1. Agropastoral livelihood system typology for coping with socio-ecological diversity: a demonstrative case in Karauzyak, Karakalpakstan, Uzbekistan Davran Niyazmetov, Akmal Akramkhanov, Timur Ibragimov, Quang Bao Le 4-6 April, 2016 Almaty, Kazakhstan

2. Photos by: Ms. Sanobar Khudaybergenova 2 Republic of Karakalpakstan is located in Northwest of Uzbekistan Embraces the vast dry lands (harsh environmental conditions) One of the regions with low income in Uzbekistan Introduction

3. Why livelihood systems? Irrelevance: “Uniform blanket” of rural development strategy vs. diversity of livelihood context that shape livelihood outcomes Need: To identify plausible livelihood typology as a basis for better targeting and out-scaling in research and development Gaps: Lack of published research on this in the study site Introduction 3

4. 1.Identify main factors discriminating agricultural livelihood types at village level; 2.Identify and characterize agricultural livelihood types in the study sites. Objectives 4

5. Two villages in Karauzyak district have been selected for the survey and analysis: Karabuga - located to the South from the district center and having more favourable conditions Algabas - located to the North from the district center and having harsh conditions Study site 5

6. Description of study site ##Village Citizen CouncilsHouseholdsPopulation Including MenWoman 1Karakol7125,2152,6152,600 2Berdakh8825,4952,7492,746 3Algabas6755,2082,6382,570 4Koybak2281,446725721 5Madaniyat8965,6402,8302,810 6Karauzyak7105,0582,5322,526 7Esimozek3702,4211,2151,206 8Karabuga7094,9202,4702,450 Out of total 1,384 households in selected two villages, 100 (7%) households were surveyed 6

7. Methods and materials The Sustainable Livelihood Framework (SLF) serves to view a household-farm as whole by taking into account all of its natural, physical, financial, human, and social assets. 7 Source: slide from IFAD SL Workshop presentation

8. Livelihood assets: Human capital (demography, education, and profession of household members); Natural capital (i.e., land, both own and leased); Physical capital (agricultural equipment, transportation assets, etc.); Financial capital (income from all sources, savings, livestock); Social capital (social status, networking, public activity). 8 Methods and materials

9. Identification of household-farm types Combined multivariate analysis and expert knowledge The multivariate analysis consisted of two steps (PCA and K-means cluster analysis) ANOVA - to characterize identified agricultural livelihood types 1, 2,..., k clusters Main components K-means cluster analysis 100 households Principal component analysis of the dataset collected using Sustainable Livelihood Framework Karabuga and Algabas villages Random sampling 9

10. Results – Site characterization Farming system characteristics Household size (members)5.5 Female Head (%)13 Illiteracy (%)1 Bachelor degree (%): female30 Bachelor degree (%): male32 Network membership (%)10 Holding (ha/person)0.07 Livestock (unit per person)0.64 Food availability concern (%)67 10

11. Results – PCA identified key variable discriminating household livelihoods 11 Principal components Livelihood asset Variable 1-Land PC (19%) 2-Mark. PC (14%) 3-Inc. PC (12%) 4-Age. Exp. PC (10%) Human H HEADAGE 0.003-0.1010.0720.821 H HEADEXP -0.11-0.109-0.0550.802 H SIZE -0.08-0.026-0.0410.106 H LABOUR 0.0130.007-0.1210.166 H DEPEND 0.087-0.202-0.151-0.718 PhysicalH DFMARKET -0.0090.9820.054-0.001 Natural H HOLDINGS 0.9690.0570.0130.01 H HOLDINGCP 0.9630.0740.013-0.025 H CULTLANDCP 0.9630.0740.013-0.025 Financial H SHRFINC -0.046-0.15-0.911-0.091 H SHREMITINC 0.0520.0340.8580.05 H SHNOFFINC -0.0230.157-0.1520.045 SocialH SATCSDM 0.0710.0090.055-0.01

12. Results: Agricultural Livelihood types in study sites Livelihood type I: Educated, land-poor, livestock- and poultry-rich, off-farm- income-oriented farms Represents 10% of the study sample; Off-farm activities for income generation (86.9%), farm income - 6.4%; Land holding – 0.01 ha per person; Networking - satisfied with community activities, and there is one community leader (11% of households). 78% of households in this type have a reliable informal source of borrowing within a community. 12

13. Livelihood type II: Farm-income-dependent, less educated, land-poor, poultry-turned farms Represents 16% of the study sample; Land holding – 0.03 ha per person: highest average yield of vegetables (68% of land use) among all types, which is 6.7 ton per ha; Dependence upon farm activities – 54% of total income and remittances 16% as income source; Households have more poultry (0.94) than livestock (0.54); Less socially active than other ones. Results: Agricultural Livelihood types in study sites 13

14. Livelihood type III: Land-rich, poultry-turned, off-farm-income-dependent farms Represents the majority of the study sample – 74% Highest availability of the land and diversification of land use; Land holding – 0.09 ha per person: this type uses less than half of its land for cultivation of vegetables (47.9%); This type of livelihood uses land much more efficiently than type I: the average annual yield of vegetables is 5.1 ton per ha (vs. 1.1 ton/ha of type 1); Social activity: 14% of it participates in public organization, with 3% have community leaders. Results: Agricultural Livelihood types in study sites 14

15. Conclusion Both surveyed villages apply certain livelihood strategies prominent in rural areas, including: 1. subsistence agriculture; 2. seasonal labor migration; 3. official jobs at state-funded or budget organizations: 4. some entrepreneurial (non-agricultural) activities. Majority of the surveyed households cultivate food crops (vegetables, beans, fruits, etc.) for own consumption; cultivate fodder crops to feed their livestock, and limited amount of fruits and vegetables for sale. The cluster analysis resulted in three agricultural livelihood types for household-farms in the study site. 15

16. Opportunities for growth in household-farms appear to be limited by very small farm sizes. Marketed surpluses by household are small and, cash earning are limited. Addressing non-agricultural income sources is essential in rural growth. Conclusion 16

17. 1.Assess livelihood type-specific performances, behavior and drivers 2.Develop farming system models capturing the livelihood typology 3.Use above results to recommend policy and community management strategies target better specific constraints, potentials and needs, as well as outscaling of farm-level findings 17 Next steps

18. Thank you very much for your kind attention!!! timur.nazarbaevich@gmail.com t.ibragimov@cigar.org Current research has been initiated within the framework of “Integrated Systems Analysis and Modelling in Aral Sea Region” activity in Uzbekistan, by ICARDA, DS-CRP