Weed vegetation, crop yield and quality and management of 26 wheat fields of Mediterranean Spain

[Method] The study was carried out in 2013–2014 across 26 wheat fields (Triticum aestivum L. and Triticum turgidum subsp. durum Desf.) under a Mediterranean climate in Spain. Fields were either organically or conventionally managed and were located in cereal areas in four Spanish regions: Andalusia (15 fields), Madrid (3 fields), Castilla-La Mancha (4 fields) and Catalonia (4 fields). In each field, ten (1 m × 1 m) plots were established during crop tillering and maintained until harvest. Weeds were sampled at the end of the crop vegetative period (April with dates slightly varying depending on field location). In each plot we recorded the weed species present, and we visually assessed the cover (in cm2) of each weed species and the crop. We also measured plant height (cm). Immediately before crop harvest, we sampled the plots again to obtain crop yield. In each plot we counted the number of wheat stems with ears and cut 30 of them. We determined the dry weight (after 48 h at 65 °C) of the 30 stems, threshed the ears and weighed the grain for each sample. In doing this we obtained an average value for the grain weight of an ear. We calculated crop grain weight in each plot multiplying mean grain weight of each sample by the total number of fertile stems in the plot. In each plot, we obtained two measures of grain quality: the percentage of total dry protein content determined using the Kjeldahl standard method at the Laboratorio Agroalimentario de Córdoba (Córdoba, Spain) and the glutenin to gliadin ratio. The amount of glutenin and gliadin in each plot, was obtained from 100 mg of wheat flour. First, we ground the wheat grains of each sample using a ball mill to obtain flour of a 100 µm particle size. The extraction of gliadin and glutenin proteins was done using a modified classical Osborne procedure based on protein solubility. The method is described in detail in Wieser et al. (1998) and Pistón et al. (2011). Gliadin protein was extracted stepwise three times, samples were centrifuged, and the supernatans were collected and pooled. The insoluble material from the previous step was used to obtain the glutenin fraction in a similar manner. Then, each of the extracts were filtered and they were applied to a 300SB-C8 reverse phase analytical column using a 1200 Series Quaternary LC System liquid chromatograph (Agilent Technologies) with a DAD UV-V detector. Absorbance was monitored with the DAD UV-V module at 210 nm. The amounts of both fractions were determined using bovine serum albumin as protein standard. Both fractions were expressed as μg/mg flour. The glutenin to gliadin ratio was calculated by dividing the amount of glutenins by gliadin content. For each field we also recorded the crop species and the variety, and collected data on management practices: sowing date (month), fertilization rate (kg N/ha), preceding crop (legume, fallow, sunflower or cereal; categories depending on the nutrient demand of the crop), type of tillage and herbicides used. We also obtained data on the average monthly temperature (ºC) and total precipitation (mm) during crop growth season from the nearest meteorological stations. [Methods for processing the data] Different plant community diversity indices were computed from data. Statistical analyses were performed to understand the relationship between crop quality and weeed community diversity as explained in: Paper accepted for publication.

Weed community structure, including composition and taxonomic and functional diversity, may explain variability in crop quality, adding to the variability accounted for by management, climatic and genetic factors. Focusing on Mediterranean rainfed wheat crops, we sampled weed communities from 26 fields in Spain that were either organically or conventionally managed. Weed communities were characterized by their abundance and taxonomic, compositional and trait-based measures. Grain protein concentration and the glutenin-to-gliadin ratio were used as indicators of wheat grain quality. Linear mixed-effects models were used to analyze the relationship between crop quality and weed community variables, while accounting for environmental factors. Nitrogen fertilization, previous crop and precipitation explained an important part of wheat grain protein concentration (R2marginal = 0.39) and composition (R2marginal = 0.26). Weed community measures had limited effects on grain quality (increasing R2marginal of models by 1% on average). The weed effects were related to the composition and the functional structure of their communities but not to their abundance. Environmental conditions promoting higher protein concentration also selected for weed species with competitive attributes, whereas the role of weed functional diversity depended on the functional trait and on the resource limiting crop grain quality. Understanding the mechanisms of weed effects on crop quality could aid in designing sustainable weed management practices.

Grants AGL2012-33736 funded by MCIN/AEI/10.13039/501100011033 and by “ERDF A way of making Europe”.

Peer reviewed