Ir direto para menu de acessibilidade.
Portal do Governo Brasileiro




Home > Sem categoria > INTERNSHIPS ABROAD BY THE DOCTORATE SANDWICH PROGRAM (PDSE)


INTERNSHIPS ABROAD BY THE DOCTORATE SANDWICH PROGRAM (PDSE)



Student: João Arthur Pompeu Pavanelli
Local: Basque Centre for Climate Change – BC3-  Basco, Espanha
Period: August/2018 – January/ 2019.

The internship was eventually trained with the ARIES semantic metamodeling framework (ARIES) for spatially explicit modeling of ecosystem services. ARIES is developed at BC3 as part of the Human and Natural Systems Coupled Modeling Research Line, which is the integrator of the Center’s other four lines, which address the physical understanding and ecological and social implications of climate change. The main activities performed with this group during the internship period were:

1) Training for the use of the ARIES tool – By integrating the team that works with the development and application of the ARIES tool, the training process is done in two ways. The first, introductory, was the International Spring University on Ecosystem Services Modeling Course (https://springuniversity.bc3research.org/isu-2018/), organized by BC3 with a 40-hour workload between 8 and October 12, 2018. This course was taught by the ARIES international team to participants from various countries in Latin America, Africa and Europe, and enabled the initiation to practical use of the modeling tool. After the course, I deepened my knowledge of the tool by working directly with the BC3 team to cover gaps that were not addressed, such as working from the interface for model development, as the course covered only the interface for end use of templates. This allows the customization and application of the tool for the thesis context and other contexts of interest for application in Brazilian landscape studies.

2) Data collection, generation and processing – The first weeks in BC3 were devoted to collecting the latest missing geographic data, refining the data and finally processing spatial data for use in the model. The data collected during the internship period were the 2017 update of Municipal Agricultural Production (PAM) and Municipal Livestock Survey (PPM), available from IBGE. In addition, land use and land cover data derived from MapBiomas collection 3.0 were made available on the platform’s website also at the beginning of the internship. The spatialization of agricultural crops surveyed by the IBGE for the last 30 years in spatial resolutions of 10 km and 1 km was possible from the use of a supercomputer provided by BC3 only for processing these data, which, together with the reprocessed maps of use and land cover in 1 km, are the input variables for the spatially explicit modeling of ecosystem services addressed in this thesis. The use of the supercomputer is justified by the amount of data, whose processing took approximately two months and required more than 2 Tb of storage and at least 80 Gb of memory. The thousands of maps generated, comprising area, productivity and production of 62 agricultural crops between 1988 and 2017 in the national territory, as well as pasture stocking during this period, can support several studies on the spatiotemporal dynamics of Brazilian agriculture and will be available in raster format for use by the scientific community. However, they are already available for application of some spatially explicit models in ARIES. As an immediate scientific product derived from these data, an article on changes in the availability of macro and micronutrients for the Brazilian population is being written, in partnership with scientists from CCST / INPE and the University of Minnessota, in the United States.

3) Case study – With the thesis data prepared for use in ARIES, we developed in BC3 a case study on pollination of agricultural crops, one of the ecosystem services that is greatly threatened by land cover conversion and has a significant impact on food. human resources and the production of the agricultural sector. We selected a region of the Cerrado for the study of temporal space and, at the moment, we are parameterizing the model with specific information about pollinators for Brazilian landscapes. This study will result in a working paper to be published by BC3 and a scientific article in a journal specializing in ecosystem services. This working paper was requested by the BC3 editorial team as a scientific contribution to the Center in return for my office.

Student: Leticia d’Agosto Miguel Fonseca
Local: Universidade de Oxford – UK
Period: August/2018 – January/2019.

The internship took place at the Department of Geography at Oxford University, in the ecosystem lab led by researcher Yadvinder Malhi. The laboratory is composed of approximately 10 researchers from various areas of formation, as well as masters and doctorate students from the department. During the period, I made two presentations to the research group, the first presentation was given on September 4, 2018, in order to expose the theme of my thesis to the group, and thus, to outline the guidelines of the work that would be held during my stay at the University. At the end of the sandwich period, I performed my second presentation on January 15, 2019, where I showed the results achieved in partnership with my mentors Yadvinder Malhi and Sami Rifai. These two presentations were configured as chapters 1 and 2 of the thesis, respectively. The exchange of knowledge during my presentations was relevant to research in Amazonian floodplains, as we delineated scientific issues related to this theme, as these areas are still poorly studied in Brazil and other tropical forests in the world. This exchange of information was extremely important for my background and also for the Oxford ecosystem research group.

The group is extremely organized and focused on results, so biweekly meetings were held with my mentors, who taught me data analysis techniques involving remote sensing, statistical modeling, field data and flow tower data ( carbon and water) in the Amazon. These techniques were crucial for the finalization of the thesis and understanding of the results, which were discussed during this period with my advisor in Brazil, Laura Borma. In addition to the work focused on the thesis theme, I participated in the group dynamics of the laboratory, where I had the opportunity to discuss weekly two scientific articles about global tropical forests, keeping the reading updated, covering different research sites. This dynamic was fundamental to understand the critical points and the main gaps of the work done in these areas.

Another dynamic held weekly was the discussion of articles under preparation, in which I had the opportunity to submit my first article of the thesis for discussion and, therefore, improve the structure, figures and writing in English. During the term, during which master classes are taught, several seminars were held, which I had the opportunity to participate and talk with researchers from other universities in the UK and Europe, learning about techniques and methodologies. research in tropical forests, besides knowing the projects that are underway in partnership with Brazil. The finalized and ongoing articles resulting from this working period were entitled: “Phenology and seasonal ecosystem productivity in an Amazonian floodplain forest”, which will be submitted to Geophysical Research Letters, and “Estimation of long-term ET / GPP over a tropical floodplain forest with complex hydrology ”, which is still in the process of discussing the results, however, I intend to submit them by August 2019.