From 1 - 5 / 5
  • Categories  

    The first part of the dissertation deals with the dynamics of assets owned by the household head, his spouse, or jointly by both in response to diverse shocks in rural agricultural households in Bangladesh, one of the most vulnerable countries to climate change. Therefore, a unique and detailed country representing household survey panel data is used, known as ‘Bangladesh Climate Change Adaptation Survey’ of 2010 and 2012. Looking at changes within rather than between households, the research shows that land is owned mostly by men, who are also wealthier than their spouses. By constructing a comprehensive index the overall effect on wealth is investigated, which does not exist in the literature yet. The results suggest that husband’s and wife’s asset holdings respond differently depending on the type of shocks. Weather shocks such as cyclones adversely affect the asset holdings of household heads in general, while predicted external events such as seasonal droughts and dowry payments reduce assets of both spouses. The results suggest that jointly owned assets are not sold in response to shocks; either due to these assets being actively protected or due to the difficulty of agreeing on this coping strategy. Women’s asset holdings and associated choices of substituting assets are shaped by their lesser involvement in agriculture. To know the changes of behavioral patterns in response to these shocks, the results suggest that households are more likely to adopt short-term coping mechanisms in response to non-climatic negative shocks rather than to climatic shocks, whereas households are more likely engage in adaptation strategies in response to the latter. Furthermore, adaptation strategies are often combined complementary efforts, whereas coping mechanisms are mutually independent across the study. In particular, group participation in general is associated with crop adaptation strategies and perceptions of climate change among women. Finally, the research seeks to explore the potential of group based approaches which is receiving a growing attention due to their possible role in securing household welfare in the presence of adverse events. The inherent endogeneity is addressed by using instrumental variables. The results suggest that household heads mainly participate in groups that are welfare augmenting and income enhancing, while their spouses are mainly active in credit groups due to less personal wealth which are more strongly negatively affected by shocks. Furthermore, evidence is found for a positive association of social and political capital with household-level welfare and with asset holdings of the household head. Interestingly, it seems that this effect is not driven by mere participation in groups, but also by other aspects of social capital, for example informal networks, of both household heads and spouses.

  • The research within the scope of which the presented data was generated was part of the funding initiative ‘Knowledge for Tomorrow-Cooperative Research Project in sub-Saharan Africa on Resource, their Dynamics, and Sustainability’ funded by the Volkswagen Foundation. The overall study aimed at investigating the uncertainties in the effectiveness of biological control of stem borers under different climate change scenarios in Kenya and Tanzania. Using the species distribution modelling approach MaxEnt, the research predicts the current and future distribution of three important lepidopteran stem borer pests of maize in eastern Africa, i.e., Busseola fusca (Fuller, 1901), Chilo partellus (Swinhoe, 1885) and Sesamia calamistis (Hampson, 1910), and two of their parasitoids used for biological control, i.e., Cotesia flavipes (Cameron, 1891) and Cotesia sesamiae (Cameron, 1906). Based on these potential distributions and data collected during household surveys with local farmers in Kenya and Tanzania, future maize yield losses are predicted considering three different Global Circulation Models (GCMs) for four different Shared Socioeconomic Pathway (SSP) scenarios (SSP1-2.6, SSP2-4.5, SSP 3-7.0, SSP5-8.5) and two time periods, i.e., 2041-2060 and 2081-2100. Stem borer-associated maize yield losses (in kg/ha) are extrapolated using data on average maize yield in the study area downloaded from MapSPAM and based on results from a 2018 household survey conducted by researchers from the International Centre of Insect Physiology and Ecology (icipe) under the cooperative project "Adaptation for Food Security and Ecosystem Resilience in Africa” (AFERIA) of icipe, the University of Helsinki and the University of York in which local farmers were asked to quantify losses in maize yield by stem borer infestation. Based on these survey data, 95% confidence intervals (CIs) for maize yield losses were calculated. Yield losses by stem borers for all scenarios are given for mean, lower and upper bound of the CI.

  • Categories  

    The research within the scope of which the presented data was generated was part of the funding initiative ‘Knowledge for Tomorrow-Cooperative Research Project in sub-Saharan Africa on Resource, their Dynamics, and Sustainability’ funded by the Volkswagen Foundation. The overall study aimed at investigating the uncertainties in the effectiveness of biological control of stem borers under different climate change scenarios in Kenya and Tanzania. Using the species distribution modelling approach MaxEnt, the research predicts the current and future distribution of three important lepidopteran stem borer pests of maize in eastern Africa, i.e., Busseola fusca (Fuller, 1901), Chilo partellus (Swinhoe, 1885) and Sesamia calamistis (Hampson, 1910), and two of their parasitoids used for biological control, i.e., Cotesia flavipes (Cameron, 1891) and Cotesia sesamiae (Cameron, 1906). Based on these potential distributions and data collected during household surveys with local farmers in Kenya and Tanzania, future maize yield losses are predicted considering three different Global Circulation Models (GCMs) for four different Shared Socioeconomic Pathway (SSP) scenarios (SSP1-2.6, SSP2-4.5, SSP 3-7.0, SSP5-8.5) and two time periods, i.e., 2041-2060 and 2081-2100. A raster in which probability of habitat suitability is separately specified for each grid cell is the immediate output from species distribution modelling with MaxEnt. Probability of habitat suitability for the respective species hereby is expressed as probability value ranging between 0 (unsuitable habitat) to 1 (perfectly suitable habitat). Probability of habitat suitability was modelled for five species for current climatic conditions, as well as for four SSPs and two time periods.

  • Categories  

    The research within the scope of which the presented data was generated was part of the funding initiative ‘Knowledge for Tomorrow-Cooperative Research Project in sub-Saharan Africa on Resource, their Dynamics, and Sustainability’ funded by the Volkswagen Foundation. The study aimed at investigating the uncertainties in the effectiveness of biological control of stem borers under different climate change scenarios in Kenya and Tanzania. Using the species distribution modelling approach MaxEnt, the research predicts the current and future distribution of three important lepidopteran stem borer pests of maize in eastern Africa, i.e., Busseola fusca (Fuller, 1901), Chilo partellus (Swinhoe, 1885) and Sesamia calamistis (Hampson, 1910), and two of their parasitoids used for biological control, i.e., Cotesia flavipes (Cameron, 1891) and Cotesia sesamiae (Cameron, 1906). Based on these potential distributions and data collected during household surveys with local farmers in Kenya and Tanzania, future maize yield losses are predicted considering three different Global Circulation Models (GCMs) for four different Shared Socioeconomic Pathway (SSP) scenarios (SSP1-2.6, SSP2-4.5, SSP 3-7.0, SSP5-8.5) and two time periods, i.e., 2041-2060 and 2081-2100. The reduction potential of stem borer-associated maize yield losses by application of parasitoids (in kg/ha) is extrapolated using previously estimated average maize yield losses in the study area and results from a 2018 household survey conducted by researchers from the International Centre of Insect Physiology and Ecology (icipe) under the cooperative project "Adaptation for Food Security and Ecosystem Resilience in Africa” (AFERIA) of icipe, the University of Helsinki and the University of York in which local farmers were asked to quantify the reduction of losses in maize yield by stem borer infestation through application of natural enemies. Based on these survey data, 95% confidence intervals (CIs) for maize yield losses were calculated. Potential to reduce yield losses by stem borers for all scenarios are given for mean, lower and upper bound of the CI.

  • Categories  

    The research within the scope of which the presented data was generated was part of the funding initiative ‘Knowledge for Tomorrow-Cooperative Research Project in sub-Saharan Africa on Resource, their Dynamics, and Sustainability’ funded by the Volkswagen Foundation. The overall study aimed at investigating the uncertainties in the effectiveness of biological control of stem borers under different climate change scenarios in Kenya and Tanzania. Using the species distribution modelling approach MaxEnt, the research predicts the current and future distribution of three important lepidopteran stem borer pests of maize in eastern Africa, i.e., Busseola fusca (Fuller, 1901), Chilo partellus (Swinhoe, 1885) and Sesamia calamistis (Hampson, 1910), and two of their parasitoids used for biological control, i.e., Cotesia flavipes (Cameron, 1891) and Cotesia sesamiae (Cameron, 1906). Based on these potential distributions and data collected during household surveys with local farmers in Kenya and Tanzania, future maize yield losses are predicted considering three different Global Circulation Models (GCMs) for four different Shared Socioeconomic Pathway (SSP) scenarios (SSP1-2.6, SSP2-4.5, SSP 3-7.0, SSP5-8.5) and two time periods, i.e., 2041-2060 and 2081-2100. The rasters show the species predicted current and future distribution for four different SSPs and time periods 2041-2060 and 2081-2100. The distribution rasters are based on a raster displaying probability of habitat suitability which was converted into binary range maps by application of different threshold levels, i.e., 1) Balance training omission, predicted area and threshold values Cloglog threshold, 2) Maximum training sensitivity plus specificity Cloglog threshold, 3) Equal training sensitivity and specificity Cloglog threshold and 4) 10th percentile training presence Cloglog threshold. Grid cells carrying a probability value above the respective threshold show species presence (cell assigned a value of 1), while grid cells with a probability value below the threshold show species absence (cell assigned a value of 0). Accordingly, 4 presence-absence rasters were obtained for a species current distribution and calculated in their sum, while 12 presence-absence rasters were calculated for each climate change scenario, subsequently also calculated in their sum. The raster therefore specifies grid cells where species distribution is predicted to be more (grid cell carrying a high sum value), or less, likely (grid cell carrying a low cell value).