Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → Espagne → 2008 → Efecto de la manipulación de la diversidad sobre el establecimiento de comunidades vegetales en tierras de cultivo abandonadas en la dehesa salmantina

Universidad de Salamanca (2008)

Efecto de la manipulación de la diversidad sobre el establecimiento de comunidades vegetales en tierras de cultivo abandonadas en la dehesa salmantina

Alvarez Díaz, Jimmy Edgard

Titre : Efecto de la manipulación de la diversidad sobre el establecimiento de comunidades vegetales en tierras de cultivo abandonadas en la dehesa salmantina.

Auteur : Alvarez Díaz, Jimmy Edgard

Université de soutenance : Universidad de Salamanca

Grade : Doctoral Tesis 2008

Résumé
Old-field succession was studied during seven years of field experiments, under the CLUE project (CHANGING LAND USAGE, Enhancement of biodiversity and ecosystem functioning). The secondary succession on former arable land is dominated by opportunistic species (mainly arable weeds from the seed bank), which are major constraints in vegetation dynamics due to their delayed force role against the restoration of existing natural conditions prior to conversion. Our aim then was to change the initial vegetation succession stages by the direct sowing of later-successional plant species in two different seed mixture treatments : one contained 4 (low diversity, BD) and the other 15 (high diversity, AD) sown plant species. We compared the plant communities’ development between both seed mixture treatments and with those of unsown plots (CN) following a randomized block design. Only four of the fifteen sown species were dominant over all the experimental period, the other half were left as subordinate species while the remain ones completely disappeared. The most successful sown species to establish in sowing treatments was the exotic grass, Bromus inermis, which had a strong and negative effect on the other sown species success such as Plantago lanceolata, Sanguisorba minor and the legume Lotus corniculatus. However, these latter species gained certain dominance degree within the established plant communities in absence of the former sown grass species. In general terms, when sown species declined in both number and abundance over time, BD treatment tended to respond in similar way to AD treatment, in spite of being more related to unsown conditions (CN), during most part of the time, due to the lack of a dominant sown species that would able to suppress weeds. The sowing effect on the suppression of unsown species was more evident at the scale of subplots (1 sqm) rather than whole-plots (100 sqm), and was mainly caused by the decrease in the number of colonizing species in plots with sown species, which was most pronounced in AD plots. The lowest scale was associated with the formation of heterogeneous patches by Bromus inermis while at highest scale this effect was not clear because weeds had mayor chances of expression among the different established patches. Therefore, only a fraction of the weed species within a whole-plot will be found in the neighborhood of B. inermis and only a few of them were able to increase its abundance. This result was according to the Elton hypothesis in the sense that more diverse communities should be less susceptible to invasion by colonizing species. Consequently, B. inermis promotes a plant functional change because it excluded both sown and unsown grasses but facilitated other non-legume plant species. Despite of the important functional effect accounted, sowing did not affect the established plant community productivity. Moreover, the plant communities exhibited the same negative relationship between diversity and productivity in similar way as many authors have reported for various dehesa’s zones. Therefore, other diversity components were more important than the number of species per se in explaining ecosystem functioning (including the productivity) ; for example, the species composition with high number of low-biomass species and low number of legumes. Finally, the diversity effect was highly species and context dependent and that the identity of B. inermis was the most significant because it changes functionally the established plant community but without increasing any species biomass in the process.

Présentation

Version intégrale (10,04 Mb)

Page publiée le 29 janvier 2016, mise à jour le 18 mars 2017