Modeling rhizosphere carbon and nitrogen cycling in Eucalypts

• Ano de defesa: 2018
• Autor(a) principal: Valadares, Rafael Vasconcelos
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de Viçosa; Solos e Nutrição de Plantas
• Palavras-chave em Português: Rizosfera; Micro-organismos; Modelos matemáticos; Ciência do Solo
• Link de acesso: https://locus.ufv.br//handle/123456789/29362

Eucalypts as well as other trees allocate large amounts of fixed carbon to produce roots and rhizodepositions, which include exudates, secretions, lysates and gases. The role of these compounds has been studied to elucidate the benefits for the trees of the transfer of matter/energy to the rhizosphere microorganisms. It is currently well known that the results of the interaction between the trees and the microbiota can be positive, neutral or negative, depending on the plant species, associated microbial population, and environmental conditions. This can ultimately determine the growth rate, planting stand, and even whether or not a given tree species will survive in the environment. Numerous studies have been carried out to investigate the effect of the activation of rhizosphere microbial communities previously in a state of oligotrophy or quasi-dormancy by the rhizodepositions, in the so-called rhizosphere priming effect. This process triggers changes in the structure and activity of rhizosphere communities that affect the dynamics of soil organic matter (SOM) decomposition and formation. In the particular case of the soil under eucalypts, this process predominantly accelerates the decomposition of the SOM inside the rhizosphere and, therefore, may present a positive return to the trees by the increase of the nitrogen supply (M.D. Costa, personal information¹). Despite this, there are still no models to measure the quantitative importance of this process for forest plantations of the Eucalyptus genus. Thus, the present thesis aimed to develop a mechanistic model to estimate the carbon (C) and nitrogen (N) cycling in the rhizosphere soil under eucalypts. Before this, for the elaboration of this mathematical representation, it was necessary to take into account that the rhizosphere system is entirely dependent on the activity of the roots and, therefore, it does not make sense its modeling without the prediction of the plant root growth and estimation of the rhizodeposition process. For this reason, the objective of the thesis was also to present and improve a growth model for Eucalyptus in the APSIM (next generation) agricultural models platform. This model has, among other advantages, greater sensitivity to the soil factor, which includes the effect of the N supply by processes of SOM mineralization and litter decomposition. Thus, in Chapter I, it was reviewed the main processes and factors involved in the subsequent chapters, which are more related to the mathematical representation of key processes for rhizosphere mineralization. In Chapter II, the APSIM model was presented and validated for 12 forest sites; out of those 12 sites, nine are geographically located in Brazil and the other three are located in Australia. The APSIM model presented a satisfactory treatment of the growth processes, considering the diverse climatic and edaphic conditions and the different forest managements. However, in order to improve its performance, adjustments were suggested regarding carbon partitioning for the roots, as well as considering other nutritional limitations besides N; and other nutrient flows such as those occurring in the rhizosphere environment. Chapter III, on the other hand, was devoted to the conceptual elaboration, presentation, evaluation and sensitivity analysis of the Forest Plantation Rhizosphere Available Nitrogen model (ForPRAN) for estimates of the C and N mineralization fluxes in the rhizosphere soil. The performance of the model was quantitatively and qualitatively satisfactory when compared to the data observed in the literature. The input variables that most influenced the increase of N by rhizosphere mineralization were (in order of decreasing importance): root diameter> rhizosphere thickness> soil temperature> clay concentration. The rhizosphere mineralization in a typical eucalypt plantation producing 42 m³ ha -1 a -1 of shoot biomass, with assumed N losses of 40 % by different processes, was estimated in 24.6 % of the amount of N accumulated in the plantations (shoot+root+litter). From this point of view, it was concluded that the rhizosphere cycling model should be considered for adaptation of other models of forest and agricultural production, such as APSIM Eucalyptus, where the inclusion of such processes offers the potential to improve the growth predictions. Finally, in Chapter IV, it was tested the mathematical modeling hypothesis that the rhizosphere priming effect (RPE) is quantitatively important for nitrogen nutrition of eucalypts in different climatic and soil conditions. To test this hypothesis, the eucalypts growth was simulated using the general APSIM Eucalyptus model and the rhizosphere processes using the ForPRAN model. Eucalyptus growth was projected at four sites, two in Brazil - Aracruz/ES and Curvelo/MG - and the other two in Australia - Wagga Wagga/NSW and Coffs Harbour/NSW. At the end of the 7-year rotation, considering the cycling in the 0- 20 cm depth layer, it was estimated that 12, 11, 10 and 5 % of the total soil was occupied by the rhizosphere at Aracruz, Coffs Harbour, Wagga Wagga and Curvelo, respectively. Because of the rhizodeposition and of the different pedoenvironments, the ForPRAN model suggested an increase in the rhizosphere microbial biomass in the four sites, whose average values were between 70.3 μg cm -3 and 246.1 μg cm -3 of C. In general, the RPE has the potential to explain between 15 and 38 % of the N demand of the forest ecosystem (shoot+root+litter), considering the studied cases. Higher temperatures and rainfall volumes cause increase in the N rhizosphere supply, in absolute values, in Brazilian sites as compared with the Australian ones, which also explains part of the observed smaller N limitation in eucalypts plantations in Brazil.