Land cover mapping of large areas is challenging due to the significant volume of satellite data to acquire and process, as well as the lack of spatial continuity due to cloud cover. Temporal aggregation—the use of metrics (i.e., mean or median) derived from satellite data over a period of time—is an approach that benefits from recent increases in the frequency of free satellite data acquisition and cloud-computing power. This enables the efficient use of multi-temporal data and the exploitation of cloud-gap filling techniques for land cover mapping. Here, we provide the first formal comparison of the accuracy between land cover maps created with temporal aggregation of Sentinel-1 (S1), Sentinel-2 (S2), and Landsat-8 (L8) data from one-year and test whether this method matches the accuracy of traditional approaches. Thirty-two datasets were created for Wales by applying automated cloud-masking and temporally aggregating data over different time intervals, using Google Earth Engine. Manually processed S2 data was used for comparison using a traditional two-date composite approach. Supervised classifications were created, and their accuracy was assessed using field-based data. Temporal aggregation only matched the accuracy of the traditional two-date composite approach (77.9%) when an optimal combination of optical and radar data was used (76.5%). Combined datasets (S1, S2 or S1, S2, and L8) outperformed single-sensor datasets, while datasets based on spectral indices obtained the lowest levels of accuracy. The analysis of cloud cover showed that to ensure at least one cloud-free pixel per time interval, a maximum of two intervals per year for temporal aggregation were possible with L8, while three or four intervals could be used for S2. This study demonstrates that temporal aggregation is a promising tool for integrating large amounts of data in an efficient way and that it can compensate for the lower quality of automatic image selection and cloud masking. It also shows that combining data from different sensors can improve classification accuracy. However, this study highlights the need for identifying optimal combinations of satellite data and aggregation parameters in order to match the accuracy of manually selected and processed image composites.

The potential to capture additional air pollutants by introducing more vegetation or changing existing short vegetation to woodland on first sight provides an attractive route for lowering urban pollution. Here, an atmospheric chemistry and transport model was run with a range of landcover scenarios to quantify pollutant removal by the existing total UK vegetation as well as the UK urban vegetation and to quantify the effect of large-scale urban tree planting on urban air pollution. UK vegetation as a whole reduces area (population)-weighted concentrations significantly, by 10% (9%) for PM _2.5 , 30% (22%) for SO ₂ , 24% (19%) for NH ₃ and 15% (13%) for O ₃ , compared with a desert scenario. By contrast, urban vegetation reduces average urban PM _2.5 by only approximately 1%. Even large-scale conversion of half of existing open urban greenspace to forest would lower urban PM _2.5 by only another 1%, suggesting that the effect on air quality needs to be considered in the context of the wider benefits of urban tree planting, e.g. on physical and mental health. The net benefits of UK vegetation for NO ₂ are small, and urban tree planting is even forecast to increase urban NO ₂ and NO x concentrations, due to the chemical interaction with changes in BVOC emissions and O ₃ , but the details depend on tree species selection. By extrapolation, green infrastructure projects focusing on non-greenspace (roadside trees, green walls, roof-top gardens) would have to be implemented at very large scales to match this effect. Downscaling of the results to micro-interventions solely aimed at pollutant removal suggests that their impact is too limited for their cost–benefit analysis to compare favourably with emission abatement measures. Urban vegetation planting is less effective for lowering pollution than measures to reduce emissions at source. The results highlight interactions that cannot be captured if benefits are quantified via deposition models using prescribed concentrations, and emission damage costs. This article is part of a discussion meeting issue ‘Air quality, past present and future’.

Pollination services are affected by landscape context, farming management and pollinator community structure, all of which impact flower visitation rates, pollen deposition and final production. We studied these processes in Argentina for highbush blueberry crops, which depend on pollinators to produce marketable yields. We studied how land cover and honeybee stocking influence the abundance of wild and managed pollinators in blueberry crops, using structural equation modelling to disentangle the cascading effects through which pollinators contribute to blueberry fruit number, size, nutritional content and overall yield. All pollinator functional groups responded to landscape changes at a spatial scale under 1000 m, and the significance or direction of the effects were modulated by the field‐level deployment of honeybee hives. Fruit diameter increased with pollen deposited, but decreased with honeybee abundance, which, had indirect effects on fruit acidity. Honeybees had a positive effect on the number of fruit produced by the plants and also benefited the overall yield (kg plant⁻¹) through independent effects on both the quality and quantity components of fruit production. Synthesis and applications. Deployment of beehives in blueberry fields can buffer, but not compensate for the negative effects on honeybee abundance produced by surrounding large scale none‐flowering crops. Such compensation would require high‐quality beehives by monitoring their health and strength. The contribution of honeybees to crop production is not equal across production metrics. That is, higher abundance of honeybees increases the number of berries produced but at the cost of smaller and more acidic fruits, potentially reducing their market value. Growers must consider this trade‐off between fruit quantity and quality when actively managing honeybee abundance.

We report the results of a structured expert elicitation to identify the most likely types of potential food system disruption scenarios for the UK, focusing on routes to civil unrest. We take a backcasting approach by defining as an end-point a societal event in which 1 in 2000 people have been injured in the UK, which 40% of experts rated as “Possible (20–50%)”, “More likely than not (50–80%)” or “Very likely (>80%)” over the coming decade. Over a timeframe of 50 years, this increased to 80% of experts. The experts considered two food system scenarios and ranked their plausibility of contributing to the given societal scenario. For a timescale of 10 years, the majority identified a food distribution problem as the most likely. Over a timescale of 50 years, the experts were more evenly split between the two scenarios, but over half thought the most likely route to civil unrest would be a lack of total food in the UK. However, the experts stressed that the various causes of food system disruption are interconnected and can create cascading risks, highlighting the importance of a systems approach. We encourage food system stakeholders to use these results in their risk planning and recommend future work to support prevention, preparedness, response and recovery planning.

There is strong evidence that landscape‐scale factors such as habitat diversity, composition and configuration are important drivers of declines in pollinators and pollination services. However, context and species‐specific responses make it challenging to draw general conclusions about the most important components of landscapes that support diverse and abundant pollinator communities. In this study, we took a functional‐traits approach to community assembly and tested the hypothesis that landscape properties act most strongly on pollinators indirectly, through their influence on flowering plant communities. Using plant and pollinator data from 96 landscapes in Britain, we tested the associations between plant and pollinator communities and local environmental factors, such as habitat cover and configuration, using path analysis based on Mantel and partial Mantel statistics. When all pollinators were considered, we found that the environmental factors had stronger links to the composition of flowering plant communities than to the composition of pollinator communities. Further, the flowering plant community was strongly linked to the pollinator community suggesting a mediating role between land use and pollinators. When separating the pollinator community into taxonomic groups, we found the same result for hoverflies, but wild bees were linked to both environmental factors and flowering plants. We further explored these links with structural equation models using the response‐effect trait framework as a guiding principle. We found strong evidence that land‐use composition and configuration influence the trait distribution and functional diversity of the pollinator community via plant community composition. These findings suggest that the indirect effect of land use on pollinators via flowering plants should be considered in informing the design of pollinator friendly landscapes and in future research of the effects of land use and management on wild pollinators. Read the free Plain Language Summary for this article on the Journal blog.