Implanting new plots of mass-flowering resources in landscapes can have both positive and negative effects on pollinator visitation rates to crops. We investigated the effects of flowering plot characteristics on the best places to locate new co-flowering plots to optimise crop pollination. First,we parameterised and validated an existing pollination model with field data. Then,we ran two simulation experiments,both of which simulated the conversion of one or multiple plots by modifying their flowering and nesting resource attributes. We quantified changes in pollinator visitation rates to "target" sunflower fields according to different conversion scenarios.The data set is composed of (i) the validation of the spatially explicit InVEST pollination model using field data collected in the Vallées et Coteaux de Gascogne (France),and (ii) the results associated with the two simulation experiments using InVEST pollination model. For the validation of model parameters,we sampled bees in 30 sites with three coloured pan traps per sampling site. We compared the predicted visitation rates vs the observed total abundance of bees in the pan traps. The visitation rate predictions were calculated in different radii around the sampling sites (from 1 to 100 m in increments of 1 m). The first experiment simulated the conversion of the flowering and nesting attributes of 1800 plots according to four scenarios (wildflower plot,sunflower field,More flowers,More nests). Changes in visitation rates induced by plot conversion were assessed in 368 target sunflower fields and then analysed according to the distance separating the converted plots from the target sunflower fields. For statistical replication,we divided our study area into 15 circular sectors. The radius of the circular sectors was 5 km and the sectors were centered on the centroid of a target sunflower field. The second simulation experiment quantified changes in visitation rates caused by the conversion of multiple plots into sunflower fields which are isolated to greater (Isolation scenario) or less extents (Aggregation scenario). The target sunflower fields used to assess changes in visitation rates were the 15 sunflower fields defining the centre of the 15 circular sectors and the converted sunflower fields.

The aim was to assess the pollens carried by wild bees hunted in three types of semi-naturel habitats :permanent grasslands,forest edges and hedgerows. The bees were netted in each of these habitats,in 30 landscapes selected according to a gradient of % woodland. Pollen grains were collected on female bee’s body with a fine tweezer and a moistened brush and were deposited directly into a droplet of water placed on a microscope slide. Pollen grains from each individual bee were observed under a x400 optical microscope and identified to genera or species level based on the pollen reference collection of the INRAe du Magneraud.

Données d'enquêtes produites dans le cadre du projet COTERRA dans l’ancien canton d'Aurignac lors du stage d'Anouk Leclerc, l'enquête portait sur l'influence des politiques publiques agricoles sur les trajectoires d'exploitation déjà enquêtées en 2006_2007.

The aim was to assess the difference in bee communities in three types of wooded and herbaceous habitats (permanent grassland,hedgerow and forest edge). The bees were netted in each of these habitats,in 30 landscapes selected according to a gradient of % of woodland. Wild bees were collected in May and June 2016 using an insect net along transects in each sampling point. Bees were hunted for 10 minutes,excluding time of capture and preparation of each individual insect. The transects were 100 m long and 5 m wide and were located in the middle of grasslands and along forest edges and hedgerows.

The aim was to assess the difference in plant communities in three types of wooded and herbaceous habitats (permanent grassland,hedgerow and forest edge). Plants were surveyed in each of these habitats,in 30 landscapes selected according to a gradient of % of woodland. Botanical surveys were conducted in May and June 2016 in forest edges and hedgerows,and in April and May 2015 and May and June 2016 in permanent grasslands. The surveys comprised abundance-dominance records of all vascular plant species according to the Braun-Blanquet scale in three vegetation layers (herbaceous:0–1 m,shrubs:1–3 m and trees:> 3 m). Data were collected along a 25-m transect on each side of the hedgerows,on a 50-m transect along the forest edges,and wandering freely in grasslands.

Ensemble des opérations

L’atlas réalisé par Jean-Louis Edouard lors de sa thèse rassemble les cartes bathymétriques de 43 des lacs étudiés par l'auteur,et des photographies de ces lacs. Vous le trouverez page 681 du manuscrit (693 du pdf) disponible sur HAL.