The application of remote sensing methods to assess crop vigor and yields has had limited applications in Sub-Saharan Africa (SSA) due largely to limitations associated with satellite images. The increasing use of unmanned aerial vehicles in recent times opens up new possibilities for remotely sensing crop status and yields even on complex smallholder farms. This study demonstrates the applicability of a vegetation index derived from UAV imagery to assess maize (Zea mays L.) crop vigor and yields at various stages of crop growth. The study employs a quadcopter flown at 100 m over farm plots and equipped with two consumer-grade cameras, one of which is modified to capture images in the near infrared. We find that UAV-derived GNDVI is a better indicator of crop vigor and a better estimator of yields—r = 0.372 and r = 0.393 for mean and maximum GNDVI respectively at about five weeks after planting compared to in-field methods like SPAD readings at the same stage (r = 0.259). Our study therefore demonstrates that GNDVI derived from UAV imagery is a reliable and timeous predictor of crop vigor and yields and that this is applicable even in complex smallholder farms in SSA.

Drones are increasingly becoming a ubiquitous feature of society. They are being used for a multiplicity of applications for military, leisure, economic, and academic purposes. Their application in academia, especially as social science research tools, has seen a sharp uptake in the last decade. This has been possible due, largely, to significant developments in computerization and miniaturization, which have culminated in safer, cheaper, lighter, and thus more accessible drones for social scientists. Despite their increasingly widespread use, there has not been an adequate reflection on their use in the spatial social sciences. There is need for a deeper reflection on their application in these fields of study. Should the drone even be considered a tool in the toolbox of the social scientist? In which fields is it most relevant? Should it be taught as a course in the social sciences much in the same way that spatially-oriented software packages have become mainstream in institutions of higher learning? What are the ethical implications of its application in spatial social science? This paper is a brief reflection on these questions. We contend that drones are a neutral tool which can be good and evil. They have actual and potentially wide applicability in academia but can be a tool through which breaches in ethics can be occasioned given their unique abilities to capture data from vantage perspectives. Researchers therefore need to be circumspect in how they deploy this powerful tool which is increasingly becoming mainstream in the social sciences.

The shortfalls in the quality, quantity, and reliability of agriculture performance data are neither new nor confined to Sub-Saharan Africa (SSA). It is, however, a more dire challenge given the overwhelming importance of agriculture in the economies of most countries in the region in terms of food security and poverty reduction. While farmers’ self-reported (SR) data on crop outputs and farm sizes remain popular variables for computing plot productivity and yields, especially in SSA, other methods such GPS measurement and remote sensing measurement of crop area, crop cuts (CC) as well as whole plot harvests have been touted as the gold standard methods for yield measurement. All these approaches to yield estimation are insufficient in capturing real agriculture productivity in rainfed farming systems due to the significant area loss that characterizes these farming systems in the course of each cropping season. This paper compares yield data of smallholder maize plots from two farming communities in the Eastern Region of Ghana based on farmer self-reported outputs and crop cuts, as well as GPS and aerial imagery measurement of plot area. The study finds a high level of agreement between GPS-measured plot area and that measured using remote sensing methods (R² = 0.80) with the minor deviations between the two measures attributable to changes in farmers’ plans in the course of the season with regards to their cultivation extent. More interestingly, the study finds a substantial disparity between measured CC yields and SR yields; 2174 kg/ha for CC yields compared to 651 kg/ha for SR yields. The significant disparity between the two measures of yield is partly attributable to the significant intra-plot variability in crop performance leading to plot area loss in the course of the season. This area loss (ranging from 15 to 30% of the planted area) is usually not taken into account in current yield measurement approaches. Delineating the productive and planted-but-unproductive sections of plots has important implications not only for yield estimation methodologies but also for shedding more light on the factors underlying current poor yields and pathways to improving productivity on smallholder rainfed maize farms.

Climate conditions, in complex interrelations with other socioenvironmental, economic, political and cultural factors, significantly play a key role in shaping population density and movement. However, the impacts of climate change on global human migration remain poorly understood. This study assesses the sensitivity of international migration to climate change across 160 countries and regions, focusing on temperature-related climate indices. The results show that migration patterns are strongly influenced by a country’s baseline climate and economic conditions. Under global warming, countries in mid- to high-latitudes with lower baseline temperatures are projected to experience substantial increases in migration inflows, while low-latitude countries are likely to see rising migration outflows. This climate sensitivity is especially pronounced in higher Gross Domestic Product (GDP) countries, whereas lower-GDP countries show little dependence on baseline climate. A simple predictive model based on projected changes in climate and GDP suggests that populations in low-GDP, low-latitude countries will face worsening conditions and increasing migration pressure by the end of the 21st century. The results underscore the growing influence of economic factors on migration and call for urgent integration of climate justice into migration policy to support vulnerable populations in a warming world.

Background: Crop yields are lowest in sub-Saharan Africa compared to other regions, and this is true even for such an important staple crop as maize. Persistence of patches of low crop vigour side-by-side to patches with healthier maize crops has been shown to significantly contribute to low yields on smallholdings. Farmers' perspectives on the presence of such poor patches are important as far as their on-farm investment attitudes are concerned. We analyse maize yield levels and farmers’ perspectives of their production levels in two farming communities in rural Ghana.Results: We find substantial potential for yield improvements; while local attainable yields (average of the yields attained by the top 10% of farmers in each village) were 4.4 t/ha and 3.6 t/ha, average crop cut yields were 2.0 t/ha and 2.4 t/ha for Asitey and Akatawia, respectively. As much as 62% of the maize fields in both study locations were unable to reach the respective average village yield level. From the photo-elicitation interviews, the general attitude of smallholders to the presence of poor patches is that of indifference. We find contradictions in farmers’ perceptions and attitudes towards low yields. While more than half (54%) perceived they were getting adequate yields relative to their expectations, an even greater proportion (88%) of farmers interviewed aver that their plots could yield much more. Similarly, a significant majority (63%) did not attempt to remedy the poor patches even though the same proportion perceive that it is worth it to invest in yield-improving inputs.Conclusions: Farmers in such contexts view investments in fertilizers on their farms as too risky. As alternatives, they would rather invest their already limited resources in non-farm ventures. Farmers opt for yield optimization rather than maximization and this has important implications for diversification off the farm. These findings have important implications for smallholder households’ ability to meet their subsistence needs and for efforts to reduce yield gaps on small farms particularly in resource-poor contexts.

Yield levels and the factors determining crop yields is an important strand of research on rainfed family farms. This is particularly true for Sub-Saharan Africa (SSA), which reports some of the lowest crop yields. This also holds for Ghana, where actual yields of maize, the most important staple crop, are currently about only a third of achievable yields. Developing a comprehensive understanding of the factors underpinning these yield levels is key to improving them. Previous research endeavours on this frontier have been incumbered by the mono-disciplinary focus and/or limitations relating to spatial scales, which do not allow the actual interactions at the farm level to be explored. Using the sustainable livelihoods framework and, to a lesser extent, the induced innovation theory as inspiring theoretical frames, the present study employs an integrated approach of multiple data sources and methods to unravel the sources of current maize yield levels on smallholder farms in two farming villages in the Eastern region of Ghana. The study relies on farm and household survey data, remotely-sensed aerial photographs of maize fields and photo-elicitation interviews (PEIs) with farmers. These data cover the 2016 major farming season that spanned the period March–August. We found that the factors that contributed to current yield levels are not consistent across yield measures and farming villages. From principal component analysis (PCA) and multiple linear regression (MLR), the timing of maize planting is the most important determinant of yield levels, explaining 25% of the variance in crop cut yields in Akatawia, and together with household income level, explaining 32% of the variance. Other statistically significant yield determinants include level of inorganic fertiliser applied, soil penetrability and phosphorus content, weed control and labour availability. However, this model only explains a third of the yields, which implies that two-thirds are explained by other factors. Our integrated approach was crucial in further shedding light on the sources of the poor yields currently achieved. The aerial photographs enabled us to demonstrate the dominance of poor crop patches on the edges and borders of maize fields, while the PEIs further improved our understanding of not just the causes of these poor patches but also the factors underpinning delayed planting despite farmers’ awareness of the ideal planting window. The present study shows that socioeconomic factors that are often not considered in crop yield analyses—land tenure and labour availability—often underpin poor crop yields in such smallholder rainfed family farms. Labour limitations, which show up strongly in both in the MLR and qualitative data analyses, for example, induces certain labour-saving technologies such as multiple uses of herbicides. Excessive herbicide use has been shown to have negative effects on maize yields.