A major aim of the HABITABLE project is to identify perceptions of climate change among migrating people and compare these perceptions to quantitative data on climate conditions. Do the connections people perceive between changes in their environments and their movements truly match? And what can we learn from the complex and sometimes unexpected relationship between perception and a measurable reality?
In her 2018 masters thesis, University of Twente graduate Dinah A. E. Ogara explored these questions as they relate specifically to drought and its role in migration decisions among (former) pastoralists in Northern Kenya. The study can be found here in full. This article presents a condensed and edited report on the research findings and connects to larger questions posed within the HABITABLE project.
Climate change and temperature rise have been associated with increasing severity, intensity, and frequency of drought (Dai 2013; Wang et al. 2018) in many parts of Africa in the recent decades (Masih et al. 2014; Spinoni et al. 2014). Droughts, referring to a dry spell relative to the area-based normal conditions (Dai 2011) are particularly frequent in Africa’s drylands, which make up 43% of Africa’s land and give a home to about 50% of Africa’s population (Cervigni and Morris 2016a). People in drylands largely depend on agriculture and livestock for their livelihoods (Cervigni and Morris 2016b; Linke et al. 2018; Stidsen 2006; Nori and Gemini 2011; Buono et al. 2016), and these sectors are particularly impacted by climatic variability (Epule et al. 2017). Impacts of drought comprise exacerbated land degradation and desertification (Barrett et al. 2006), crop yield and forage reductions (Adow 2008), animal diseases and livestock losses (Rain et al. 2011). In very affected communities drought can result in food insecurity, malnutrition, famine, and resource conflict (Raleigh and Kniveton 2012; Middleton and Sternberg 2013; Kubik 2018; Linke et al. 2018).
An essential coping strategy to these hardships is mobility, particularly seasonal migration, whereby pastoralists move to other locations to find forage and water for their livestock (Perch-Nielsen, Bättig, and Imboden 2008). Traditionally, mobility and migration have been the prime strategy for pastoralists to adapt to drought (de Bruijn and van Dijk 2003). Pastoralists also occasionally engaged in herd splitting, herd merging, and livestock sales or other livelihood diversification strategies such as the temporary adoption of hunting and gathering (Fratkin 2001), trading, farming with drought-tolerant crops, and water harvesting (Barton, Morton, and Hendy 2001; Fratkin 2001). Nowadays, however, an increase in population and remaining land pressure as well as the increased frequency and intensity of droughts, render traditional coping strategies ineffective (Magal and Wambua 2017; Mogotsi, Nyangito, and Nyariki 2011). Pastoralists are forced to change their adaptation regimes and now more often choose to migrate to nearby towns (Fratkin, Roth, and Nathan 2004).
Migration is commonly perceived as a coping strategy to reduce the increased risk from environmental hazards (Webber and Barnett 2010; IPCC 2012; Linke et al. 2018; Foresight 2011). However, the relation between drought and migration is not straightforward. Instead, rural to urban migration is often characterized by a complex interplay of push and pull factors (Foresight 2011).
Moreover, climatic variability, drought, or land degradation may have different effects on different population groups, such as the young, the elderly, women or men (Rigaud et al. 2018; Hunter and David 2009; Reckien et al. 2017; Bruno Soares, S. Gagnon, and M. Doherty 2012). Due to existing inequalities in access to resources, decision making, institutional access, household burden and exposure to violence, women are particularly impacted by climate change (Morrow 2008; UNDP 2009), which may compound their inherent difficulties and injustices to adapt, cope and respond to climate-related risks (Alston 2015; Gaillard 2010). However, the differential impacts of climate change on the migration decisions of, for example, different gender and age groups remain scanty.
Kniveton et al. (2008) propose the use of quantitative methods such as statistical regression, specifically, multiple regression analysis and agent-based modelling, as a means to disaggregate the multiple migration variables and obtain simulations of future migration patterns. The long-term assessment of environmental factors may provide a possible approach to what Gemenne (2011) terms as a key methodological caveat in estimating the relation between slow-onset environmental changes and migrations. However, this method is limited as some of the variables such as culture are unquantifiable and most significantly lacks time-sensitive migration flow data, which is crucial due to the seasonal nature of both climate and migration variables. Time-series data for such studies are therefore imperative.
This article addresses the above-mentioned research gaps, particularly the lack of i) empirically-based studies involving data of historical environmental variations and migration flows, ii) knowledge about the influence of gender and age on the migration decision, and iii) a comparison of opportunities and challenges involved in the migration decision in the source as well as destination areas. The main objective is to quantify the relationship between drought and rural to urban migration—overall, as well as differentiated by gender and age—and to assess the effects of that migration on livelihood perception. This study focuses on Marsabit Town in Northern Kenya. We answer the following three research questions:
- What are the timings, causes, and source areas of rural to urban migration in the region?
- How do these migration fluxes quantitatively relate to the severity of drought across the entire sample as well as gender and age groups?
- What are the perceived consequences of that migration on the livelihood perception of pastoralist households?
Our research takes a multi-disciplinary approach through a knowledge blend of urban and regional planning and natural resource management discourses. We investigate aspects of questions 1 and 3 with the help of interview data from field surveys (primary data sources), and question 2 by quantifying the relationship between the number of migrations and satellite-derived, spatio-temporal drought indices, i.e. relative seasonal forage scarcity proxies derived from NDVI time series. These data sources were based on a household survey and direct interactions with - now urban pastoralists currently residing in Marsabit Town, Key Informant Interviews (KII) and Focus Group Discussions (FGD) with the elderly (males and females above 65 years old at the time of the survey) to gather their life trajectories, experiences and perceptions about drought and migration patterns.
The main objective of the study is to quantitatively assess how drought contributes to rural-urban migration and the livelihood dynamics of pastoralists in Northern Kenya. The study was conducted in Marsabit town, Marsabit County, Kenya. Marsabit County is the second-largest county in Kenya spanning a total area of 70,961 km2. Marsabit town is the headquarters of Marsabit County. The town is a trading and commercial centre. It has a total population of about 14,907 people (as of 2009 census) projected to 19,179 people in 2017, at the time of the study (Republic of Kenya, County Government of Marsabit 2017).
The key questions that guided our methodology were: What are the main causes of migration? Can seasonal forage scarcity proxies derived from the NDVI time series explain the temporal peaks or dominant timings of urban migration of pastoral populations? And what are the current livelihoods and perceived quality of life before and after the urban migration of pastoral populations of Northern Kenya?
With the help of the 5 local area chiefs of each of the five administrative subloclations within Marsabit Town boundary (Marsabit Township, Sagante, Dinb-gombo, Songa, Hula Hula Karere), 295 respondents were sampled. These represent 295 pastoralist households comprising residents from all five administrative sublocations within the Marsabit town boundary as well as from different ethnic groupings (i.e. Burji, Borana, Rendille, and Gabbra). Respondents were asked about reasons for migration when they left their sprevious rural location (if possible, day, month and year of migration), as well as the effects of drought on the communities’ socio-economic well-being.
In addition to the household survey, key informant interviews, and focus group discussions were carried out. Two Focus Group Discussions, one for male and one for female participants, were conducted using a participatory Geographic Information System (GIS) process. Participatory GIS - also known as community or indigeneous mapping (McCall 2006) - was used for sketching migration routes and drought hotspots during the group discussions. Lastly, five Key Informant Interviews2 were conducted that solicited knowledge of expertise and experience related to drought, migration and urban development.
The time series of the Normalized Difference Vegetation Index (NDVI) was used to obtain an independent temporal indicator of the relative severity of drought conditions within a specific season. When comparing NDVI between different years, it can provide an indicator for reduced vegetation productivity, which in (semi-)arid areas is normally due to drought (Rojas, Vrieling, and Rembold 2011).
Both sources of the NDVI time series were translated into spatially-aggregated indices that reflect the relative seasonal forage scarcity in a spatial unit (as an outcome of drought impact) following Vrieling et al. (2016).
To create a drought index we followed method as per Vrieling et al. (2016), the forage scarcity index was computed to indicate how forage conditions for a specific season compared to the multi-annual average conditions. The method consists of a) spatially averaging NDVI per IBLI unit for each time step; b) temporally accumulating the result between the unit-specific start and end of the season, as determined from phenological analysis, resulting in a cumulative NDVI (cNDVI) that provided a relative measure of forage abundance as a result of the long rains (L: March-September) and the short rains (S: October-February); c) comparing cNDVI values (for L and S separately) between the available years of the dataset and using its mean and standard deviation to calculate a z-score (zcNDVI)3.
To assess the relationship between migration numbers and drought, the cumulative percentage of migrations was plotted against the zcNDVI values. This way we could evaluate if more migrations were reported during seasons for which zcNDVI indicated below-normal forage availability. To generate this relationship, first, we temporally counted all reported migration instances per season and year. Second, the zcNDVI was sorted from lowest to highest. Third, the migration numbers were converted to percentages of migrations for the total timeframe considered, and subsequently converted to cumulative percentages of the sorted zcNDVI. This then resulted in a graph of cumulated migration against zcNDVI, allowing us to determine what percentage of total migrations occurred below different zcNDVI thresholds.
The normal distribution curve which contains z-scores and cumulative percentages were used to interpret the cumulative percentage graphs and make inferences about the data. To test the normality of distribution of zcNDVI data, we performed the Shapiro-Wilk test4. For each data set, the test indicated no significant departure from normality. Following a normal distribution of zcNDVI, one would expect about 2 % of all migrations to occur below a zcNDVI threshold of -2—hereafter ‘severe drought’, 16 % of migrations below a threshold of -1—hereafter ‘severe and medium drought’, and 50 % below a threshold of zero—hereafter ‘severe, medium and mild drought’. If we find a higher percentage as mentioned at these thresholds, we infer that the current-season drought has affected the migration decision.
In terms of temporal peaks of migration, our study reveals that the majority of migrations (72%; 214 cases) took place during the long rain season as compared to the short rain season. Migration counts of more than ten pastoralist households are reported for 1995L, 1997L, 1998L, 1999L, 1999S, 2000L, 2009L, 2011L, and 2017L (Figure 3).
Out of the 295 respondents in our study, the vast majority (91%) reported a drought-related cause (e.g., lack of forage, rain, or water availability for livestock) as the main driver. The remaining push factors included causes such as poverty (6%) and tribal clashes and cattle raiding (3%). Additionally, resource-based conflict and human-wildlife conflict were mentioned in the FGD.
Assessment of the relationship between migration frequency and zcNDVI indicates that under severe drought conditions about 4 times as many pastoralist households moved than expected.
Similarly, a cumulative percentage of migrations of 18%, 28%, and 25% are registered at zcNDVI of below -1, which is still higher than the expected 16%. We here infer that almost twice as many households moved under severe and medium drought conditions, as would have been expected following a normal distribution.
We also see larger than expected (50%) migration frequencies for the cumulative percentage of migrations at a zcNDVI value of below zero (0), i.e. 58%, 50% and 52% for GIMMS(1981-2015), GIMMS(2001-2015) and eMODIS(2001-2016), respectively. We, therefore, infer that current-season drought relates to urban migration of pastoralist communities under all, mild, medium and severe drought conditions in Marsabit County.
We also tested if migration has a differential link with drought conditions based on gender. By comparing cumulative migration frequency against the zcNDVI thresholds (GIMMS dataset 1981-2015), separated for male- and female-headed households, we observe that under medium drought conditions more men-headed households move than expected; under severe drought conditions three times as expected. This suggests that male-headed households respond to mild, medium and severe drought conditions, much in contrast to women-headed households. For the latter, we did not find an indication of drought-related migration.
We assessed migration by age groups to the drought indices. Furthermore, data analysis indicates that compared to the normal distribution more migrations were registered for young people under severe and medium drought conditions and about twice as many under severe drought conditions. Comparing young and old, about twice as many young people and their households moved under severe, medium, and mild drought conditions (zcNDVI under 0) and about half more under severe and medium drought conditions. A similar but slightly less pronounced migration pattern is seen for the age group 31-50 years old, whereas the elderly migrates much less.
Finally, we measured reported quality of life after migration: While the urban migrants are more involved in informal employment, prone to unemployment and high dependence on relief aid, they seem to be generally more pleased with their current quality of life. Most pastoralists drop-outs (96%) were more comfortable in the urban centres than in their previous rural locations, and not willing to return to their rural areas, while 3% were willing to move back.
The relationship between drought and rural to urban migration is multi-faceted and not yet well-understood. To contribute to the growing body of knowledge on this complex topic, we found that temporal peaks of rural to urban migration coincided with low values of the satellite-based forage scarcity indices. Our results, therefore, show that drought contributes to rural to urban migration decisions: 91% of migrants list drought-related reasons as the main trigger for migration into Marsabit Town, an overwhelming majority. However, about 6% of respondents in our study named poverty as a cause of migration. This validates prior studies showing that cities also radiate economic, infrastructural and social push factors (Piguet 2011). Our discussions with key informants also revealed poverty.
In addition, conflict and insecurity are determinants and push factors of migration. In this research, 3% of pastoralists relocated because of ethnic clashes particularly resource conflict, cattle rustling, and livestock raiding as noted by Jónsson (2010).
Under severe drought conditions, more pastoralist households move than under normal conditions. Male-headed households strongly respond to drought. However, migrations of female-headed households did not show a relation to drought in our sample. Age is also an important factor in environmental migration. Young people and their households moved under (severe or medium) drought conditions as compared to the elderly.
When moving to town, livelihood and income profiles of pastoralist drop-outs gradually shifted from natural-resource-based livelihoods to more sedentary, income-generating activities—often casual labour. Poverty was in some cases the direct consequence—and, hence, can be both, a cause and a result of migrations. Nonetheless, most respondents are satisfied with their new urban location and would not want to move back to their rural origins, although a majority has an interest to practice pastoralism while keeping their urban basis. For the urban planning discourse, these results highlight the need to plan for migration as a major factor of urbanization in the destination areas (towns) within dryland areas. Planning for pastoral community resilience at source (rural) areas (to stand pressures of drought, potentially diversify livelihoods and reduce conflict; increase infrastructure services) is also paramount.
This study offers a methodological advantage to previous studies by using satellite-derived drought indices in combination with a community-based survey. Through this approach, the study was able to blend methods that explain migration patterns and decisions of pastoralists based on the seasonal performance of vegetation growth. It provides a quantitative way of comparing migration counts to satellite-derived drought indices. Our study thereby weighed the contribution of environmental factors (drought) in migration dynamics.
The authors would like to thank Sixtus Odumbe (SNV-Turkana), Ochieng Mc’Okeyo (Highlands Surveyors) and Joseph Yaw Frimpong.
(1) For a more detailed elaboration of the methodology, please see: Ogara, D. (2018) Assessing the fluxes and impacts of drought-induced migration of pastoralist ommunities into urban areas: A case of Marsabit Town, Northern Kenya. Enschede: University of Twente. Online: http://purl.utwente.nl/essays/83772.
(2) These interviews were held with a representative of the National Drought Management Authority, the Director of Forest Services, a Physical County Planner, and two representatives of NGO’s that conduct drought-affiliated activities.
(3) From a z-scored value per IBLI unit, per season, per year, for both datasets of GIMMS (1981-2015) and eMODIS (2001-2016), we computed a single averaged zcNDVI value for all the 14 IBLI units per season within a given year, to get an estimated zcNDVI value for the whole of Marsabit County. This yielded separate values for the GIMMS and MODIS datasets.
(4)The Shapiro Wilk test results for the different data sets were: GIMMS 1981-2015 (w=0.98177; p-value=0.4213); MODIS 2001-2016 (w=0.95876; p-value=0.254); GIMMS 2000-2015 (w=0.9597; p-value=0.2865). For each data set, the test indicated no significant departure from normality.
- Adow, Mohamed. 2008. “Pastoralists in Kenya.” Forced Migration Review. http://www.fmreview.org/sites/fmr/files/FMRdownloads/en/climatechange/adow.pdf.
- Aklilu, Y, and M Wekesa. 2002. “Drought, Livestock and Livelihoods: Lessons from the 1999–2001 Emergency Response in the Pastoral Sector in Kenya.” HPN Network Paper, no. December 2002: 40.
- Alston, Margaret. 2015. Women and Climate Change in Bangladesh. Routledge. https://doi.org/10.4324/9781315774589.
- Barrett, Christopher B., Paswel Phiri Marenya, John Mcpeak, Bart Minten, Festus Murithi, Willis Oluoch-Kosura, Frank Place, Jean Claude Randrianarisoa, Jhon Rasambainarivo, and Justine Wangila. 2006. “Welfare Dynamics in Rural Kenya and Madagascar.” Journal of Development Studies 42 (2): 248–77. https://doi.org/10.1080/00220380500405394.
- Barton, David, John Morton, and Cary Hendy. 2001. “Policy Series15: Drought Contingency Planning for Pastoral Livelihoods.” Natural Resource Institue. University of Greenwich.
- Bruijn, M de, and H van Dijk. 2003. “Changing Population Mobility in West Africa: Fulbe Pastoralist in Central and South Mali.” African Affairs 102 (407): 285–307. http://afraf.oxfordjournals.org/content/102/407/285.abstract.
- Bruno Soares, Marta, Alexandre S. Gagnon, and Ruth M. Doherty. 2012. “Conceptual Elements of Climate Change Vulnerability Assessments: A Review.” International Journal of Climate Change Strategies and Management 4 (1): 6–35. https://doi.org/10.1108/17568691211200191.
- Buono, Nicoletta, Sara Di Lello, Margherita Gomarasca, Cornelia Heine, Stefano Mason, Michele Nori, Rita Saavedra, and Koen Van Troos. 2016. “Pastoralism, the Backbone of the World’s Drylands.” Vétérinaires Sans Frontières International. http://vsf-international.org/wp-content/uploads/2016/09/REPORT-pastoralism-2017-pag1-1401.pdf.
- Cervigni, Raffaello, and Michael L. Morris. 2016a. Confronting Drought in Africa’s Drylands : Opportunities for Enhancing Resilience. The World Bank Group. https://doi.org/10.1596/978-1-4648-0817-3.
- ———. 2016b. “Confronting Drought in Africa’s Drylands : Opportunities for Enhancing Resilience,” 257.
- Dai, Aiguo. 2011. “Drought under Global Warming: A Review.” Wiley Interdisciplinary Reviews: Climate Change 2 (1): 45–65. https://doi.org/10.1002/wcc.81.
- ———. 2013. “Increasing Drought under Global Warming in Observations and Models.” Nature Climate Change 3 (1): 52–58. https://doi.org/10.1038/nclimate1633.
- Epule, Terence Epule, James D Ford, Shuaib Lwasa, and Laurent Lepage. 2017. “Climate Change Adaptation in the Sahel.” Environmental Science and Policy 75 (March): 121–37. https://doi.org/10.1016/j.envsci.2017.05.018.
- Foresight. 2011. “Migration and Global Environmental Change: Future Challenges and Opportunities.” Government Office for Science - Forsight, 234. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/287717/11-1116-migration-and-global-environmental-change.pdf.
- Fratkin, Elliot. 2001. “East African Pastoralism in Transition: Maasai, Boran, and Rendille Cases.” African Studies Review 44 (3): 1. https://doi.org/10.2307/525591.
- Fratkin, Elliot, Eric Abella Roth, and Martha A. Nathan. 2004. “Pastoral Sedentarization and Its Effects on Children’s Diet, Health, and Growth among Rendille of Northern Kenya.” Human Ecology 32 (5): 531–59. https://doi.org/10.1007/s10745-004-6096-8.
- Gaillard, J.C. 2010. “Vulnerability, Capacity and Resilience: Perspectives for Climate and Development Policy.” Journal of International Development 22 (2): 218–32. https://doi.org/10.1002/jid.1675.
- Gemenne, François. 2011. “Why the Numbers Don’t Add up: A Review of Estimates and Predictions of People Displaced by Environmental Changes.” Global Environmental Change 21 (SUPPL. 1): 1–9. https://doi.org/10.1016/j.gloenvcha.2011.09.005.
- Gemenne, François, Jon Barnett, W. Neil Adger, and Geoffrey D. Dabelko. 2014. “Climate and Security: Evidence, Emerging Risks, and a New Agenda.” Climatic Change 123 (1): 1–9. https://doi.org/10.1007/s10584-014-1074-7.
- Henry, Sabine, Bruno Schoumaker, and Cris Beauchemin. 2004. “The Impact of Rainfall on the First Out-Migration: A Multi-Level Event-History Analysis in Burkina Faso.” Population and Environment 25 (5): 423–60. https://doi.org/10.1023/B:POEN.0000036928.17696.e8.
- Huho, Julius M., and Edward M. Mugalavai. 2010. “The Effects of Droughts on Food Security in Kenya.” The International Journal of Climate Change: Impacts and Responses 2 (2): 61–72. https://doi.org/10.18848/1835-7156/CGP/v02i02/37312.
- Hunter, Lm, and E David. 2009. “Climate Change and Migration: Considering Gender Dimensions.” IBS Population Program, no. November. http://www.colorado.edu/ibs/pubs/pop/pop2009-0013.pdf.
- IPCC. 2012. “Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation. A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change [Field, C.B., V. Barros, T.F. Stocker, D. Qin, D.J. Dokken, K.L. Ebi, M.D.” Cambridge, UK, and New York, NY, USA.
- KIRA. 2014. “Marsabit Secondary Data Review” 1 (April): 1–7.
- Kubik, Zaneta. 2018. “Weather Shocks , Migration and Food Security : Evidence from Sub-Saharan Africa.”
- Linke, Andrew, Frank DW Witmer, John O’Loughlin, J. Terrence McCabe, and Jaroslav Tir. 2018. “Relocating in Response to Drought: Human Mobility and Conflict in Contemporary Kenya.” Environmental Research Letters, August. https://doi.org/10.1088/1748-9326/aad8cc.
- Magal, Raphael P, and Boniface N Wambua. 2017. “Risk Assessment of the Coping and Adaptation Mechanisms for Pastoralists to Climate Change and Variability : A Case Study of Kongelai Ward , West Pokot County , Kenya.” Asian Journal of Agriculture and Food Sciences 05 (02): 62–73.
- Masih, I., S. Maskey, F. E.F. Mussá, and P. Trambauer. 2014. “A Review of Droughts on the African Continent: A Geospatial and Long-Term Perspective.” Hydrology and Earth System Sciences 18 (9): 3635–49. https://doi.org/10.5194/hess-18-3635-2014.
- McCall, Mike. 2006. “Precision for Whom? Mapping Ambiguity and Certainty in (Participatory) GIS.” In Participatory Learning and Action, 54:114–119.
- Middleton, N J, and T Sternberg. 2013. “Climate Hazards in Drylands: A Review.” https://doi.org/10.1016/j.earscirev.2013.07.008.
- Mogotsi, K, M M Nyangito, and D M Nyariki. 2011. “The Perfect Drought? Constraints Limiting Kalahari Agro-Pastoral Communities from Coping and Adapting.” African Journal of Environmental Science and Technology 5 (March): 168–77. https://doi.org/10.5897/AJEST10.216.
- Morrow, Betty Hearn. 2008. “Community Resilience: A Social Justice Perspective,” 31.
- Nori, Silvia, and Michele Gemini. 2011. “The Common Agricultural Policy Vis-a-Vis European Pastoralists: Principles and Practices.” Pastoralism: Research, Policy and Practice 1 (1): 27. https://doi.org/10.1186/2041-7136-1-27.
- Perch-Nielsen, Sabine L., Michèle B. Bättig, and Dieter Imboden. 2008. “Exploring the Link between Climate Change and Migration.” Climatic Change 91 (3–4): 375–93. https://doi.org/10.1007/s10584-008-9416-y.
- Piguet, Etienne. 2011. “The Migration/Climate Change Nexus: An Assessment.” Rethinking Migration: Climate, Resource Conflicts and Migration in Europe, 1–27. www.network-migration.org and www.geographie.uni-bremen.de.
- Rain, David, Ryan Engstrom, Christianna Ludlow, and Sarah Antos. 2011. “Accra Ghana: A City Vulnerable to Flooding and Drought-Induced Migration.” Global Report on Human Settlements 2011, no. May 2009: 21.
- Raleigh, Clionadh, and Dominic Kniveton. 2012. “Come Rain or Shine: An Analysis of Conflict and Climate Variability in East Africa.” Journal of Peace Research 49 (1): 51–64. https://doi.org/10.1177/0022343311427754.
- Reckien, Diana, Felix Creutzig, Blanca Fernandez, Shuaib Lwasa, Marcela Tovar-Restrepo, Darryn Mcevoy, and David Satterthwaite. 2017. “Climate Change, Equity and the Sustainable Development Goals: An Urban Perspective.” Environment and Urbanization 29 (1): 159–82. https://doi.org/10.1177/0956247816677778.
- Republic of Kenya, County Government of Marsabit, UNDP. 2017. Revised First County Integrated Development Plan (2013-2017).
- Rigaud, Kanta Kumari, Alex De Sherbinin, Bryan Jones, Jonas Bergmann, Viviane Clement, Kayly Ober, Jacob Schewe, et al. 2018. “Groundswell - Preparing for Internal Climate Migration.” Washington, DC: The World Bank, 256. https://openknowledge.worldbank.org/bitstream/handle/10986/29461/WBG_ClimateChange_Final.pdf.
- Rojas, O., A. Vrieling, and F. Rembold. 2011. “Assessing Drought Probability for Agricultural Areas in Africa with Coarse Resolution Remote Sensing Imagery.” Remote Sensing of Environment 115 (2): 343–52. https://doi.org/10.1016/j.rse.2010.09.006.
- Spinoni, Jonathan, Gustavo Naumann, Hugo Carrao, Paulo Barbosa, and Jürgen Vogt. 2014. “World Drought Frequency, Duration, and Severity for 1951-2010.” International Journal of Climatology 34 (8): 2792–2804. https://doi.org/10.1002/joc.3875.
- Stidsen, Sille. 2006. The Indigenous World; 2006. Copenhagen: IWGIA. https://books.google.nl/books?redir_esc=y&output=html&id=SdbwzF4k-msC&dq=%22Turbi%22+%22massacre%22&focus=searchwithinvolume&q=%22Turbi%22+%22massacre%22.
- UNDP. 2009. Resource Guide on Gender and Climate Change. United Nations Development Programme.
- Vrieling, Anton, Michele Meroni, Andrew G. Mude, Sommarat Chantarat, Caroline C. Ummenhofer, and Kees C.A.J.M. de Bie. 2016. “Early Assessment of Seasonal Forage Availability for Mitigating the Impact of Drought on East African Pastoralists.” Remote Sensing of Environment 174: 44–55. https://doi.org/10.1016/j.rse.2015.12.003.
- Wang, Zhaoli, Ruida Zhong, Chengguang Lai, Zhaoyang Zeng, Yanqing Lian, and Xiaoyan Bai. 2018. “Climate Change Enhances the Severity and Variability of Drought in the Pearl River Basin in South China in the 21st Century.” Agricultural and Forest Meteorology 249 (September 2017): 149–62. https://doi.org/10.1016/j.agrformet.2017.12.077.
- Webber, Michael, and Jon Barnett. 2010. “Accommodating Migration To Promote Adaptation To Climate Change,” no. April. https://doi.org/10.1596/1813-9450-5270.
Table and figures legend
- Figure 1: Location of Marasabit County in Kenya, and Marsabit Town within Marsabit County; google earth image of Marsabit Town. Source: Author.
- Figure 2: Marsabit Index-Based Livestock Insurance (IBLI) Spatial Units based on aggregated sub locations.
- Figure 3:Number of documented migrations aggregated per bimodal seasonality (1980-2017). Key: Light green shows the number of migrations in the Long rain season (April-October); dark blue shows the migrations in the Short rain season (November-March). Source: Author.
- Figure 4: Cumulated migrations plotted against the zcNDVI per season. zcNDVI values of 0 indicate normal/average conditions. zcNDVI values below -1 indicate dry conditions; the lower the value the poorer the performance of the vegetation in that season.
- Figure 5: Gender of Household Heads plotted against average of current season of zcNDVI. Dataset used: GIMMS 1981-2015. Source: Author.
- Figure 6: Cumulated migrations plotted against the zcNDVI per season. zcNDVI values of 0 indicate normal/average conditions. zcNDVI values below -1 indicate dry conditions; the lower the value the poorer the performance of the vegetation in that season (i.e. drier conditions). Dataset used: GIMMS 1981-2015, GIMMS 2001-2015 and eMODIS 2001-2016. Source: Author.
- Table 1: A comparison of pastoralist drop-outs perceived quality of life before and after migration.