Togolese agriculture is predominantly rain-fed and hence fundamentally dependent on the vagaries of weather. Thus, it is negatively affected by climate change. The present study assesses farmers’ perceptions and adaptation to climate change to enhance policy towards tackling the challenges climate change poses to the farmers in the study area. Descriptive statistics and multinomial logit (MNL) were used to analyze data obtained from a cross-sectional survey executed during the 2013/2014 agricultural production year in the maritime, plateau and savannah regions of Togo. The analysis of farmers’ perception to climate change reveals high increase in temperature and decrease in rainfall. These results are in line with the trend analysis of climate data that records from 1961 to 2013 about the study area especially on the temperature. Furthermore, the results show that crop diversification, change in crops, find off-farm jobs, change of the amount of land, change of the planting date and plant short season variety are the adaptation methods employed by the farmers. Moreover, with respect to the multinomial logit analysis, the results highlight that education level, farming experience, access extension services, access to credit and access to climate information are the factors that enhance farmers’ adaptive capacity to climate change and variability. Thus, there is room for better adaptation if government intensifies activities of extension workers and ensures that farmers have access to affordable credit schemes to increase their ability and flexibility to adopt adaptation measures. There is also a need to include climate change communication to facilitate exchange of climatic information that could enable smallholder farmers to adapt to changing planting dates. Finally, investment in education systems and creation of off-farm job opportunities in the rural areas can be underlined as a good policy option.
The Intergovernmental Panel on Climate Change [
Agriculture in Togo stands to be a major economic sector as it contributes about 38% of the nation’s GDP. More than 70% of the population of the country depends on agricultural sector for their livelihood. In addition, agriculture supplies more than 20% of the exportation revenues of the country. Despite its high contribution to the overall economy, agriculture in Togo is predominantly rain-fed and hence fundamentally dependent on the vagaries of weather [
Adaptation is widely recognized as a vital component of any policy response to climate change. It is a way of reducing vulnerability, increasing resilience, moderating the risk of climate impacts on lives and livelihoods, and taking advantage of opportunities posed by actual or expected climate change [
In addition, [
Despite the importance of perceptions and adaptations to climate change, in the context of Togo, a very few studies have examined rural smallholder farmers’ perceptions and adaptations to climate change. This study therefore analyses how farmers perceive and adapt to climate change. Especially, this paper intends to capture the extent of farmers’ awareness and perceptions of climate variability and change and the types of adjustments they have made in their farming practices in response to these changes.
The study was conducted in the maritime, plateau and savannah regions of Togo (Appendix 1). The maritime and plateau regions are located at the southern part of Togo while the savannah region is at the extreme northern part. The maritime region covers an area of about 6329 km2 of land and has 373 people per km2 as population density, whereas, the plateau region covers 17,323 km2 and has 75 people per km2. Last but not least, the savannah region covers 8688 km2 of land and has 99 people per km2 as population density. Furthermore, according to [
The current study is based on a cross-sectional household survey data of mixed crops and livestock farmers collected during the month of August 2014 in the Maritime, Plateaux and Savannah regions of Togo. The sample regions were purposely selected for this study based on a study by [
Besides collecting data on different socioeconomic and environmental attributes, the survey also included information on farmers’ perceptions of climate change and adaptation methods. The surveyed farmers were asked questions about their observation in the temperature and rainfall patterns over the past 20 years. Also, monthly rainfall and temperature data were obtained from the Togolese main Meteorological Service in Lomé. The data cover the period from January 1961 to December 2013 for all the meteorological services located within each of three regions selected for this study.
Descriptive statistics and logistic regression analysis were the main analytical techniques used in this study. Data were analyzed using the Stata 13.1 software. Correlation analysis was used to analyse the association between different variables. The hypothesized explanatory variables were checked for the existence of multi-col- linearity problem. When the absolute value of Pearson correlation coefficient between two variables is greater than 0.8, there is multi-collinearity problem. So, one of these two variables will be drop from the model.
Farmers’ Perception of Climate Change
The logit model was employed due to the nature of the decision variable; whether farmers perceived change in the temperature and/or the rainfall or not. The logit model considers the relationship between a binary dependent variable and a set of independent variables, whether binary or continuous. It is given by [
where Pi is the probability of perceiving a change in the climate and Xi an independent variables. Therefore, the parameter βi gives the log odds of the dependent variable and β0 is a constant.
The probability of occurrence of an event relative to non-occurrence is called odds ratio and is given by [
Farmers’ Adaptation to Climate Change
Given that we investigate several adaptation choices, the appropriate econometric model would, thus, be either a multinomial logit (MNL) or multinomial probit (MNP) regression model. Both models estimate the effect of explanatory variables on a dependent variable involving multiple choices with unordered response categories. In this study, therefore, an MNL specification is adopted to model climate change adaptation behaviour of farmers involving discrete dependent variables with multiple choices. The advantage of the MNL is that it permits the analysis of decisions across more than two categories, allowing the determination of choice probabilities for different categories [
The multinomial log it model is useful in investigating consumer choice behaviour and has become increasingly popular in marketing research. Let C be a set of n choices, denoted by {1; 2; ∙∙∙; n}. A subject is present with alternatives in C and is asked to choose the most preferred alternative. Let xi be a covariate vector associated with the alternative i. The multinomial logit model for the choice probabilities is given by
where β is a vector of unknown regression parameters.
Unbiased and consistent parameters estimates of the MNL model in Equation (3) require the assumption of independence of irrelevant alternatives (IIA) to hold. The property of the logit model whereby Pj/Pk is independent of the remaining probabilities is called the independence from irrelevant alternatives (IIA) [
Across the three regions, about 85% of the farmers interviewed perceive changes in temperature. In the Maritime region, this percentage is 97, while in the Plateaux region it is 80 and 76 in Savannah region (Appendix 2). About 72% of the farmers perceive increases in temperature, while only 12.85% notice the contrary, a decrease in temperature. However, 9.72% of the farmers do not perceive any change in temperature (
In total, 85.58% of the respondents observed changes in rainfall patterns over the past 20 years. The distribution of the farmers’ perceptions regarding changes in rainfall patterns revealed that 74.61% perceived a decrease in rainfall. In the Maritime region, 95% of farmers perceived decrease in rainfall, while in the Plateaux region it is 62% and 63% in the Savannah region (Appendix 3). Despite higher perception of the farmers interviewed on changes in rainfall patterns, 6.58% of the farmers interviewed did not see any change in rainfall patterns (
Yearly Temperature | Maritime Region | Plateaux Region | Savannah Region |
---|---|---|---|
Mean (˚C) | 27.54 | 25.45 | 28.27 |
Standard deviation (˚C) | 0.574 | 0.405 | 0.560 |
Minimum temperature (˚C) | 26.4 | 24.5 | 27.1 |
Maximum temperature (˚C) | 28.8 | 26.2 | 29.5 |
Trend (˚C/year) | 0.0334*** | 0.0125*** | 0.0286*** |
Correlation | 0.8813 | 0.4882 | 0.7907 |
Total change calculated from the trend (˚C/53 years) | 1.737 | 0.650 | 1.487 |
***p < 0.01 Student’s t-test, N = 53. Total change is the difference between the trend line value of the first and last year.
statistically significant. The correlation between rainfall and time is also insignificant. Indeed, there is a large variability in the amount of precipitation from year to year. The same pattern is observed in each district (
Among the variables, the age of the farmer was found to be correlated inversely with education (ρ = −0.035), while it was highly positive and significant at p < 0.01 level of significance with farming experience (ρ = 0.825). By the same token, there has been a strong positive association between gender and land tenure at p < 0.01.
Most importantly, the analysis showed that there is multi-collinearity problem between age and farming experience. Thus, the variable age was dropped from the model because most of farmers are old and variable farming experience is more relevant for the study than the latter.
The independent variables are gender, education, farming experience, farm size, land tenure, soil fertility, access to extension services, access to climate information, access to credit, farmers’ group membership, and region dummy for Plateaux and Savannah with Maritime being the reference region for comparison.
The results displayed in
Farming experience seems to decrease the probability that the farmer will perceive long-term changes in rainfall and temperature. Thus, educated farmers are more likely to see that rainfall does not have a significant trend and less likely to perceive that temperature does not have a significant trend over the long run.
Yearly total rainfall | Maritime region | Plateaux region | Savannah region |
---|---|---|---|
Mean (mm) | 942.7 | 1514.2 | 1054.4 |
Standard deviation (mm) | 193.06 | 263.86 | 120.99 |
Minimum rainfall (mm) | 557.1 | 982.6 | 808.6 |
Maximum rainfall (mm) | 1528.2 | 2150.7 | 1323.4 |
Trend (mm/year) | −1.142 | −2.625 | 0.181 |
Correlation | −0.0913 | −0.1537 | 0.0231 |
Total change calculated from the trend (mm/53 years) | −59.38 | −136.52 | 9.42 |
Total change calculated from the trend (%) | −6.11 | −8.63 | 0.89 |
Gender | Age | Education | Farming experience | Farm | Land tenure | Soil fertility | Extension | Credit | Farmers’ group | Climate information | |
---|---|---|---|---|---|---|---|---|---|---|---|
Gender | 1.0000 | ||||||||||
Age | −0.0959 | 1.0000 | |||||||||
Education | 0.1767* | −0.0351 | 1.0000 | ||||||||
Farming experience | −0.1311* | 0.8253* | −0.0466 | 1.0000 | |||||||
Farm size | −0.0186 | 0.1274* | 0.0912 | 0.1372* | 1.0000 | ||||||
Land tenure | 0.3535* | 0.0445 | −0.0639 | −0.0420 | −0.1305* | 1.0000 | |||||
Soil fertility | 0.1150* | 0.0485 | −0.0470 | 0.0343 | −0.0210 | 0.2594* | 1.0000 | ||||
Extension | −0.0292 | 0.1840* | 0.0252 | 0.2648* | 0.2433* | −0.0798 | −0.0515 | 1.0000 | |||
Credit | −0.0348 | 0.1524* | 0.1183* | 0.1294* | 0.1294* | −0.0003 | −0.0342 | 0.3576* | 1.0000 | ||
Farmers’ Group | 0.2197* | −0.0046 | 0.0047 | −0.0957 | −0.1052 | 0.2409* | 0.1068 | 0.0496 | 0.1057 | 1.0000 | |
Climate information | 0.0839 | 0.0860 | 0.0734 | 0.1098 | 0.2011* | 0.0008 | 0.0763 | 0.3085* | 0.1534* | −0.0202 | 1.0000 |
*p < 0.01. All correlations are Pearson’s r.
COEFFICIENTS(in log-odds unit ) | ||
---|---|---|
VARIABLES | Perceive change in temperature | Perceive change in rainfall |
Gender | 0.80* (1.73) | 0.41 (0.95) |
Education level | −0.06 (−1.04) | −0.02 (−0.40) |
Farming experience | −0.13** (−2.29) | −0.19*** (−3.41) |
Farm size | 0.32 (0.91) | 0.17 (0.59) |
Land tenure | 1.22*** (3.00) | 0.17 (0.45) |
Soil fertility | 0.47 (0.75) | 0.82 (1.52) |
Access to extension | 0.60 (1.19) | 0.33 (0.74) |
Access to credit | 0.07 (0.11) | −0.45 (−0.79) |
Farmers’ group membership | 0.33 (0.76) | 0.50 (1.15) |
Access to climate information | −0.58 (−1.44) | −0.54 (−1.35) |
Plateaux region | −2.52** (−2.54) | −3.14*** (−3.48) |
Savannah region | −3.04*** (−3.30) | −3.40*** (−3.89) |
Constant | 0.12 (0.09) | 1.22 (0.83) |
Observations | 316 | 316 |
***p < 0.01, **p < 0.05, *p < 0.1; Robust z-statistics in parentheses.
Total change is the difference between the trend line value of the first and last year
Male farmers are more likely to perceive change in temperature than female farmers;
Farm land ownership, on the other hand, increases the probability of perceiving change in temperature;
The results also confirm that being in the Plateaux Region or the Savannah Region decreases the probability of perceiving climate change (in temperature and rainfall) than being in the Maritime region;
Also, farm size, access to credit, access to extension services, being member of farmers’ association, and soil fertility influence positively farmers’ perception of changes in the climate of the study area.
The adaptation methods employed by farmers in the study area are indicated in
Indeed, seven adaptation measures could be identified in the study area as farmers’ responses to increased temperature, reduced rainfall and disrupted rainfall patterns. Planting short season variety (20.38%) and changing crop planting dates (17.87%) were identified as the major adaptation strategies to climate change in the study area, while only a few (9.72%) opted for crop diversification. As indicated, planting short season variety is most commonly used method, whereas changing type of crops is the least practiced among the major adaptation methods identified in the study area. Greater use of planting short varieties as an adaptation method could be associated with the access to extension services (ICAT and NGOs) and the ongoing PNIASA project in agriculture sector in Togo that provided farmers with improved seeds.
In this section, the MNL model for adaptation choices to climate change in the study area was estimated. The MNL adaptation model was run and tested for the IIA assumption, using the Hausman specification test. As a result, the test failed to reject the null hypothesis of independence of odds of other alternative (Appendix 4), suggesting that there is no evidence against the correct specification for the adaptation model. Therefore, the application of the MNL specification to the data set for modelling climate change adaptation behavior of farmers is justified. The estimation of the multinomial logit model for this study was undertaken by normalizing one category, which is normally referred to as the “reference state,” or the “base category.” In this analysis, the first category (no adaptation) is the reference state. Thus,
Adaptation strategies | Increase in temperature and Decrease in rainfall (%) |
---|---|
Crop diversification | 9.72 |
Change in crops | 0.94 |
Find off-farm jobs | 3.76 |
Change the amount of land | 1.88 |
Change planting dates | 17.87 |
Plant short season variety | 20.38 |
Other | 3.76 |
No adaptation | 41.69 |
Total | 100 |
COEFFICIENTS (in log-odds unit ) | |||||||
---|---|---|---|---|---|---|---|
VARIABLES | Crop diversification | Change in crops | Find off-farm jobs | Changed the amount of land | Changed planting date | Plant short season variety | Others |
Gender | 0.12 | 0.21 | 0.37 | −0.97 | −0.54 | −0.72 | 0.05 |
(0.21) | (0.14) | (0.50) | (−0.93) | (−1.09) | (−1.51) | (0.06) | |
Education level | −0.04 | 0.00 | −0.11 | −0.18 | −0.03 | 0.13** | −0.17 |
(−0.51) | (0.02) | (−0.85) | (−0.75) | (−0.45) | (2.02) | (−1.07) | |
Farming experience | 0.09* | 0.13 | 0.03 | −0.11 | 0.11*** | 0.09** | 0.03 |
(1.94) | (0.78) | (0.54) | (−0.62) | (2.69) | (2.20) | (0.42) | |
Farm size | 0.32 | −0.68 | −0.63 | 0.41 | 0.37 | 0.37 | −0.50 |
(1.26) | (−0.39) | (−0.85) | (0.86) | (1.53) | (1.57) | (−0.59) | |
Land tenure | −1.14** | −1.45 | −2.17** | 0.99 | −1.31*** | −0.45 | −0.79 |
(−2.07) | (−0.77) | (−2.55) | (0.95) | (−2.70) | (−0.97) | (−0.97) | |
Soil fertility | −2.47** | −15.04 | 0.79 | −16.98 | −1.54** | −0.76 | 0.77 |
(−2.22) | (−0.01) | (1.07) | (−0.00) | (−2.36) | (−1.38) | (1.13) | |
Access to extension | 1.00* | 1.84 | −0.40 | 0.81 | 0.82* | 1.94*** | −0.25 |
(1.82) | (0.88) | (−0.45) | (0.79) | (1.69) | (4.14) | (−0.30) | |
Access to credit | 0.43 | 2.80 | 2.41** | −16.02 | 0.95 | 1.63*** | 1.68* |
(0.50) | (1.17) | (2.48) | (−0.00) | (1.32) | (2.61) | (1.66) | |
Farmers’ group membership | −2.32*** | −18.36 | −0.52 | −0.22 | −2.23*** | −1.01* | 16.22 |
(−4.09) | (−0.01) | (−0.56) | (−0.19) | (−4.27) | (−1.85) | (0.01) | |
Access to climate information | 0.84 | 1.76 | 0.43 | 0.55 | 2.65*** | 1.93*** | 0.15 |
(1.51) | (0.82) | (0.55) | (0.54) | (5.44) | (4.34) | (0.20) | |
Constant | −2.40** | −0.16 | −3.82** | 0.08 | −1.59* | −2.43** | −18.29 |
(−2.29) | (−0.05) | (−2.35) | (0.03) | (−1.65) | (−2.53) | (−0.01) | |
Observations | 316 | 316 | 316 | 316 | 316 | 316 | 316 |
***p < 0.01, **p < 0.05, *p < 0.1; Z-statistics in parentheses.
Education level of the farmers increases the probability of uptake of adaption options to climate change. As can be observed in
Farmer experience increases the probability of uptake of crop diversification, changing planting dates and planting short season variety as adaptation measures. Experienced farmers are more likely to adopt changing planting dates and planting short season variety and less likely to diversify crops in the study area. These results confirm the findings of [
Access to extension services: the results show that Access to extension services significantly increases the probability of taking up adaptation options in the study area. Indeed, farmers who have access to extension services are more likely to adopt planting short season variety and less likely to diversify crops and to change planting dates as adaptation options. Extension services provide an important source of information on climate change as well as agricultural production and management practices. Farmers who have significant extension contacts have better chances to be aware of changing climatic conditions and also of the various management practices that they can use to adapt to changes in climatic conditions. Similarly, [
Access to credit: As expected, the results show that having access to credit increases the propensity of farmers to adapt to climate change. Farmers who have access to credit are more likely to adopt planting short season variety and less likely to find off-farm jobs in the study area. The import is that poverty or lack of financial resources is one of the main constraints to adjustment to climate change and thus having access to credit counteracts these constraints. Also, with more financial and other resources at their disposal, farmers are able to change their management practices in response to changing climatic conditions. Similarly, the reports from [
Access to climate information: As expected, the results show that access to information on climate change (temperature and rainfall) has a significant and positive impact on farmers’ adoption of changing planting dates and planting short season varieties. These results are in line with findings by [
The study analyzed the factors affecting the farmers’ perceptions and choice of adaptation methods to climate change based on a cross-sectional survey data collected during the 2013/2014 agricultural production year in the maritime, plateau and savannah regions of Togo. The surveyed farmers were asked if they had observed any change in temperature and rainfall over the past 20 years. As a result, about 85% of the farmers perceived increase in temperature while in total, 85.58% of the respondents observed changes in rainfall patterns over the past 20 years. These results are in line with the climatic data that records in the study area because the statistical analysis of temperature data from 1961 to 2013 showed an increasing trend in the three regions and rainfall data showed decreasing trends for maritime and plateau regions while for savannah region, the trend was slightly increasing.
Regarding the determinants of farmers’ perceptions of climate change, male gender farmers are more likely to perceive change in temperature than female gender; owing a farm land, on the other hand, increases the probability of perceiving change in temperature; and farming in plateau region or savannah region decreases the probability of perceiving climate change (in temperature or rainfall) unlike farming in maritime region.
Although farmers appear to be well aware of climate change, few seem to actively undertake adaptation measures to counteract climate change. Indeed, almost 42% did not undertake any remedial actions. The adaptation options observed in the study area are manifold but the main adaptation strategies of farmers identified include planting short season variety and changing crop planting dates.
The study used the multinomial logit (MNL) model to assess the factors influencing farmers’ choices of climate change and variability adaptation methods. In the model, the dependent variables include different adaptation methods and the explanatory variables include household, farm and institutional characteristics and other factors. The results highlighted that education level, farming experience, access extension services, access to credit and access to climate information are the factors that enhance farmers’ adaptive capacity to climate change and variability.
This study demonstrates the importance of government policies and strategic investment plans which should support improved access to climate forecasting and dissemination, ensure that farmers have access to affordable credit schemes to increase their ability and flexibility to adopt adaptation measures in response to the forecasted climate conditions. Moreover, given that extension services are inadequate in the study area, improving the knowledge and skills of extension service personnel and making the extension services more accessible to farmers appear to be some of the key elements of a fruitful adaptation program. Finally, investment in education systems and creation of off-farm job opportunities in the rural areas can be underlined as a policy option regarding reduction of the adverse impacts of climate change in the study area.
I wish to express my profound gratitude to West African Science Service Centre on Climate Change and Adapted Land Use (WASCAL) for financing this research. I also thank the WASCAL management team at Université de Lomé for their technical support.
AgossouGadédjisso-Tossou, (2015) Understanding Farmers’ Perceptions of and Adaptations to Climate Change and Variability: The Case of the Maritime, Plateau and Savannah Regions of Togo. Agricultural Sciences,06,1441-1454. doi: 10.4236/as.2015.612140
Appendix 1. Study area.
Appendix 2. Farmers’ perceptions of changes in temperature per region.
Appendix 3. Farmers’ perceptions of changes in rainfall per region.
Appendix 4. Hausman Tests of IIA Assumption (MNL Model).