Search

Non‐crop (segetal) plants in arable systems are commonly perceived simply as ‘weeds’, that is, harmful at worst and undesirable at best. The increase in management intensity in European arable systems has vastly reduced the populations of all but the most disturbance‐tolerant plant species, negatively impacting the whole agricultural food web. In recent years, efforts have been made to promote agricultural biodiversity through measures such as flower strips and unsprayed field margins. However, studies of their impacts on the arable flora have rarely considered their spatial variation within the crop field. We investigated the spatial distribution of vascular plant species richness and their contribution to the food web via biomass and flower units in conventional and agri‐environment cereal fields in six regions of Germany. We studied two types of in‐crop measures (extensive cereals without pesticides or fertiliser, and with or without intercropping with flowering species) and one adjacent measure (neighbouring flower strip), recording at 1‐m intervals from the field edge to the interior. These results were then extrapolated to illustrate the effects of these measures on resource provision at the field scale. Species richness and plant biomass dropped off sharply after the first metre in the conventional treatments, regardless of the adjacent habitat. The ‘extensive’ treatments maintained a much higher level of diversity and resource provision in the field interior. At the field level, this can mean more than a 60‐fold difference in the provision of flowering resources between conventional management (1900 flower units/ha) and agri‐environment measures (AEMs) (127,000 units/ha for extensive cereals). Synthesis and applications. The strong edge effects we found in conventional cultivation support the premise that reducing field sizes could play a role in promoting in‐crop biodiversity. However, incorporating extensive field margins as an AEM would be more efficient at maximising the diversity of generalists whilst maintaining high yields. [ABSTRACT FROM AUTHOR]

Floral resources for native pollinators that live in wildland settings are diverse and vary across and within growing seasons. Understanding floral resource dynamics and management is becoming increasingly important as honeybee farms seek public land for summer pasture. Small Unmanned Aircraft Systems (sUASs) present a viable approach for accurate broad floristic surveys and present an additional solution to more traditional alternative methods of vegetation assessment. This methodology was designed as a simplified approach using tools frequently available to land managers. The images of three subalpine meadows were captured from a DJI Phantom 4 Pro drone platform three times over the growing season in 2019 in Sanpete County, Utah. The images were composited using Pix4D software 4.5.6 and classified using a simple supervised approach in ENVI 4.8 and ArcGIS Pro 2.4.3 These same meadows were assessed using two traditional ocular methods of vegetation cover–meter-squared quadrats and macroplot estimation. The areas assessed with these methods were compared side by side with their classified counterparts from drone imagery. Classified images were not only found to be highly accurate when detecting overall floral cover and floral color groups (76–100%), but they were also strongly correlated with quadrat estimations, suggesting that these methods used in tandem may be a conducive strategy toward increased accuracy and efficiency when determining floral cover at broad spatial scales. [ABSTRACT FROM AUTHOR]

Floral resources for native pollinators that live in wildland settings are diverse and vary across and within growing seasons. Understanding floral resource dynamics and management is becoming increasingly important as honeybee farms seek public land for summer pasture. Small Unmanned Aircraft Systems (sUASs) present a viable approach for accurate broad floristic surveys and present an additional solution to more traditional alternative methods of vegetation assessment. This methodology was designed as a simplified approach using tools frequently available to land managers. The images of three subalpine meadows were captured from a DJI Phantom 4 Pro drone platform three times over the growing season in 2019 in Sanpete County, Utah. The images were composited using Pix4D software 4.5.6 and classified using a simple supervised approach in ENVI 4.8 and ArcGIS Pro 2.4.3 These same meadows were assessed using two traditional ocular methods of vegetation cover–meter-squared quadrats and macroplot estimation. The areas assessed with these methods were compared side by side with their classified counterparts from drone imagery. Classified images were not only found to be highly accurate when detecting overall floral cover and floral color groups (76–100%), but they were also strongly correlated with quadrat estimations, suggesting that these methods used in tandem may be a conducive strategy toward increased accuracy and efficiency when determining floral cover at broad spatial scales.