@Article{AlmeidaCampFranEbec:2022:NeNeDa,
               author = "Almeida, Vin{\'{\i}}cius Albuquerque de and Campos Velho, 
                         Haroldo Fraga de and Fran{\c{c}}a, Gutemberg Borges and Ebecken, 
                         Nelson Francisco Favilla",
          affiliation = "{Universidade Federal do Rio de Janeiro (UFRJ)} and {Instituto 
                         Nacional de Pesquisas Espaciais (INPE)} and {Universidade Federal 
                         do Rio de Janeiro (UFRJ)} and {Universidade Federal do Rio de 
                         Janeiro (UFRJ)}",
                title = "Neural networks for data assimilation of surface and upper-air 
                         data in Rio de Janeiro",
              journal = "Geoscientific Model Development Discussions",
                 year = "2022",
               volume = "2022",
                month = "Sept.",
             abstract = "The practical feasibility of neural networks models for data 
                         assimilation using local observations data in the WRF model for 
                         the Rio de Janeiro metropolitan region in Brazil is evaluated. 
                         Surface and multi-level variables retrieved from airport 
                         meteorological stations are used: air temperature, relative 
                         humidity, and wind (speed and direction). Also, 6-hour forecast 
                         from WRF high-resolution simulations are used domain centered in 
                         the Rio de Janeiro city with nested grids of 8 and 2.6 km. Periods 
                         of 168h from 2015-2019 are used with 6h and 12h assimilation 
                         cycles for surface and upper-air data, respectively, applied to 
                         6-hour forecast fields. The observed data (interpolated to grid 
                         points close to airport locations and influence computed in its 
                         surroundings) and short-range forecasts are used as input for 
                         training model and the 3D-Var analysis on 6-hour forecast fields 
                         for each grid point is used as target variable. The neural network 
                         models are built using two different approaches: WEKA multilayer 
                         perceptron model and TensorFlows deep learning implementation. The 
                         year of 2019 is used as an independent dataset for forecast 
                         validation from the trained models. Results employing 6-hour 
                         forecast fields with neural network models are able to emulate the 
                         3D-Var results for surface and multi-level variables, with better 
                         results for the NN-TensoFlow implementation. The main result 
                         refers to CPU time reduction enabled by the neural networks 
                         models, reducing the data assimilation CPU-time by 121 times and 
                         25 times for NN-TensorFlow and NN-WEKA, respectively, in 
                         comparison to the 3D-Var method under the same hardware 
                         configurations.",
                  doi = "10.5194/gmd-2022-50",
                  url = "http://dx.doi.org/10.5194/gmd-2022-50",
                 issn = "1991-962X and 1991-9611",
             language = "en",
           targetfile = "gmd-2022-50.pdf",
        urlaccessdate = "20 maio 2024"
}