Barriers to and determinants of the choice of crop management strategies to combat climate change in Dejen District, Nile Basin of Ethiopia

Barriers to and determinants of the choice of crop management strategies to combat climate change... Background: Climate change without adaptation is projected to impact strongly the livelihoods of the rural com- munities. Adaptation to climate change is crucial for least developed country like Ethiopia due to high population and dependency on agriculture. Hence, this study was initiated to examine the barriers to and determinants of the choice of crop management strategies to combat climate change. The Intergovernmental Panel on Climate Change concepts of climate change response provided the framework. Stratified and snowball sampling techniques were employed to select a sample of 398 households. The household survey was employed to collect data on current adaptation strategies. Logistic regression was used to analyse the determinants of the choice of adaptation strategies. Logistic regression analyses were carried out at p ≤ 0.05. Results: Small farmland size, agro-ecology, farmland location, financial constraints, and lack of skills were the major barriers to adoption of crop management strategies. Age, farming experience, income, family size, government experts’ extension services, agro-ecology setting, and crop failure history of households significantly affect the choice of most of the crop management strategies. Conclusions: Socio-economic and institutional factors determined rural communities’ ability and willingness to choose effective adaptation strategies. Policy priority should be given based on agro-ecology and households demand of policy intervention such as providing extension services and subsidizing the least adopted strategies due to financial constraints. Keywords: Adaptation, Climate change impact, Crop management practices, Nile Basin of Ethiopia Background average of 14 °C. It was 0.08 °C above the average anom- The warming trends observed over the past few decades aly of 0.50 °C for the past 10 years (2005–2014) [1]. continued in 2014. World Meteorological Organization According to Food and Agricultural Organization [2], (WMO) has ranked as nominally the warmest year since due to climate change and variability almost one billion modern instrumental measurements began in the mid- people experienced hunger in 2010 globally. This implies 1800s [1]. The global average near-surface temperature the most marginalized people cannot access enough of for 2014 was comparable to the warmest years in the the primary macronutrients. Perhaps, other billions are 165-year instrumental record. In 2014, the global aver- thought to suffer from hidden hunger, in which essential age temperature was 0.57 ± 0.09 °C above the 1961–1990 micronutrients are missing from their diet, with conse- quent risks of physical and mental impairment [3]. The majority (85%) of the Ethiopian population is dependent *Correspondence: Zerihun.yohannes19@gmail.com on agriculture. As a result, agriculture will continue to be Institute of Disaster Risk Management and Food Security Studies, Bahir Dar University, BahirDar, Ethiopia Full list of author information is available at the end of the article © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/ publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Amare et al. Agric & Food Secur (2018) 7:37 Page 2 of 11 the most important sector in its need to adapt to climate of 1071 and 3000  m above sea level (m.a.s.l). The aver - change. age temperature and total annual rainfall of the district There are many reasons and convincing arguments for range between 20 and 24  °C and 800 and 1200  mm, a more comprehensive consideration of adaptation as respectively. Dejen District is located in the valley of the a response measure to climate change. Firstly, given the Blue Nile which is highly undulated topography and fre- amount of past greenhouse gas emissions and the inertia quent susceptibility to climate-related problems such as of the climate system, we are already bound to some level erratic rainfall, crop pests, livestock diseases, and malaria of climate change, which can no longer be prevented outbreaks. In the past 8  years (2009–2016) due to cli- even by the most ambitious emission reductions [4]. Sec- mate change impacts, the district loses 50,555 quintals of ond, the effect of emission reductions takes several dec - crops, which has the potential to feed 27,701 individuals. ades to fully manifest, whereas most adaptation measures The study district is categorized into three agro-ecolog - have more immediate and sustainable benefits [5]. ical zones, 41% Dega (highland), 31% Woinadega (mid- Third, adaptations can be effectively implemented on land), and 28% Kolla (lowland) [11, 12]. a local or regional scale such that its efficiency is less dependent on the actions of others, whereas mitiga- Research design and sampling procedure tion of climate change requires international coopera- The study employed cross-sectional research design with tion. Fourth, most adaptations to climate change also both quantitative and qualitative research methods. This reduce the risks associated with current climate variabil- study used a multi-stage sampling technique to select the ity, which is a significant hazard in many world regions. agro-ecology, Kebeles (the lower administrative unit next According to Gbetibouo [6], there are two adaptation to district), and households. At the first stage, Dejen Dis - assessment approaches, namely top-down and bottom- trict of the Nile Basin was selected purposely due to its up assessment approaches. The top-down approach highly undulated topography and frequent susceptibil- starts with climate change scenarios, and estimates ity to extreme events and representativeness of the three impact through scenario analysis, based on which pos- agro-ecological zones such as highland, midland, and sible adaptation practices are identified. Most of the lowland. In the second stage, six Kebeles (two from each top-down adaptations represent possible or potential agro-ecological setting) were selected purposely based on measures, rather than those that have been used [6]. the above-listed district selection criteria. Most studies, e.g. [7–9], carried out in Ethiopia and Climate change affects the rural communities differ - Africa using top-down approach predicted the impact ently in different agro-ecological zones. As a result, com - of climate change on the agricultural sector with adverse munities’ knowledge and skill to adapt to the climate effects on crop yields. The bottom-up approach takes a change impacts varies from different agro-ecological set - vulnerability perspective where adaptation strategies are tings. In the third stage, stratified sampling was employed considered more as a process involving the socio-eco- to select households. Under the stratified sampling, the nomic, and policy environments, and elements of deci- population was divided into male- and female-headed sion-making [6]. In line with this notion, Schröter et  al. households, and then the sample was selected from each [10] argue that in choosing adaptation options to climate male- and female-headed household to constitute a rep- change and developing policies to implement these pos- resentative sample. sibilities the affected community should actively par - The sample size was determined proportionately ticipate. This study adopts the bottom-up approach that (Table 1). In Ethiopia context, female-headed households seeks to identify actual adaptation strategies at the local are those who do not have husband due to either being level and the factors that determine the choice of crop divorced, widowed, or separated. In Ethiopia, in some of management strategies in Dejen District, Nile Basin of the rural communities, disclosing of the marital status of Ethiopia. older females is culturally not feeling comfort them. Thus, to get female-headed households, snowball sampling was Methods employed. Based on Yamane [13] at the 95% confidence The study area interval and 5%, level of precision, 398 households were The study district is located in west-central Ethiopia selected at the six Kebeles of the district (Table 1). (Fig. 1) at a road distance of 335 km south of the regional state capital, Bahir Dar City, and 230  km northwest of Data sources and data collection methods the capital city of Ethiopia, Addis Ababa, in the Amhara The primary and secondary data sources were quantita - Regional State at the edge of the canyon of the Blue Nile. tive and qualitative in nature. The study used two main The district lies between longitude 38°6′E and 38°10′E, data sources to analyse the barriers and determinants of and between latitude 10°7′N 10°11′N, with an elevation the adoption of crop management strategies to climate Amare et al. Agric & Food Secur (2018) 7:37 Page 3 of 11 Fig. 1 Study area change impacts. The main data source was the socio-eco - language for data processing and analysis. The house - nomic data collected through household survey. hold survey was initially pretested to check the appro- priateness and validity of the questions from the three agro-ecological zones. Pretested Kebeles and partici- Household survey pant households were not involved in the actual survey. The household survey was used to collect quantitative After pretesting, ambiguous words were rephrased data on households’ current adaptation strategies, the and inappropriate questions were modified. The expe - barriers to adapt, and factors that affect rural com - rience of the author in the study area played a para- munities’ choice of crop management strategies. The mount role in choosing the data collectors who have survey questions were prepared in English language. been working for many years in the rural community in It was translated into local language (Amharic) during the field of agriculture, environment, and land admin - data collection and then encoded into SPSS in English istration. The data were collected by the trained data Amare et al. Agric & Food Secur (2018) 7:37 Page 4 of 11 Table 1 Distribution of  sample size based on  the  total logistic regression was inappropriate. To fix such prob - study population. (Source: computed based on [14]) lems, the possible remedy suggested by Bryan et  al. [15] was to combine similar measures into single categories. Sample Number % of study Share of Kebeles Kebeles of household heads population from 398 HHs However, such grouping into self-defined categories may lead to miss interpretation [15, 16]. Besides, the set of Male Female Total explanatory variables affecting the households’ decision Koncher (high- 995 225 1220 16.5 66 was also expected to be different for different adaptation land) strategies. Therefore, this study employed binary logis - Yetnora (high- 1388 767 2155 29 115 tic regression technique to analyse the determinants of land) the choice of crop management strategies. Analysis of Zemeten (mid- 652 280 932 12.6 50 the logistic regression of binary response model can be land) defined as: Borebor (mid- 644 281 925 12.5 50 land) β0+β1(x1)+β2(x2)+β3(x3)+βn(xn) Kurar (lowland) 726 363 1089 14.7 59 y = e , −[β0+β1(x1)+β2(x2)+βn(xn) Gelgele (low- 732 345 1077 14.6 58 (1) land) 5137 2261 7398 100 398 where y is response variable; y = 1, outcome is present HHs Households (adopt); y = 0, outcome is not present (not adopt); β is constant. β + β + β + ··· β are regression coefficients that 1 2 3 n collectors under close supervision of the author in the explain the change in the log odds for each unit change period March to October 2016. in x. x + x + ··· x = set of predictor variables included 1 2 n in the model (Table  2). e represents the change in the Methods of data analysis odds of the outcome (multiplicatively) by increasing x by The descriptive statistics such as percentage and fre - 1 unit. The current crop management adaptation strate - quencies were used to summarize and categorize the gies in the study area, such as using crop diversification, information gathered from households. The binary logis - improved seeds, changing planting date, and replant- tic regression was used to analyse determinants of the ing failed crops, were used as dependent variables. The choice of crop management strategies. choices of these adaptation strategies were determined by Before the data collection, a multinomial logistic socio-economic and demographic characteristics, insti- (MNL) model was proposed based on aspects of the tutional factors, and agro-ecological settings (Table  2). literature. However, in the course of this study, the sur- Based on Agresti [17], logistic regression method can be veyed households chose more than one adaptation strate- used when the dependent variable Y is dichotomous. Y gies simultaneously. As a result, the use of multinomial Table 2 Description of  explanatory variables that  affect households’ choice of  crop management strategies. Source: Author (2016) Explanatory variables Description Sex Dummy, 1 = Male, 0 = Female Age Discrete (years), 18–35, 36–55, > 55 Education Discrete (years), cannot read and write, primary school and above) Farming experience Continuous (years), 10, 10–20, 21–30, > 30 Income Continuous (ETB), < 10,000, 10,001–30,000, 30,001–50,000, > 50,000 Family size Discrete (number), < 4, and > 4 Access to weather information Dichotomous, 1 = Yes, 0 = No Access to farmer to farmer extension services Dichotomous, 1 = Yes, 0 = No Access to government experts’ extension services Dichotomous, 1 = Yes, 0 = No Agro-ecology settings Dummy, 1 = highland; 2 = midland; 3 = lowland Farmland size Continuous (hectare), < 1.2, and > 1.2 Crop failure history of households Dichotomous, 1 = Yes, 0 = No Taken as reference (base for analysis), ETB is Ethiopian currency ($1 = 22.3ETB) [18] Amare et al. Agric & Food Secur (2018) 7:37 Page 5 of 11 is coded as 1 when the outcome is present and coded 0 independent variable is lower than the reference/base when the outcome is not present. variable (Table 2). Multicollinearity was assessed by examining toler- Test of goodness of fit and multicollinearity ance and variance inflation factors (VIF). The variance The fitness of the logistic regression model to the data inflation factor (VIF) quantifies how much the variance was measured by using the SPSS classification table is inflated. The tolerance is the percentage of the vari - (crosstabs) and the Hosmer–Lemeshow test. Besides, ance in a given predictor that cannot be explained by collinearity among predictor variables was checked using the other predictors. When VIF > 5, X (the explanatory multicollinearity statistics. The Hosmer–Lemeshow test variable) is highly correlated with the other explanatory used 95% confidence interval (CI) and asymptotically fol - variables [20, 21]. Mathematically, variance inflation lows a X distribution. factor (VIF) can be expressed by: 1 2 Empty cells or small frequency was checked by doing VIF = , where R j is the coefficient of determina - 1−R j crosstabs between categorical predictor variables and the outcome variables. When the cell has very few cases, the tion of a linear regression model that uses X as the model becomes unstable. Based on Kothari [19] and SPSS response variable and all other X variables as the [20], the Hosmer–Lemeshow statistics indicate a poor fit explanatory variables. Tolerance is the reciprocal of if the significance value (p) is less than 0.05 (Table 3). VIF (i.e. tolerance = ). A tolerance of less than 0.20 VIF Techniques to remedy poor fit of the model were by: and/or a variance inflation factor (VIF) greater than 5 (1) using re-categorized (for example educational levels and above indicates a multicollinearity problem. How- were categorized as cannot read and write and primary ever, in this study, the results of the variance inflation school level and above), (2) dropping the least theoreti- factor indicated that there was no multicollinearity cally important explanatory variables that contribute to problem. the model a poor fit to the data. For example, explanatory variables such as family size and crop failure history of households were excluded from entering and competing Results in the model in the improved seed and crop diversifica - Barriers to crop management strategies tion adaptation options, respectively. Implementation of crop management strategies used by In the logistic regression model, the Exp (B) is the the rural communities varied among households. The “Odds ratio” which explains the effect of the independ - study identified a number of constraints faced by the ent variable (X ) on the dependent variable. The beta households to adopt crop management strategies to com- coefficient (β ) is the estimated logit coefficient which bat climate change impact. is the rate of change in the Y (the dependent variables) as X (independent variable) changes. When the beta coefficient (β ) is negative, it shows that the dependent Crop diversification and independent variables have an inverse relationship, The major constraints identified by respondents to not and when it has a positive coefficient, there is a positive to adopt crop diversification were: small land size (57%) relationship. Odds ratio = 1 indicates the same proba- followed by soil fertility decrement (25.3%), shortage bility of an event occurring between the two situations. of money to buy some expensive crop varieties (1.3%), Odds ratio > 1 probability of an event occurring with shortage of labour (2.5%) to implement some labour- a unit increase in the independent variable is higher intensive farming practices, and they prefer the easiest than the reference/base variable. Odds ratio < 1 prob- crop type, lack of skill (3.8%) to sow different crops and ability of an event occurring with a unit increase in the stick to one types of crop, topography of farmland loca- tion (2.5%) which permits only some crop types to grow, and the remaining (2.5%) do not/have small land size. Table 3 Hosmer and  Lemeshow goodness-of-fit test The majority (58%) of rural communities have less than results of  logistic regression model. (Source: Computed 1.2 hectares. As a result, small land size/no land at all based on household survey data, March–October (2016)) consequences is barrier to the choice of crop manage- Dependent variables Chi-square df P value (> 0.05) ment strategies. On the other hand, rural communities Crop diversification 7.763 8 0.457 who are located in the lowlands of the undulated topog- Improved seeds 10.78 8 0.214 raphy encountered farm soil fertility decrement. Due to Changing planting date 9.239 8 0.323 this reason, farmers were obliged to leave their farmland (fallowing) for some periods instead of diversifying dif- Replanting failed crops 2.416 8 0.966 ferent crops as an adaptation strategy to climate change. df degree of freedom Amare et al. Agric & Food Secur (2018) 7:37 Page 6 of 11 Improved seeds logistic regression model results indicated that adult- The study communities adopt improved seeds (84.4%). headed households have a significant (p = 0.010) effect on However, the remaining households did not use improved adopting crop diversification (Table  4). This means adult seeds for one or another reason: financial constraints household heads (age 36–55) are 3.506 times more likely (42.6%), lack of skills (18%) on how to use, compatibility to use crop diversification than young-headed households problems with their farmland (16.4), small/no land at all (age 18–35). Adult-headed households (36–55) have a (13.1%), and lack of information (9.8%). Rural communi- significant (p = 0.010) effect on adopting improved seeds ties need all the support they can get to fight the adverse (p = 0.011). The beta coefficient (+ 1.415) shows posi- impacts of climate change and extreme weather events. tive relationships in explaining adopting improved seeds. Improved seed varieties developed by research institutes This indicates there is an increase in the log of odds using offer higher yields and stronger resistance to challenges improved seeds by 1.415 in adult-headed households related to climate change such as drought. Improved (HHH1). Exp (B) of 4.115 indicates that adult-headed seed tolerates weeds and other climate change-related households are 4.115 times more likely to use improved diseases. seeds than young- and old-headed households. Old- headed households have a significant (p = 0.038) effect on Changing planting dates adopting changing planting date (age > 55 years) (HHH2); Planting dates are growing season during which the Exp (B) of 5.985 indicates 5.985 times more likely to rainfall and temperature allow plants to grow. Among adopt changing planting date than (age 18–35). This indi - interviewed households, 78.9% used changing planting cates that, as age increases, the probability of adopting dates, whilst others do not use this method. Among those changing planting date as adaptation strategy increased. who did not use changing planting dates, 87.6% of the Adult- and old-headed households have a significant respondents attributed lack of skills as a barrier to adap- (p = 0.005; 0.014) power in explaining replanting failed tation methods, whilst 12.3% have not/small land size. crops. Age shows positive relationship (Beta = HHH1, + 1.689 and HHH2, + 3.470). This indicates that there is Replanting damaged crops an increase in the log of odds by 1.689 and 3.470. Adult- Weather events such as flooding, hailstorms, disease headed households (HHH1) Exp (B) of 5.416 indicate outbreaks can damage previously planted crops in all 5.4416 times more likely, and old-headed households or a portion of farm fields. This requires technical assis - (HHH2) Exp (B) of 32.143 times much more likely to use tance for decision-making in replanting. The major - replanting their failed crops than young-headed house- ity (92.5%) of respondents replant their failed crops. holds (age 18–35 years). The possible explanation is that However, among those who did not use, the majority age of household head increases the possibility of pursu- (63%) of households indicated lack of skills about future ing replanting failed crops as climate change adaptation weather forecast and economic return of the replanting strategy. crops, suitability of land and cropping season (7.4%) and small land size/no land (29.6%) contributes for not using Farming experience replanting damaged crops. Farmers in the range of farming experience 10–20  years (HHH1) have a significant effect on adopting improved Determinants of the choice of crop Management strategies seeds (p = 0.000). The beta coefficient indicates positive The determinant factors of the choice of adaptation strat - relationships in adopting improved seeds. This implies egies are presented in Table 2. Analyses were carried out there is an increase in the log of odds by 2.319 in using at p ≤ 0.05. improved seeds. The EXP (B) of 10.166 indicates that Changing crop management practice is one of the households having farming experience of 10–20 (HHH2) adaptation practices to climate change impacts. For are 10.166 times more likely to adopt improved seeds this study using crop diversification, improved seeds, than households having farming experience of fewer than changing planting date, and replanting failed crops were 10  years (HHH), 21–30  years (HHH2), and > 30  years selected in the context of the study sites. The applications (HHH3). of these strategies have been determined by a number The farming experience of 21–30  years and > 30  years of socio-economic, biophysical, and institutional factors has a significant effect (p = 0.007 and 0.018) on adopt- (Table 4). ing changing planting date. Farming experience of 21–30  years (HHH2, Beta = − 1.567) and > 30  years Age of the household head (HHH3, Beta = − 1.659) indicates that there is a decrease Crop diversification is considered as an important adap - in the log of odds by 1.567 and 1.659 (inverse relation- tation strategy to combat climate change impacts. The ships). The Exp (B) of 0.209 and 0.190 indicates farming Amare et al. Agric & Food Secur (2018) 7:37 Page 7 of 11 Table 4 Determinants of  households’ choice of  crop management strategies to  climate change impact. (Source: Computed from household survey, March–October (2016)) Explanatory variables Crop diversification Improved seed Changing planting dates Replanting failed crops p Exp (B) p Exp (B) p Exp (B) p Exp (B) Sex_HHH (1) .170 1.850 .983 .991 .055 .376 .104 .332 Age_HHH .035 .039 .102 .008 Age_HHH (1) .010* 3.506 .011* 4.115 .092 2.360 .005* 5.416 Age_HHH (2) .110 3.662 .111 3.886 .038* 5.985 .014* 32.143 Edu_HHH (1) .731 1.149 .226 1.690 .725 .867 .456 1.513 Farm_exp_HHH .233 .004 .046 .603 Farm_exp_HHH (1) .108 2.384 .000* 10.166 .120 .428 .261 2.246 Farm_exp_HHH (2) .884 .917 .238 2.100 .007* .209 .496 1.796 Farm_exp_HHH (3) .533 .636 .078 3.943 .018* .190 880 .864 Income_HHs .000 .013 .097 .697 Income_HHs (1) .000* 8.481 .012* 3.408 .639 1.378 .558 1.490 Income_HHs (2) .000* 17.510 .029* 3.632 .051 4.163 .241 2.701 Income_HHs (3) .000* 18.539 .002* 9.064 .281 2.275 .444 1.998 Family_size_HHs (1) .820 .902 N/C N/C .416 1.490 .002* .101 Weather_info_HHH (1) .364 .668 .182 .526 .102 .533 .774 .848 Farmer to farmer extension (1) .809 1.118 .424 1.455 .658 .806 .093 3.286 Government experts extension (1) .003* .271 .635 .815 .124 .520 .064 .290 Agro-ecol._HHs .005 .000 .000 .022 Agro-ecol._HHs (1) .036* 4.082 .055 5.446 .002* 5.412 .200 2.922 Agro-ecol._HHs (2) .093 .496 .002* .218 .000* 145.815 .006* 11.247 Farmlandsize (1) .001* 4.286 .865 .931 .000* .150 .02* 4.570 Crop failure N/C N/C .205 1.656 .044* .345 .211 0.146 Constant .121 .245 .305 .347 .002 23.795 .894 1.196 N/C not computed, HHH household head, HHs households *Significant at 0.05 Family size experience of (HHH2) only 0.209 times and (HHH3) only Family size has a significant (p = 0.002) effect on adopt - 0.190 times (much less) likely to adopt changing planting ing replanting failed crops. Family size has an inverse date. (Beta = − 2.297) in the log of odds by 2.297. The Exp (B) of 0.101 indicates households having family size > 4, only 0.101 times (much less) likely to replant failed crops. The Income inverse of Exp (B) of 0.101 indicates small family sizes Income has a positive and significant (p = 0.000) effect (< 4) are 9.91 times (much more) likely to replant failed on adopting crop diversification. Exp (B) of income crops than large family sizes. (10,001–30,000) 8.481 times, income (30,001–50,000) 17.510 times, and income (> 50,000) 18.539 times Access to government experts’ extension services more likely to diversify crops than low-income groups Formal extension services from government experts have (< 10,000). Income of households has a positive (Beta, a significant (p = 0.003) effect on adopting crop diversifi - + 1.226, + 1.290, and + 2.204) and significant (p = 0.012, cation to combat climate change impacts. The beta coef - 0.029, and 0.002) effect on adopting improved seeds. ficient shows an inverse relation (− 1.306) in adopting This indicates there is an increase in the log of odds by adaptation strategies. This indicates households who did 1.226, 1.290, and 2.204 in adopting improved seeds as an not get extension service; there is a decrease in the log of adaptation strategy. The EXP (B) of income (HHs1) 3.408 odd in diversifying crops by 1.306. The Exp (B) of 0.271 times, income (HHs2) 3.632 times, and income (HHs3) indicates that households who did not get formal exten- 9.064 times is more likely to adopt improved seeds than sion services were only 0.271 times (i.e. much less) likely low-income households (< 10,000). Amare et al. Agric & Food Secur (2018) 7:37 Page 8 of 11 to diversify crops than households who have got exten- having farmland size > 1.2 hectares are 4.286 times more sion service during 12 months of the year. likely to diversify crops than households with < 1.2 hec- tare farmland. Farmland size (> 1.2 hectares) of house- Agro-ecology settings holds has an inverse (beta, − 1.898) and significant Significant variation in the adoption of crop diversifica - (p = 0.000) effect on adopting changing planting date. tion was observed across agro-ecological zones (mid- This indicates a decrease (inverse relationships) in the land, p = 0.036). For example, higher crop diversification log of odds by 1.898 in changing planting date. The Exp was identified in the midland than highland and lowland (B) of 0.150 indicates households having large farmland agro-ecological settings. The Exp (B) of 4.082 indicates size (> 1.2 hectares) are only 0.150 times changed their households who live in the midland 4.082 times more planting date. Households having small land size( < 1.2 likely to use crop diversification than households reside hectare) are 6.76 times more likely to use changing plant- in highland. The Exp (B) of 0.496 indicates the lowland ing date. Farmland size has a positive relationship (Beta, households are only 0.496 times much less likely diversi- + 1.519) with no significant effect on adopting replant - fying crops than highlands. This means highland resident ing failed crops. There is an increase in the log of odds households are 2.016 times more likely to diversify crops by 1.519. The Exp (B) of 4.570 indicates households hav - than lowland households (i.e. invert, 1/0.496 = 2.016). ing > 1.2-hectare land are 4.2570 times more likely to The lowland agro-ecology with an inverse beta value replant failed crops than households with < 1.2 hectares. (− 1.552) has a significant effect (p = 0.002) on adopt- ing improved seeds. This indicates a decrease (1.522) in Crop failure history of households the log of odds on adopting improved seeds in the low- Crop failure has a significant (p = 0.044) effect on adopt - land agro-ecology zones. The EXP (B) of 0.218 indicates ing changing planting date. Households who never faced the lowland households are only 0.218 times (much less) crop failure in the past 10  years have an inverse rela- likely to use improved seeds than highland households. tionship (Beta = − 1.064) in employing changing plant- This indicates that highland households are 4.587 times ing date. The Exp (B) of 0.345 indicates households who (much more) likely to use improved seeds than lowland never faced crop failure in the past 10 years are only 0.345 resident households (inverse of 1/0.218 = 4.587). times adopted changing planting date. The invert of Exp The midland and lowland agro-ecologies have a sig - (B) of 0.345 is 2.8986, which indicates households who nificant effect on changing cropping date (p = 0.002 and faced crop failure are 2.8986 times more likely to change p = 0.000), respectively. The coefficient of beta (+ 1.689 planting date. This implies most farmers learn only when midland (HHs1) and + 4.982 lowland (HHs2)) indicates they faced problems. there is an increase in the log of odds by 1.689 and 4.982 on adopting changing planting date. The Exp (B) of 5.412 Discussion indicates that households who reside in the midland The rural communities of Dejen District adopt crop man - agro-ecology are 5.412 times much more likely to change agement strategies to combat climate change impacts. their cropping date than highland households. However, the key barriers identified in the study district The mid- and lowland agro-ecologies have a signifi - were shortage of money, lack of access to information, cant (p = 0.006; 0.020) effect on adopting replanting and small land size. Previous studies (e.g. [22–24]) stated failed crops. There is an increase (Beta =+ 1.072 midland that financial barriers are one of the barriers that restrict (HHs2) and + 2.420 lowland (HHs2)) in the log of odds implementation of adaptation strategies. This implies by 1.072 and 2.420. The Exp (B) of 2.922 indicates house - every form of adaptation requires some direct or indirect holds who live in the midland agro-ecologies are 2.922 costs. For instance, the use of improved varieties of crops times more likely to use replanting than highland house- has been reported as one of the key adaptation strate- holds. The Exp (B) of 11.247 indicates the lowland house - gies for farmers in Dejen District, Nile Basin of Ethiopia, holds are 11.247 times (much more) likely to replant their where this study confirmed. In the context of this study, failed crops. improved seeds include high yielding varieties, drought tolerant, short maturing, pest- and disease-resistant spe- Farmland size cies either induced or indigenous. When improved seed Farmland is the most significant (p = 0.001) factor to varieties are available, their price may be prohibiting diversify crops in the study communities. The beta coef - making it difficult for many rural households to access. ficient of households having farm size of > 1.2 hectare u Th s, framers have often sought to use their own saved has a positive relation to diversify crops. This implies seeds. One of the possible causes of financial barriers in there is an increase in the log of odds in diversifying the study area could be due to lack of credit facilities to crops by 1.455. The Exp (B) of 4.286 means households rural communities. Amare et al. Agric & Food Secur (2018) 7:37 Page 9 of 11 Access to information on weather and climate change Family size is negatively and significantly associated is an important tool that can be used to enhance the with the households’ decision to pursue replanting failed adaptation and implementation of adaptation strategies crops. Households who have large family size are sup- by rural communities of the study area. Access to infor- posed to have an opportunity in pursuing various adap- mation is particularly important for Africa [25] and Ethi- tation options to combat impacts of clime change and opia in particular, where there are few climate projections variability. This argument is raised by previous studies due to lack of appropriate climate data. This is crucially [26, 27] who argued that large family size is associated important, considering that most farming systems in with higher labour endowment which would enable a Dejen District depend on rain-fed agricultural systems. household to accomplish various agricultural tasks. The Hence, lack of appropriate climate information could be possible reason for an inverse relationship might be due crucial for rural communities’ food security. to the fact that community’s expectation of the benefits Age of household heads significantly determined crop of using adaptation strategy. In this regard, Barungi and diversification, improved seeds and changing planting Maonga [28] based on the rational choice theory; argue date, and replanting failed crops. Crop diversification that the behaviour of human beings is motivated by the and replanting of the failed crops require more energy possibility of gaining benefit. The possible explanation and experience. Thus, adult household heads are more could be households who have large family size have the matured and active in sowing different crops than old possibility to engage in off-farm activities, and they will and young household heads. The possible reason for the ignore the failed crops to replant. Therefore, rural com - positive and significant association is due to the fact that munities are rationale consumers of new technologies, age is the proxy indicator that may likely to endow the and they will only adopt technology as they foresee it will farmers with the requisite experience that enables them result in increased productivity. to make a better decision in the choice of crop manage- Access to government extension services has a nega- ment strategies. This is in line with studies by Deressa tive and significant association with the likelihood of et  al. [26] who found that an increase in age of house- choosing crop diversification to combat climate change hold head does mean an increase in farming experience impacts. This result is in contrary with previous stud - which would increase rural communities’ local knowl- ies [29, 30] who noted that farmers who obtain agricul- edge to respond to hazards resulted in climate change tural extension services through extension workers are and variability. more likely informed about the climatic situation and Farming experience is one of the significant variables the responses followed. The contributing factors for that affect the rural communities’ choice of adaptation this inverse relationship could be barriers to adopting strategies. Farming experience is a proxy indicator of age. crop diversification such as inadequate extension ser - Like crop diversification, the middle age household heads vices, constraints of money, labour, skills, and farmland have ability and willingness to adopt improved seeds locations. to adjust climate change impacts. This implies as one Households who live in the midland and highland agro- become more experienced in farming, the probability of ecologies have a significant and positive effect on adop - one to use improved seeds increases more than a farmer tion of crop diversification. This is because the suitability with less farming experience. of highland agro-ecology to sow different types of crops On the other hand, farming experience has an inverse and access to government extension services due to prox- relationship with changing planting date. The reason for imity to the administration. For instance, in this study an inverse relationship might be that experienced farm- finding, the midland agro-ecology has got more access to ers will have access to irrigation and water harvesting for extension services (77%) than the lowland agro-ecology their agricultural activities and plant their seeds without (47%) communities by the government extension experts changing the planting date. This implies a farmer with in the past cropping season. more experience would know when climate variability The lowland agro-ecology resident households have a is occurring in the area and which method of adapta- negative and significant effect on adoption of improved tion strategies works well in that specific agro-ecology seed varieties. The possible explanation is that lowland zone. As expected, income is positively and significantly households did not use improved seeds because of suit- associated with the household decision to pursue crop ability problem of the lowland agro-ecology and topog- diversification and improved seeds. This means crop raphy to use improved seeds to their farmland. This was diversification and purchasing of improved varieties of confirmed by households report on the barriers to adopt seeds require money. This implies the rate of using crop improved seeds as crop management strategy. On the diversification and improved seeds is increased as income other hand, the lowland agro-ecology has a positive and of households increased. significant effect on pursuing changing planting dates Amare et al. Agric & Food Secur (2018) 7:37 Page 10 of 11 to combat climate change impacts. This is because low - the government bodies in the office of agriculture did land agro-ecologies are characterized by erratic rainfall not realize the problems. This implies, in the process and other extreme events that lead households to change of diffusion of adaptation strategies, climate change their planting date. The midland and lowland agro-ecol - adaptation process should require close collaboration ogies have a significant effect on employing replanting and active participation of climate change research- failed crops as crop management strategy to combat ers, decision-makers, policy analysts, the community, climate change impacts. This is due to the fact that the and partners. Government policies should strengthen midland and lowland households are characterized by cli- the current adaptation strategies practised by rural mate variability such as erratic rainfall than the highland community households and support the adoption of agro-ecology households. The exposure of climatic vari - crop management strategies. Besides, the less adopted ability gave them more experience in adopting replanting crop management strategies due to financial con- their failed crops than highland households. straints should be subsidized by government and aid As expected, farm size has a significant and positive organizations. This study contributes to the academic effect on adopting crop diversification to combat climate discourse on climate change impact adaptations by change impacts. Households with larger farm sizes were providing empirical evidence to deepen understanding more likely to diversify their crops. On the other hand, of the barriers and determinants that confronts rural larger farmland size has a negative and significant effect communities in their attempt to implement adaptation on using changing planting dates as crop management strategies to manage the negative impacts of climate strategies. This means households having small land size change and variability. (< 1.2 hectares) are more likely use changing planting dates. This reminds us “a hunter who has only one arrow Abbreviations does not shoot with careless aim”. This implies house - df: degree of freedom; HHs: households; HHHs: household heads; NC: not holds who have small land size took care of their farm- computed; m.a.s.l: metres above sea level; VIF: variance inflation factor. land and changed their planting dates when there is a Authors’ contributions change in weather conditions. Even if farmland size has ZYA designed the data collection tools, conducted fieldwork and analysis, no significant effect on changing failed crops, it shows a and developed the manuscript. JOA contributed in commenting the data collection tools, recommending data analysis methods, reviewed and made positive effect on using changing failed crops. The pos - editorial comments on the draft manuscript. IOA contributed in commenting sible explanation could be the more farmland plot they the data collection tools, data analysis methods, reviewed and made editorial have, the more is the probability of having failed crop- comments on the draft manuscript. MTZ contributed to developing the data collection tools, commenting data analysis methods, reviewed, and made lands that could lead them to replant their failed crops. editorial comments on the draft manuscript. All authors read and approved the final manuscript. Conclusion and policy recommendations Author details The study communities have tremendous ideas to adapt Institute of Disaster Risk Management and Food Security Studies, Bahir Dar University, BahirDar, Ethiopia. Department of Geography, Faculty of the Social for current and future climate change impacts with a Sciences, University of Ibadan, Ibadan, Nigeria. Department of Geography strong motivation to move out of poverty. However, and Environmental Studies, Debre Tabor University, Debre Tabor, Ethiopia. the mere willingness to adopt climate change adapta- Acknowledgements tion strategies was not enough. Their ability to adopt is This study was made possible by the financial support of Pan-African Univer - constrained by many internal and external factors. Rural sity (PAU), a continental initiative of the African Union Commission (AU), Addis communities who did not employ adaptation strate- Ababa, Ethiopia. Further, authors would like to thank the data collectors and field assistants for their effective coordination, support, and time spent in gies gave many reasons for their failure to adopt. These organizing and conducting successful household interviews. Special thanks to includes poor or no access to water sources, limited the Dejen District rural communities who willingly volunteered the informa- knowledge, and skill, shortage of labour, lack of and/or tion in this study. shortage of farmland, lack of money, lack of information, Competing interests lack of agricultural extension services, and other institu- The authors declare they have no competing interests. tional factors. Availability of data and materials The most significant determinant factors of the The datasets used and/or analysed during the current study available from the choice of crop management strategies were age, farm- corresponding author on reasonable request. ing experience, income, agro-ecology setting, and Consent for publications farmland size. Agro-ecology setting has a significant The authors obtained permission from all participants in Dejen District, to effect on all adaptation strategies. Due to the soil publish their data. characteristics, the lowland agro-ecology zones were not suitable for adopting improved seeds. However, Amare et al. Agric & Food Secur (2018) 7:37 Page 11 of 11 Ethics approval and consent to participate 12. DDEPO. Dejen District Environmental Protection Office (DDEPO), East Consent to participate was received from everyone interviewed in Dejen Gojjam zone, Dejen, Ethiopia; 2016. District, Ethiopia. The Pan-African University, Life and Earth Sciences Institute 13. Yamane T. Statistics: an introductory analysis. New York: Harper and Row; (PAULESI) (including Health and Agriculture), University of Ibadan were 1967. informed of the study. 14. DDFEDO. Dejen District Finance and Economic Development Office (DDFEDO) Population Projection (DDFED, 2014); 2014. Funding 15. Bryan E, et al. Adapting agriculture to climate change in Kenya: house- This study was sponsored by the Pan-African University (PAU), a continental hold strategies and determinants. J Environ Manag. 2013;114:26–35. initiative of the African Union Commission (AU), Addis Ababa, Ethiopia. 16. Abid M, et al. Farmers’ perceptions of and adaptation strategies to climate change and their determinants: the case of Punjab province, Pakistan. Received: 11 November 2017 Accepted: 22 May 2018 Earth Syst Dyn. 2015;6(1):225. 17. Agresti A. An introduction to categorical data analysis. 2nd ed. Hoboken: Wiley; 2007. 18. NBE National Bank of Ethiopia. Commercial Banks’ Exchange rate; 2016. 19. Kothari CR. Research methodology: methods and techniques. Chennai: New Age International; 2004. References 20. SPSS. Statistical Package for the Social Sciences (SPSS). Version20. 1. WMO. Statement on the status of the global environment, vol. 1152. 21. Kothari GG. Research methodology. 3rd ed. New Delhi: New Age Interna- Geneva: Japan Meteorological Agency, in Cooperation with the World tional Publishers; 2014. Meteorological Organization; 2015. p. 250. 22. Bryan E, et al. Adaptation to climate change in Ethiopia and South Africa: 2. FAO. Climate change and food safety: a review. Food Res Int. options and constraints. Environ Sci Policy. 2009;12(4):413–26. 2010;43(7):1745–65. 23. Kithiia J. Climate change risk responses in East African cities: need, barri- 3. Foresight U. The future of food and farming. Final project report. London: ers and opportunities. Curr Opin Environ Sustain. 2011;3(3):176–80. The Government Office for Science; 2011. 24. Peterson C. Fast-growing groundnuts keep Ghana’s farmers afloat amid 4. Füssel H-M, Klein RJ. Climate change vulnerability assessments: an evolu- climate shifts. Retrieved July, 2013, vol. 16, p. 2013. tion of conceptual thinking. Clim Change. 2006;75(3):301–29. 25. IPCC. Climate change: climate change impacts, adaptation and vulner- 5. Rahman MI-U. Climate change: a theoretical review. Interdiscip Descr ability. The fourth assessment report of the Intergovernmental Panel on Complex Syst. 2013;11(1):1–13. Climate Change. Geneva, Switzerland; 2007. 6. Gbetibouo GA. Understanding farmers’ perceptions and adaptations to 26. Deressa TT, et al. Determinants of farmers’ choice of adaptation methods climate change and variability: the case of the Limpopo Basin, South to climate change in the Nile Basin of Ethiopia. Glob Environ Change. Africa, vol. 849. Washington: The International Food Policy Research 2009;19(2):248–55. Institute; 2009. 27. Menberu TZ, Aberra Y. Determinants of the adoption of land manage- 7. Segele ZT, Lamb PJ. Characterization and variability of Kiremt rainy season ment strategies against climate change in Northwest Ethiopia. Ethiop over Ethiopia. Meteorol Atmos Phys. 2005;89(1):153–80. Renaiss J Soc Sci Humanit. 2014;1:93–118. 8. NMA. Climate change national adaptation programme of action (Napa) 28. Barungi M, Maonga BB. Adoption of soil management technologies by of Ethiopia. National Meteorological Services Agency (NMA), Ministry of smallholder farmers in central and southern Malawi. J Sustain Dev Afr. Water Resources, Federal Democratic Republic of Ethiopia, Addis Ababa; 2011;13(3):28–38. 29. Maddison D. The perception of and adaptation to climate change in 9. You GJ-Y, Ringler C. Hydro-economic modeling of climate change Africa, vol. 4308. Washington: World Bank Publications; 2007. impacts in Ethiopia. International Food Policy Research Institute (IFPRI); 30. Nhemachena C, Hassan R. Micro-level analysis of farmers adaption to climate change in Southern Africa. Washington: The International Food 10. Schröter D, Polsky C, Patt AG. Assessing vulnerabilities to the effects of Policy Research Institute; 2007. global change: an eight step approach. Mitig Adapt Strat Glob Change. 2005;10(4):573–95. 11. DDARDO. Dejen District Agricultural and Rural Development Office (DDRDO), Annual Report, East Gojjam zone, Dejen, Ethiopia; 2016. Ready to submit your research ? Choose BMC and benefit from: fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Agriculture & Food Security Springer Journals

Barriers to and determinants of the choice of crop management strategies to combat climate change in Dejen District, Nile Basin of Ethiopia

Free
11 pages
Loading next page...
 
/lp/springer_journal/barriers-to-and-determinants-of-the-choice-of-crop-management-bamjLRVccG
Publisher
BioMed Central
Copyright
Copyright © 2018 by The Author(s)
Subject
Life Sciences; Agriculture; Biotechnology; Plant Sciences; Ecology; Agricultural Economics; Epidemiology
eISSN
2048-7010
D.O.I.
10.1186/s40066-018-0188-y
Publisher site
See Article on Publisher Site

Abstract

Background: Climate change without adaptation is projected to impact strongly the livelihoods of the rural com- munities. Adaptation to climate change is crucial for least developed country like Ethiopia due to high population and dependency on agriculture. Hence, this study was initiated to examine the barriers to and determinants of the choice of crop management strategies to combat climate change. The Intergovernmental Panel on Climate Change concepts of climate change response provided the framework. Stratified and snowball sampling techniques were employed to select a sample of 398 households. The household survey was employed to collect data on current adaptation strategies. Logistic regression was used to analyse the determinants of the choice of adaptation strategies. Logistic regression analyses were carried out at p ≤ 0.05. Results: Small farmland size, agro-ecology, farmland location, financial constraints, and lack of skills were the major barriers to adoption of crop management strategies. Age, farming experience, income, family size, government experts’ extension services, agro-ecology setting, and crop failure history of households significantly affect the choice of most of the crop management strategies. Conclusions: Socio-economic and institutional factors determined rural communities’ ability and willingness to choose effective adaptation strategies. Policy priority should be given based on agro-ecology and households demand of policy intervention such as providing extension services and subsidizing the least adopted strategies due to financial constraints. Keywords: Adaptation, Climate change impact, Crop management practices, Nile Basin of Ethiopia Background average of 14 °C. It was 0.08 °C above the average anom- The warming trends observed over the past few decades aly of 0.50 °C for the past 10 years (2005–2014) [1]. continued in 2014. World Meteorological Organization According to Food and Agricultural Organization [2], (WMO) has ranked as nominally the warmest year since due to climate change and variability almost one billion modern instrumental measurements began in the mid- people experienced hunger in 2010 globally. This implies 1800s [1]. The global average near-surface temperature the most marginalized people cannot access enough of for 2014 was comparable to the warmest years in the the primary macronutrients. Perhaps, other billions are 165-year instrumental record. In 2014, the global aver- thought to suffer from hidden hunger, in which essential age temperature was 0.57 ± 0.09 °C above the 1961–1990 micronutrients are missing from their diet, with conse- quent risks of physical and mental impairment [3]. The majority (85%) of the Ethiopian population is dependent *Correspondence: Zerihun.yohannes19@gmail.com on agriculture. As a result, agriculture will continue to be Institute of Disaster Risk Management and Food Security Studies, Bahir Dar University, BahirDar, Ethiopia Full list of author information is available at the end of the article © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/ publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Amare et al. Agric & Food Secur (2018) 7:37 Page 2 of 11 the most important sector in its need to adapt to climate of 1071 and 3000  m above sea level (m.a.s.l). The aver - change. age temperature and total annual rainfall of the district There are many reasons and convincing arguments for range between 20 and 24  °C and 800 and 1200  mm, a more comprehensive consideration of adaptation as respectively. Dejen District is located in the valley of the a response measure to climate change. Firstly, given the Blue Nile which is highly undulated topography and fre- amount of past greenhouse gas emissions and the inertia quent susceptibility to climate-related problems such as of the climate system, we are already bound to some level erratic rainfall, crop pests, livestock diseases, and malaria of climate change, which can no longer be prevented outbreaks. In the past 8  years (2009–2016) due to cli- even by the most ambitious emission reductions [4]. Sec- mate change impacts, the district loses 50,555 quintals of ond, the effect of emission reductions takes several dec - crops, which has the potential to feed 27,701 individuals. ades to fully manifest, whereas most adaptation measures The study district is categorized into three agro-ecolog - have more immediate and sustainable benefits [5]. ical zones, 41% Dega (highland), 31% Woinadega (mid- Third, adaptations can be effectively implemented on land), and 28% Kolla (lowland) [11, 12]. a local or regional scale such that its efficiency is less dependent on the actions of others, whereas mitiga- Research design and sampling procedure tion of climate change requires international coopera- The study employed cross-sectional research design with tion. Fourth, most adaptations to climate change also both quantitative and qualitative research methods. This reduce the risks associated with current climate variabil- study used a multi-stage sampling technique to select the ity, which is a significant hazard in many world regions. agro-ecology, Kebeles (the lower administrative unit next According to Gbetibouo [6], there are two adaptation to district), and households. At the first stage, Dejen Dis - assessment approaches, namely top-down and bottom- trict of the Nile Basin was selected purposely due to its up assessment approaches. The top-down approach highly undulated topography and frequent susceptibil- starts with climate change scenarios, and estimates ity to extreme events and representativeness of the three impact through scenario analysis, based on which pos- agro-ecological zones such as highland, midland, and sible adaptation practices are identified. Most of the lowland. In the second stage, six Kebeles (two from each top-down adaptations represent possible or potential agro-ecological setting) were selected purposely based on measures, rather than those that have been used [6]. the above-listed district selection criteria. Most studies, e.g. [7–9], carried out in Ethiopia and Climate change affects the rural communities differ - Africa using top-down approach predicted the impact ently in different agro-ecological zones. As a result, com - of climate change on the agricultural sector with adverse munities’ knowledge and skill to adapt to the climate effects on crop yields. The bottom-up approach takes a change impacts varies from different agro-ecological set - vulnerability perspective where adaptation strategies are tings. In the third stage, stratified sampling was employed considered more as a process involving the socio-eco- to select households. Under the stratified sampling, the nomic, and policy environments, and elements of deci- population was divided into male- and female-headed sion-making [6]. In line with this notion, Schröter et  al. households, and then the sample was selected from each [10] argue that in choosing adaptation options to climate male- and female-headed household to constitute a rep- change and developing policies to implement these pos- resentative sample. sibilities the affected community should actively par - The sample size was determined proportionately ticipate. This study adopts the bottom-up approach that (Table 1). In Ethiopia context, female-headed households seeks to identify actual adaptation strategies at the local are those who do not have husband due to either being level and the factors that determine the choice of crop divorced, widowed, or separated. In Ethiopia, in some of management strategies in Dejen District, Nile Basin of the rural communities, disclosing of the marital status of Ethiopia. older females is culturally not feeling comfort them. Thus, to get female-headed households, snowball sampling was Methods employed. Based on Yamane [13] at the 95% confidence The study area interval and 5%, level of precision, 398 households were The study district is located in west-central Ethiopia selected at the six Kebeles of the district (Table 1). (Fig. 1) at a road distance of 335 km south of the regional state capital, Bahir Dar City, and 230  km northwest of Data sources and data collection methods the capital city of Ethiopia, Addis Ababa, in the Amhara The primary and secondary data sources were quantita - Regional State at the edge of the canyon of the Blue Nile. tive and qualitative in nature. The study used two main The district lies between longitude 38°6′E and 38°10′E, data sources to analyse the barriers and determinants of and between latitude 10°7′N 10°11′N, with an elevation the adoption of crop management strategies to climate Amare et al. Agric & Food Secur (2018) 7:37 Page 3 of 11 Fig. 1 Study area change impacts. The main data source was the socio-eco - language for data processing and analysis. The house - nomic data collected through household survey. hold survey was initially pretested to check the appro- priateness and validity of the questions from the three agro-ecological zones. Pretested Kebeles and partici- Household survey pant households were not involved in the actual survey. The household survey was used to collect quantitative After pretesting, ambiguous words were rephrased data on households’ current adaptation strategies, the and inappropriate questions were modified. The expe - barriers to adapt, and factors that affect rural com - rience of the author in the study area played a para- munities’ choice of crop management strategies. The mount role in choosing the data collectors who have survey questions were prepared in English language. been working for many years in the rural community in It was translated into local language (Amharic) during the field of agriculture, environment, and land admin - data collection and then encoded into SPSS in English istration. The data were collected by the trained data Amare et al. Agric & Food Secur (2018) 7:37 Page 4 of 11 Table 1 Distribution of  sample size based on  the  total logistic regression was inappropriate. To fix such prob - study population. (Source: computed based on [14]) lems, the possible remedy suggested by Bryan et  al. [15] was to combine similar measures into single categories. Sample Number % of study Share of Kebeles Kebeles of household heads population from 398 HHs However, such grouping into self-defined categories may lead to miss interpretation [15, 16]. Besides, the set of Male Female Total explanatory variables affecting the households’ decision Koncher (high- 995 225 1220 16.5 66 was also expected to be different for different adaptation land) strategies. Therefore, this study employed binary logis - Yetnora (high- 1388 767 2155 29 115 tic regression technique to analyse the determinants of land) the choice of crop management strategies. Analysis of Zemeten (mid- 652 280 932 12.6 50 the logistic regression of binary response model can be land) defined as: Borebor (mid- 644 281 925 12.5 50 land) β0+β1(x1)+β2(x2)+β3(x3)+βn(xn) Kurar (lowland) 726 363 1089 14.7 59 y = e , −[β0+β1(x1)+β2(x2)+βn(xn) Gelgele (low- 732 345 1077 14.6 58 (1) land) 5137 2261 7398 100 398 where y is response variable; y = 1, outcome is present HHs Households (adopt); y = 0, outcome is not present (not adopt); β is constant. β + β + β + ··· β are regression coefficients that 1 2 3 n collectors under close supervision of the author in the explain the change in the log odds for each unit change period March to October 2016. in x. x + x + ··· x = set of predictor variables included 1 2 n in the model (Table  2). e represents the change in the Methods of data analysis odds of the outcome (multiplicatively) by increasing x by The descriptive statistics such as percentage and fre - 1 unit. The current crop management adaptation strate - quencies were used to summarize and categorize the gies in the study area, such as using crop diversification, information gathered from households. The binary logis - improved seeds, changing planting date, and replant- tic regression was used to analyse determinants of the ing failed crops, were used as dependent variables. The choice of crop management strategies. choices of these adaptation strategies were determined by Before the data collection, a multinomial logistic socio-economic and demographic characteristics, insti- (MNL) model was proposed based on aspects of the tutional factors, and agro-ecological settings (Table  2). literature. However, in the course of this study, the sur- Based on Agresti [17], logistic regression method can be veyed households chose more than one adaptation strate- used when the dependent variable Y is dichotomous. Y gies simultaneously. As a result, the use of multinomial Table 2 Description of  explanatory variables that  affect households’ choice of  crop management strategies. Source: Author (2016) Explanatory variables Description Sex Dummy, 1 = Male, 0 = Female Age Discrete (years), 18–35, 36–55, > 55 Education Discrete (years), cannot read and write, primary school and above) Farming experience Continuous (years), 10, 10–20, 21–30, > 30 Income Continuous (ETB), < 10,000, 10,001–30,000, 30,001–50,000, > 50,000 Family size Discrete (number), < 4, and > 4 Access to weather information Dichotomous, 1 = Yes, 0 = No Access to farmer to farmer extension services Dichotomous, 1 = Yes, 0 = No Access to government experts’ extension services Dichotomous, 1 = Yes, 0 = No Agro-ecology settings Dummy, 1 = highland; 2 = midland; 3 = lowland Farmland size Continuous (hectare), < 1.2, and > 1.2 Crop failure history of households Dichotomous, 1 = Yes, 0 = No Taken as reference (base for analysis), ETB is Ethiopian currency ($1 = 22.3ETB) [18] Amare et al. Agric & Food Secur (2018) 7:37 Page 5 of 11 is coded as 1 when the outcome is present and coded 0 independent variable is lower than the reference/base when the outcome is not present. variable (Table 2). Multicollinearity was assessed by examining toler- Test of goodness of fit and multicollinearity ance and variance inflation factors (VIF). The variance The fitness of the logistic regression model to the data inflation factor (VIF) quantifies how much the variance was measured by using the SPSS classification table is inflated. The tolerance is the percentage of the vari - (crosstabs) and the Hosmer–Lemeshow test. Besides, ance in a given predictor that cannot be explained by collinearity among predictor variables was checked using the other predictors. When VIF > 5, X (the explanatory multicollinearity statistics. The Hosmer–Lemeshow test variable) is highly correlated with the other explanatory used 95% confidence interval (CI) and asymptotically fol - variables [20, 21]. Mathematically, variance inflation lows a X distribution. factor (VIF) can be expressed by: 1 2 Empty cells or small frequency was checked by doing VIF = , where R j is the coefficient of determina - 1−R j crosstabs between categorical predictor variables and the outcome variables. When the cell has very few cases, the tion of a linear regression model that uses X as the model becomes unstable. Based on Kothari [19] and SPSS response variable and all other X variables as the [20], the Hosmer–Lemeshow statistics indicate a poor fit explanatory variables. Tolerance is the reciprocal of if the significance value (p) is less than 0.05 (Table 3). VIF (i.e. tolerance = ). A tolerance of less than 0.20 VIF Techniques to remedy poor fit of the model were by: and/or a variance inflation factor (VIF) greater than 5 (1) using re-categorized (for example educational levels and above indicates a multicollinearity problem. How- were categorized as cannot read and write and primary ever, in this study, the results of the variance inflation school level and above), (2) dropping the least theoreti- factor indicated that there was no multicollinearity cally important explanatory variables that contribute to problem. the model a poor fit to the data. For example, explanatory variables such as family size and crop failure history of households were excluded from entering and competing Results in the model in the improved seed and crop diversifica - Barriers to crop management strategies tion adaptation options, respectively. Implementation of crop management strategies used by In the logistic regression model, the Exp (B) is the the rural communities varied among households. The “Odds ratio” which explains the effect of the independ - study identified a number of constraints faced by the ent variable (X ) on the dependent variable. The beta households to adopt crop management strategies to com- coefficient (β ) is the estimated logit coefficient which bat climate change impact. is the rate of change in the Y (the dependent variables) as X (independent variable) changes. When the beta coefficient (β ) is negative, it shows that the dependent Crop diversification and independent variables have an inverse relationship, The major constraints identified by respondents to not and when it has a positive coefficient, there is a positive to adopt crop diversification were: small land size (57%) relationship. Odds ratio = 1 indicates the same proba- followed by soil fertility decrement (25.3%), shortage bility of an event occurring between the two situations. of money to buy some expensive crop varieties (1.3%), Odds ratio > 1 probability of an event occurring with shortage of labour (2.5%) to implement some labour- a unit increase in the independent variable is higher intensive farming practices, and they prefer the easiest than the reference/base variable. Odds ratio < 1 prob- crop type, lack of skill (3.8%) to sow different crops and ability of an event occurring with a unit increase in the stick to one types of crop, topography of farmland loca- tion (2.5%) which permits only some crop types to grow, and the remaining (2.5%) do not/have small land size. Table 3 Hosmer and  Lemeshow goodness-of-fit test The majority (58%) of rural communities have less than results of  logistic regression model. (Source: Computed 1.2 hectares. As a result, small land size/no land at all based on household survey data, March–October (2016)) consequences is barrier to the choice of crop manage- Dependent variables Chi-square df P value (> 0.05) ment strategies. On the other hand, rural communities Crop diversification 7.763 8 0.457 who are located in the lowlands of the undulated topog- Improved seeds 10.78 8 0.214 raphy encountered farm soil fertility decrement. Due to Changing planting date 9.239 8 0.323 this reason, farmers were obliged to leave their farmland (fallowing) for some periods instead of diversifying dif- Replanting failed crops 2.416 8 0.966 ferent crops as an adaptation strategy to climate change. df degree of freedom Amare et al. Agric & Food Secur (2018) 7:37 Page 6 of 11 Improved seeds logistic regression model results indicated that adult- The study communities adopt improved seeds (84.4%). headed households have a significant (p = 0.010) effect on However, the remaining households did not use improved adopting crop diversification (Table  4). This means adult seeds for one or another reason: financial constraints household heads (age 36–55) are 3.506 times more likely (42.6%), lack of skills (18%) on how to use, compatibility to use crop diversification than young-headed households problems with their farmland (16.4), small/no land at all (age 18–35). Adult-headed households (36–55) have a (13.1%), and lack of information (9.8%). Rural communi- significant (p = 0.010) effect on adopting improved seeds ties need all the support they can get to fight the adverse (p = 0.011). The beta coefficient (+ 1.415) shows posi- impacts of climate change and extreme weather events. tive relationships in explaining adopting improved seeds. Improved seed varieties developed by research institutes This indicates there is an increase in the log of odds using offer higher yields and stronger resistance to challenges improved seeds by 1.415 in adult-headed households related to climate change such as drought. Improved (HHH1). Exp (B) of 4.115 indicates that adult-headed seed tolerates weeds and other climate change-related households are 4.115 times more likely to use improved diseases. seeds than young- and old-headed households. Old- headed households have a significant (p = 0.038) effect on Changing planting dates adopting changing planting date (age > 55 years) (HHH2); Planting dates are growing season during which the Exp (B) of 5.985 indicates 5.985 times more likely to rainfall and temperature allow plants to grow. Among adopt changing planting date than (age 18–35). This indi - interviewed households, 78.9% used changing planting cates that, as age increases, the probability of adopting dates, whilst others do not use this method. Among those changing planting date as adaptation strategy increased. who did not use changing planting dates, 87.6% of the Adult- and old-headed households have a significant respondents attributed lack of skills as a barrier to adap- (p = 0.005; 0.014) power in explaining replanting failed tation methods, whilst 12.3% have not/small land size. crops. Age shows positive relationship (Beta = HHH1, + 1.689 and HHH2, + 3.470). This indicates that there is Replanting damaged crops an increase in the log of odds by 1.689 and 3.470. Adult- Weather events such as flooding, hailstorms, disease headed households (HHH1) Exp (B) of 5.416 indicate outbreaks can damage previously planted crops in all 5.4416 times more likely, and old-headed households or a portion of farm fields. This requires technical assis - (HHH2) Exp (B) of 32.143 times much more likely to use tance for decision-making in replanting. The major - replanting their failed crops than young-headed house- ity (92.5%) of respondents replant their failed crops. holds (age 18–35 years). The possible explanation is that However, among those who did not use, the majority age of household head increases the possibility of pursu- (63%) of households indicated lack of skills about future ing replanting failed crops as climate change adaptation weather forecast and economic return of the replanting strategy. crops, suitability of land and cropping season (7.4%) and small land size/no land (29.6%) contributes for not using Farming experience replanting damaged crops. Farmers in the range of farming experience 10–20  years (HHH1) have a significant effect on adopting improved Determinants of the choice of crop Management strategies seeds (p = 0.000). The beta coefficient indicates positive The determinant factors of the choice of adaptation strat - relationships in adopting improved seeds. This implies egies are presented in Table 2. Analyses were carried out there is an increase in the log of odds by 2.319 in using at p ≤ 0.05. improved seeds. The EXP (B) of 10.166 indicates that Changing crop management practice is one of the households having farming experience of 10–20 (HHH2) adaptation practices to climate change impacts. For are 10.166 times more likely to adopt improved seeds this study using crop diversification, improved seeds, than households having farming experience of fewer than changing planting date, and replanting failed crops were 10  years (HHH), 21–30  years (HHH2), and > 30  years selected in the context of the study sites. The applications (HHH3). of these strategies have been determined by a number The farming experience of 21–30  years and > 30  years of socio-economic, biophysical, and institutional factors has a significant effect (p = 0.007 and 0.018) on adopt- (Table 4). ing changing planting date. Farming experience of 21–30  years (HHH2, Beta = − 1.567) and > 30  years Age of the household head (HHH3, Beta = − 1.659) indicates that there is a decrease Crop diversification is considered as an important adap - in the log of odds by 1.567 and 1.659 (inverse relation- tation strategy to combat climate change impacts. The ships). The Exp (B) of 0.209 and 0.190 indicates farming Amare et al. Agric & Food Secur (2018) 7:37 Page 7 of 11 Table 4 Determinants of  households’ choice of  crop management strategies to  climate change impact. (Source: Computed from household survey, March–October (2016)) Explanatory variables Crop diversification Improved seed Changing planting dates Replanting failed crops p Exp (B) p Exp (B) p Exp (B) p Exp (B) Sex_HHH (1) .170 1.850 .983 .991 .055 .376 .104 .332 Age_HHH .035 .039 .102 .008 Age_HHH (1) .010* 3.506 .011* 4.115 .092 2.360 .005* 5.416 Age_HHH (2) .110 3.662 .111 3.886 .038* 5.985 .014* 32.143 Edu_HHH (1) .731 1.149 .226 1.690 .725 .867 .456 1.513 Farm_exp_HHH .233 .004 .046 .603 Farm_exp_HHH (1) .108 2.384 .000* 10.166 .120 .428 .261 2.246 Farm_exp_HHH (2) .884 .917 .238 2.100 .007* .209 .496 1.796 Farm_exp_HHH (3) .533 .636 .078 3.943 .018* .190 880 .864 Income_HHs .000 .013 .097 .697 Income_HHs (1) .000* 8.481 .012* 3.408 .639 1.378 .558 1.490 Income_HHs (2) .000* 17.510 .029* 3.632 .051 4.163 .241 2.701 Income_HHs (3) .000* 18.539 .002* 9.064 .281 2.275 .444 1.998 Family_size_HHs (1) .820 .902 N/C N/C .416 1.490 .002* .101 Weather_info_HHH (1) .364 .668 .182 .526 .102 .533 .774 .848 Farmer to farmer extension (1) .809 1.118 .424 1.455 .658 .806 .093 3.286 Government experts extension (1) .003* .271 .635 .815 .124 .520 .064 .290 Agro-ecol._HHs .005 .000 .000 .022 Agro-ecol._HHs (1) .036* 4.082 .055 5.446 .002* 5.412 .200 2.922 Agro-ecol._HHs (2) .093 .496 .002* .218 .000* 145.815 .006* 11.247 Farmlandsize (1) .001* 4.286 .865 .931 .000* .150 .02* 4.570 Crop failure N/C N/C .205 1.656 .044* .345 .211 0.146 Constant .121 .245 .305 .347 .002 23.795 .894 1.196 N/C not computed, HHH household head, HHs households *Significant at 0.05 Family size experience of (HHH2) only 0.209 times and (HHH3) only Family size has a significant (p = 0.002) effect on adopt - 0.190 times (much less) likely to adopt changing planting ing replanting failed crops. Family size has an inverse date. (Beta = − 2.297) in the log of odds by 2.297. The Exp (B) of 0.101 indicates households having family size > 4, only 0.101 times (much less) likely to replant failed crops. The Income inverse of Exp (B) of 0.101 indicates small family sizes Income has a positive and significant (p = 0.000) effect (< 4) are 9.91 times (much more) likely to replant failed on adopting crop diversification. Exp (B) of income crops than large family sizes. (10,001–30,000) 8.481 times, income (30,001–50,000) 17.510 times, and income (> 50,000) 18.539 times Access to government experts’ extension services more likely to diversify crops than low-income groups Formal extension services from government experts have (< 10,000). Income of households has a positive (Beta, a significant (p = 0.003) effect on adopting crop diversifi - + 1.226, + 1.290, and + 2.204) and significant (p = 0.012, cation to combat climate change impacts. The beta coef - 0.029, and 0.002) effect on adopting improved seeds. ficient shows an inverse relation (− 1.306) in adopting This indicates there is an increase in the log of odds by adaptation strategies. This indicates households who did 1.226, 1.290, and 2.204 in adopting improved seeds as an not get extension service; there is a decrease in the log of adaptation strategy. The EXP (B) of income (HHs1) 3.408 odd in diversifying crops by 1.306. The Exp (B) of 0.271 times, income (HHs2) 3.632 times, and income (HHs3) indicates that households who did not get formal exten- 9.064 times is more likely to adopt improved seeds than sion services were only 0.271 times (i.e. much less) likely low-income households (< 10,000). Amare et al. Agric & Food Secur (2018) 7:37 Page 8 of 11 to diversify crops than households who have got exten- having farmland size > 1.2 hectares are 4.286 times more sion service during 12 months of the year. likely to diversify crops than households with < 1.2 hec- tare farmland. Farmland size (> 1.2 hectares) of house- Agro-ecology settings holds has an inverse (beta, − 1.898) and significant Significant variation in the adoption of crop diversifica - (p = 0.000) effect on adopting changing planting date. tion was observed across agro-ecological zones (mid- This indicates a decrease (inverse relationships) in the land, p = 0.036). For example, higher crop diversification log of odds by 1.898 in changing planting date. The Exp was identified in the midland than highland and lowland (B) of 0.150 indicates households having large farmland agro-ecological settings. The Exp (B) of 4.082 indicates size (> 1.2 hectares) are only 0.150 times changed their households who live in the midland 4.082 times more planting date. Households having small land size( < 1.2 likely to use crop diversification than households reside hectare) are 6.76 times more likely to use changing plant- in highland. The Exp (B) of 0.496 indicates the lowland ing date. Farmland size has a positive relationship (Beta, households are only 0.496 times much less likely diversi- + 1.519) with no significant effect on adopting replant - fying crops than highlands. This means highland resident ing failed crops. There is an increase in the log of odds households are 2.016 times more likely to diversify crops by 1.519. The Exp (B) of 4.570 indicates households hav - than lowland households (i.e. invert, 1/0.496 = 2.016). ing > 1.2-hectare land are 4.2570 times more likely to The lowland agro-ecology with an inverse beta value replant failed crops than households with < 1.2 hectares. (− 1.552) has a significant effect (p = 0.002) on adopt- ing improved seeds. This indicates a decrease (1.522) in Crop failure history of households the log of odds on adopting improved seeds in the low- Crop failure has a significant (p = 0.044) effect on adopt - land agro-ecology zones. The EXP (B) of 0.218 indicates ing changing planting date. Households who never faced the lowland households are only 0.218 times (much less) crop failure in the past 10  years have an inverse rela- likely to use improved seeds than highland households. tionship (Beta = − 1.064) in employing changing plant- This indicates that highland households are 4.587 times ing date. The Exp (B) of 0.345 indicates households who (much more) likely to use improved seeds than lowland never faced crop failure in the past 10 years are only 0.345 resident households (inverse of 1/0.218 = 4.587). times adopted changing planting date. The invert of Exp The midland and lowland agro-ecologies have a sig - (B) of 0.345 is 2.8986, which indicates households who nificant effect on changing cropping date (p = 0.002 and faced crop failure are 2.8986 times more likely to change p = 0.000), respectively. The coefficient of beta (+ 1.689 planting date. This implies most farmers learn only when midland (HHs1) and + 4.982 lowland (HHs2)) indicates they faced problems. there is an increase in the log of odds by 1.689 and 4.982 on adopting changing planting date. The Exp (B) of 5.412 Discussion indicates that households who reside in the midland The rural communities of Dejen District adopt crop man - agro-ecology are 5.412 times much more likely to change agement strategies to combat climate change impacts. their cropping date than highland households. However, the key barriers identified in the study district The mid- and lowland agro-ecologies have a signifi - were shortage of money, lack of access to information, cant (p = 0.006; 0.020) effect on adopting replanting and small land size. Previous studies (e.g. [22–24]) stated failed crops. There is an increase (Beta =+ 1.072 midland that financial barriers are one of the barriers that restrict (HHs2) and + 2.420 lowland (HHs2)) in the log of odds implementation of adaptation strategies. This implies by 1.072 and 2.420. The Exp (B) of 2.922 indicates house - every form of adaptation requires some direct or indirect holds who live in the midland agro-ecologies are 2.922 costs. For instance, the use of improved varieties of crops times more likely to use replanting than highland house- has been reported as one of the key adaptation strate- holds. The Exp (B) of 11.247 indicates the lowland house - gies for farmers in Dejen District, Nile Basin of Ethiopia, holds are 11.247 times (much more) likely to replant their where this study confirmed. In the context of this study, failed crops. improved seeds include high yielding varieties, drought tolerant, short maturing, pest- and disease-resistant spe- Farmland size cies either induced or indigenous. When improved seed Farmland is the most significant (p = 0.001) factor to varieties are available, their price may be prohibiting diversify crops in the study communities. The beta coef - making it difficult for many rural households to access. ficient of households having farm size of > 1.2 hectare u Th s, framers have often sought to use their own saved has a positive relation to diversify crops. This implies seeds. One of the possible causes of financial barriers in there is an increase in the log of odds in diversifying the study area could be due to lack of credit facilities to crops by 1.455. The Exp (B) of 4.286 means households rural communities. Amare et al. Agric & Food Secur (2018) 7:37 Page 9 of 11 Access to information on weather and climate change Family size is negatively and significantly associated is an important tool that can be used to enhance the with the households’ decision to pursue replanting failed adaptation and implementation of adaptation strategies crops. Households who have large family size are sup- by rural communities of the study area. Access to infor- posed to have an opportunity in pursuing various adap- mation is particularly important for Africa [25] and Ethi- tation options to combat impacts of clime change and opia in particular, where there are few climate projections variability. This argument is raised by previous studies due to lack of appropriate climate data. This is crucially [26, 27] who argued that large family size is associated important, considering that most farming systems in with higher labour endowment which would enable a Dejen District depend on rain-fed agricultural systems. household to accomplish various agricultural tasks. The Hence, lack of appropriate climate information could be possible reason for an inverse relationship might be due crucial for rural communities’ food security. to the fact that community’s expectation of the benefits Age of household heads significantly determined crop of using adaptation strategy. In this regard, Barungi and diversification, improved seeds and changing planting Maonga [28] based on the rational choice theory; argue date, and replanting failed crops. Crop diversification that the behaviour of human beings is motivated by the and replanting of the failed crops require more energy possibility of gaining benefit. The possible explanation and experience. Thus, adult household heads are more could be households who have large family size have the matured and active in sowing different crops than old possibility to engage in off-farm activities, and they will and young household heads. The possible reason for the ignore the failed crops to replant. Therefore, rural com - positive and significant association is due to the fact that munities are rationale consumers of new technologies, age is the proxy indicator that may likely to endow the and they will only adopt technology as they foresee it will farmers with the requisite experience that enables them result in increased productivity. to make a better decision in the choice of crop manage- Access to government extension services has a nega- ment strategies. This is in line with studies by Deressa tive and significant association with the likelihood of et  al. [26] who found that an increase in age of house- choosing crop diversification to combat climate change hold head does mean an increase in farming experience impacts. This result is in contrary with previous stud - which would increase rural communities’ local knowl- ies [29, 30] who noted that farmers who obtain agricul- edge to respond to hazards resulted in climate change tural extension services through extension workers are and variability. more likely informed about the climatic situation and Farming experience is one of the significant variables the responses followed. The contributing factors for that affect the rural communities’ choice of adaptation this inverse relationship could be barriers to adopting strategies. Farming experience is a proxy indicator of age. crop diversification such as inadequate extension ser - Like crop diversification, the middle age household heads vices, constraints of money, labour, skills, and farmland have ability and willingness to adopt improved seeds locations. to adjust climate change impacts. This implies as one Households who live in the midland and highland agro- become more experienced in farming, the probability of ecologies have a significant and positive effect on adop - one to use improved seeds increases more than a farmer tion of crop diversification. This is because the suitability with less farming experience. of highland agro-ecology to sow different types of crops On the other hand, farming experience has an inverse and access to government extension services due to prox- relationship with changing planting date. The reason for imity to the administration. For instance, in this study an inverse relationship might be that experienced farm- finding, the midland agro-ecology has got more access to ers will have access to irrigation and water harvesting for extension services (77%) than the lowland agro-ecology their agricultural activities and plant their seeds without (47%) communities by the government extension experts changing the planting date. This implies a farmer with in the past cropping season. more experience would know when climate variability The lowland agro-ecology resident households have a is occurring in the area and which method of adapta- negative and significant effect on adoption of improved tion strategies works well in that specific agro-ecology seed varieties. The possible explanation is that lowland zone. As expected, income is positively and significantly households did not use improved seeds because of suit- associated with the household decision to pursue crop ability problem of the lowland agro-ecology and topog- diversification and improved seeds. This means crop raphy to use improved seeds to their farmland. This was diversification and purchasing of improved varieties of confirmed by households report on the barriers to adopt seeds require money. This implies the rate of using crop improved seeds as crop management strategy. On the diversification and improved seeds is increased as income other hand, the lowland agro-ecology has a positive and of households increased. significant effect on pursuing changing planting dates Amare et al. Agric & Food Secur (2018) 7:37 Page 10 of 11 to combat climate change impacts. This is because low - the government bodies in the office of agriculture did land agro-ecologies are characterized by erratic rainfall not realize the problems. This implies, in the process and other extreme events that lead households to change of diffusion of adaptation strategies, climate change their planting date. The midland and lowland agro-ecol - adaptation process should require close collaboration ogies have a significant effect on employing replanting and active participation of climate change research- failed crops as crop management strategy to combat ers, decision-makers, policy analysts, the community, climate change impacts. This is due to the fact that the and partners. Government policies should strengthen midland and lowland households are characterized by cli- the current adaptation strategies practised by rural mate variability such as erratic rainfall than the highland community households and support the adoption of agro-ecology households. The exposure of climatic vari - crop management strategies. Besides, the less adopted ability gave them more experience in adopting replanting crop management strategies due to financial con- their failed crops than highland households. straints should be subsidized by government and aid As expected, farm size has a significant and positive organizations. This study contributes to the academic effect on adopting crop diversification to combat climate discourse on climate change impact adaptations by change impacts. Households with larger farm sizes were providing empirical evidence to deepen understanding more likely to diversify their crops. On the other hand, of the barriers and determinants that confronts rural larger farmland size has a negative and significant effect communities in their attempt to implement adaptation on using changing planting dates as crop management strategies to manage the negative impacts of climate strategies. This means households having small land size change and variability. (< 1.2 hectares) are more likely use changing planting dates. This reminds us “a hunter who has only one arrow Abbreviations does not shoot with careless aim”. This implies house - df: degree of freedom; HHs: households; HHHs: household heads; NC: not holds who have small land size took care of their farm- computed; m.a.s.l: metres above sea level; VIF: variance inflation factor. land and changed their planting dates when there is a Authors’ contributions change in weather conditions. Even if farmland size has ZYA designed the data collection tools, conducted fieldwork and analysis, no significant effect on changing failed crops, it shows a and developed the manuscript. JOA contributed in commenting the data collection tools, recommending data analysis methods, reviewed and made positive effect on using changing failed crops. The pos - editorial comments on the draft manuscript. IOA contributed in commenting sible explanation could be the more farmland plot they the data collection tools, data analysis methods, reviewed and made editorial have, the more is the probability of having failed crop- comments on the draft manuscript. MTZ contributed to developing the data collection tools, commenting data analysis methods, reviewed, and made lands that could lead them to replant their failed crops. editorial comments on the draft manuscript. All authors read and approved the final manuscript. Conclusion and policy recommendations Author details The study communities have tremendous ideas to adapt Institute of Disaster Risk Management and Food Security Studies, Bahir Dar University, BahirDar, Ethiopia. Department of Geography, Faculty of the Social for current and future climate change impacts with a Sciences, University of Ibadan, Ibadan, Nigeria. Department of Geography strong motivation to move out of poverty. However, and Environmental Studies, Debre Tabor University, Debre Tabor, Ethiopia. the mere willingness to adopt climate change adapta- Acknowledgements tion strategies was not enough. Their ability to adopt is This study was made possible by the financial support of Pan-African Univer - constrained by many internal and external factors. Rural sity (PAU), a continental initiative of the African Union Commission (AU), Addis communities who did not employ adaptation strate- Ababa, Ethiopia. Further, authors would like to thank the data collectors and field assistants for their effective coordination, support, and time spent in gies gave many reasons for their failure to adopt. These organizing and conducting successful household interviews. Special thanks to includes poor or no access to water sources, limited the Dejen District rural communities who willingly volunteered the informa- knowledge, and skill, shortage of labour, lack of and/or tion in this study. shortage of farmland, lack of money, lack of information, Competing interests lack of agricultural extension services, and other institu- The authors declare they have no competing interests. tional factors. Availability of data and materials The most significant determinant factors of the The datasets used and/or analysed during the current study available from the choice of crop management strategies were age, farm- corresponding author on reasonable request. ing experience, income, agro-ecology setting, and Consent for publications farmland size. Agro-ecology setting has a significant The authors obtained permission from all participants in Dejen District, to effect on all adaptation strategies. Due to the soil publish their data. characteristics, the lowland agro-ecology zones were not suitable for adopting improved seeds. However, Amare et al. Agric & Food Secur (2018) 7:37 Page 11 of 11 Ethics approval and consent to participate 12. DDEPO. Dejen District Environmental Protection Office (DDEPO), East Consent to participate was received from everyone interviewed in Dejen Gojjam zone, Dejen, Ethiopia; 2016. District, Ethiopia. The Pan-African University, Life and Earth Sciences Institute 13. Yamane T. Statistics: an introductory analysis. New York: Harper and Row; (PAULESI) (including Health and Agriculture), University of Ibadan were 1967. informed of the study. 14. DDFEDO. Dejen District Finance and Economic Development Office (DDFEDO) Population Projection (DDFED, 2014); 2014. Funding 15. Bryan E, et al. Adapting agriculture to climate change in Kenya: house- This study was sponsored by the Pan-African University (PAU), a continental hold strategies and determinants. J Environ Manag. 2013;114:26–35. initiative of the African Union Commission (AU), Addis Ababa, Ethiopia. 16. Abid M, et al. Farmers’ perceptions of and adaptation strategies to climate change and their determinants: the case of Punjab province, Pakistan. Received: 11 November 2017 Accepted: 22 May 2018 Earth Syst Dyn. 2015;6(1):225. 17. Agresti A. An introduction to categorical data analysis. 2nd ed. Hoboken: Wiley; 2007. 18. NBE National Bank of Ethiopia. Commercial Banks’ Exchange rate; 2016. 19. Kothari CR. Research methodology: methods and techniques. Chennai: New Age International; 2004. References 20. SPSS. Statistical Package for the Social Sciences (SPSS). Version20. 1. WMO. Statement on the status of the global environment, vol. 1152. 21. Kothari GG. Research methodology. 3rd ed. New Delhi: New Age Interna- Geneva: Japan Meteorological Agency, in Cooperation with the World tional Publishers; 2014. Meteorological Organization; 2015. p. 250. 22. Bryan E, et al. Adaptation to climate change in Ethiopia and South Africa: 2. FAO. Climate change and food safety: a review. Food Res Int. options and constraints. Environ Sci Policy. 2009;12(4):413–26. 2010;43(7):1745–65. 23. Kithiia J. Climate change risk responses in East African cities: need, barri- 3. Foresight U. The future of food and farming. Final project report. London: ers and opportunities. Curr Opin Environ Sustain. 2011;3(3):176–80. The Government Office for Science; 2011. 24. Peterson C. Fast-growing groundnuts keep Ghana’s farmers afloat amid 4. Füssel H-M, Klein RJ. Climate change vulnerability assessments: an evolu- climate shifts. Retrieved July, 2013, vol. 16, p. 2013. tion of conceptual thinking. Clim Change. 2006;75(3):301–29. 25. IPCC. Climate change: climate change impacts, adaptation and vulner- 5. Rahman MI-U. Climate change: a theoretical review. Interdiscip Descr ability. The fourth assessment report of the Intergovernmental Panel on Complex Syst. 2013;11(1):1–13. Climate Change. Geneva, Switzerland; 2007. 6. Gbetibouo GA. Understanding farmers’ perceptions and adaptations to 26. Deressa TT, et al. Determinants of farmers’ choice of adaptation methods climate change and variability: the case of the Limpopo Basin, South to climate change in the Nile Basin of Ethiopia. Glob Environ Change. Africa, vol. 849. Washington: The International Food Policy Research 2009;19(2):248–55. Institute; 2009. 27. Menberu TZ, Aberra Y. Determinants of the adoption of land manage- 7. Segele ZT, Lamb PJ. Characterization and variability of Kiremt rainy season ment strategies against climate change in Northwest Ethiopia. Ethiop over Ethiopia. Meteorol Atmos Phys. 2005;89(1):153–80. Renaiss J Soc Sci Humanit. 2014;1:93–118. 8. NMA. Climate change national adaptation programme of action (Napa) 28. Barungi M, Maonga BB. Adoption of soil management technologies by of Ethiopia. National Meteorological Services Agency (NMA), Ministry of smallholder farmers in central and southern Malawi. J Sustain Dev Afr. Water Resources, Federal Democratic Republic of Ethiopia, Addis Ababa; 2011;13(3):28–38. 29. Maddison D. The perception of and adaptation to climate change in 9. You GJ-Y, Ringler C. Hydro-economic modeling of climate change Africa, vol. 4308. Washington: World Bank Publications; 2007. impacts in Ethiopia. International Food Policy Research Institute (IFPRI); 30. Nhemachena C, Hassan R. Micro-level analysis of farmers adaption to climate change in Southern Africa. Washington: The International Food 10. Schröter D, Polsky C, Patt AG. Assessing vulnerabilities to the effects of Policy Research Institute; 2007. global change: an eight step approach. Mitig Adapt Strat Glob Change. 2005;10(4):573–95. 11. DDARDO. Dejen District Agricultural and Rural Development Office (DDRDO), Annual Report, East Gojjam zone, Dejen, Ethiopia; 2016. Ready to submit your research ? Choose BMC and benefit from: fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Agriculture & Food SecuritySpringer Journals

Published: Jun 5, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off