrence of extreme events were not included in this index. These factors should also be analysed in future studies. Exposure can also be mapped in terms of individual events overlying exposed populations and areas at risk. These types of variables should be integrated in the future and deserve further investigation on how to be included. Acknowledgements To the National Science and Technology Council of Mexico – CONACYT – for the grant received and the two institutions: Universidad Autonoma Chapingo and Universidad Nacional Autonoma de Mexico.
References Anzaldo C. 2008. Migraci on interna, distribuci on territorial de la poblaci on y desarrollo sustentable. Mexico: CONAPO. Aymone GG, Ringler C. 2009. Mapping South African farming sector vulnerability to climate change and variability. In: Environmental and production technology division. Washington, DC: International Food Polity Research Institute; p. 31.
Exposure to climate and climate change in Mexico
287
Beck M, Shepard C. 2012. World risk report. Focus: environmental degradation and disasters. Germany. [cited 2012 Mar]. Available from: worldriskreport.org Brenkert AL, Malone EL. 2005. Modeling vulnerability and resilience to climate change: a case study of India and Indian States. Climatic Change. 72:57–102. doi: 10.1007/ s10584-005-5930-3 Brooks N, Adger WN, Kelly PM. 2005. The determinant of vulnerability and adaptive capacity at the national level and the implications for adaptation. Global Environ Change. 15(2):151–163. Burton I. 1997. Vulnerability and adaptive response in the context of climate and climate change. Climatic Change. 36:185–196. CENAPRED. 2010. Base de datos de declaratorias de riesgos ambientales. [cited 2010 Jul 15]. Available from: http://www.cenapred.gob.mx/es/ Cuttler SL. 1996. Vulnerability to environmental hazards. Prog Hum Geography. 20:529–539. Cuttler SL, Mitchell JT, Scott MS. 2000. Revealing the vulnerability of people and places: a case study of Georgetown County. Ann Assoc Am Geographers. 90:713–737. DESINVENTAR. 2012. Sistema de inventario de efectos de desastres. [cited 2012 Mar 3]. Available from: www.desinventar.org Downing TE, Butterfield R, Cohen S, Huq S, Moss R, Rahman A, Sokona Y, Stephen L. 2001. Vulnerability indices: climate change impacts and adaptations. New York: United Nations. Eakin H, Tapia B. 2008. Insights into the composition of household vulnerability from multicriteria decision analysis. Global Environ Change. 18(1):112–127. Easter C. 1999. Small states development: a commonwealth vulnerability index. Round Table. 351(1):403–422. Fernandez EA, Zavala HJ, Romero, CR. 2009. Atlas clim atico digital de Mexico. [cited 2009 Jun 15]. Available from: http://uniatmos.atmosfera.unam.mx F€ ussel HM, Klein RJT. 2006. Climate change vulnerability assessments: an evolution of conceptual thinking. Climatic Change. 75(3):301–329. Hebb A, Mortsch L. 2007. Floods: mapping vulnerability in the upper Thames watershed under changing climate assessment of water resources risk and vulnerability to changing climatic conditions. Canada: Canadian Foundation for Climate and Atmospheric Sciences. Heltberg R, Siegel P, Jorgensen SL. 2009. Addressing human vulnerability to climate change: toward a ‘no regrets’ approach. Global Environ Change. 19:89–99. doi: 10.1016/j. gloenvcha.2008.11.003 IPCC. 2007. Summary for Policymakers. In: Parry ML, Canziani OF, Palutikof JP, Linden PJ, Hanson CE, editors. Climate change 2007: impacts, adaptation and vulnerability. Contribution of working group II to the fourth assessment report of the intergovernmental panel on climate change. Cambridge (UK): Cambridge University Press. IPCC. 2012. Managing the risks of extreme events and disasters to advance climate change adaptation. In: Field CB, Barros V, Stocker TF, Qin D, Dokken DJ, Ebi KL, Mastrandrea MD, Mach KJ, Plattner G-K, Allen SK, Tignor M, Midgley PM, editors. A special report of working groups I and II of the intergovernmental panel on climate change. Cambridge (UK): Cambridge University Press. Kaly U, Brigugilo L, McLeod H, Schmsall S, Pratt C, Pal R. 1999. Environmental vulnerability index (EVI) to summarize national environmental vulnerability profiles. Technical report Suva, Fiji. United Nations Environment Programme (UNEP): SOPAC. Kaly U, Pratt C. 2000. Environmental vulnerability index: development and provisional indices and profiles for Fiji, Samoa, Tuvala and Vanuatu. Technical report 306. United Nations Environment Programme (UNEP): SOPAC; p. 89. Lonergan, S. 1998. The role of environmental degradation in population displacement: global environmental change and human security project. Research report 1. British Columbia: University of Victoria.
288
A. Monterroso and C. Conde
Luers AL, Lobell DB, Skar LS, Addams CL, Matson PA. 2003. A method for quantifying vulnerability, applied to the agricultural system of the Yaqui Valley, Mexico. Global Environ Change. 13:255–267. doi: 10.1016/S0959-3780(03)00054-2 Mansury G, Healy A. 2002. Vulnerability prediction in rural Pakistan. Paper presented at: The IFPRI World Bank Conference on Risk and Vulnerability: Estimation and Policy Implications; Washington, DC. [cited 2012 Feb]. Available from: www.ifpri.org/ events/conferences/2002/092302/mansuri.pdf McCarthy J, Canziani O, Leary N, Dokken D, White K. 2001. Climate change 2001: impacts, adaptation and vulnerability. Cambridge (UK): Cambridge University Press. Monterroso RAI, Conde C, Gay C, G omez D, L opez J. 2013. Two methods to assess vulnerability to climate change in the Mexican agricultural sector. Mitig Adapt Strateg Glob Change. doi: 10.1007/s11027-012-9442-y Nakicenovic N, Alcamo J, Davis G, de Vries B, Fenhann J, Gaffin S, Gregory K, Gr€ ubler A, Jung TY, Kram T, et al. (2000). Special report on emissions scenarios: a special report of working group III of the intergovernmental panel on climate change. Cambridge (UK): Cambridge University Press. O’Brien KL, Leichenko RM, Kelkar U, Venema HM, Aandahl G, Tompkins H, Javed A, Bhadwal S, Barg S, Nygaard L, West, J. 2004. Mapping vulnerability to multiple stressors: climate change and globalization in India. Global Environ Change. 14:303–313. doi: 10.1016/j.gloenvcha.2004.01.001 Parry M, Canziani O, Palutikof JP, Coautores. 2007. Resumen tecnico. Cambio Clim atico 2007: Impactos, Adaptaci on y Vulnerabilidad. APortes del Grupo de Trabajo II al Cuarto Informe del Panel Intergubernamental sobre Cambio Clim atico. Cambridge, UK. Peduzzi P, Dao H, Herold C, Mouton F. 2009. Assessing global exposure and vulnerability towards natural hazards: the Disaster Risk Index. Nat Hazards and Earth Syst Sci. 9(1):1149–1159. SEDESOL. 2010. Observatorio de Informaci on Municipal. [cited 2012 Jan]. Available from: www.observatoriomunicipal.org.mx/ Smit B. 2006. Adaptation, adaptive capacity and vulnerability. Global Environ Change. 16:282–292. doi: 10.1016/j.gloenvcha.2006.03.008 TERI. 2003. Coping with global change: vulnerability and adaptation in Indian agriculture. New Delhi: The Energy Research Institute (TERI). UNDP. 1990. Human development report. New York: United Nations Development Programme. WorldBank. 2012. Improving the assessment of disaster risks to strengthen financial resilience: a special joint G20 publication by the Government of Mexico and the World Bank. Washington, DC. [cited 2012 Feb]. Available from: www.worldbank.org
