A study on the impact of data balance on rainfall prediction through artificial neural networks using surface microwave radiometers

Lourenço José Cavalcante Neto; Alan James Peixoto Calheiros

Lourenço José Cavalcante Neto National Institute for Space Research http://orcid.org/0000-0001-9915-7726
Alan James Peixoto Calheiros National Institute for Space Research (INPE) https://orcid.org/0000-0002-6408-3410

Keywords: Rainfall prediction, Data balancing, Machine learning, Amazon, ATTO Campina

Abstract

The National Institute for Space Research (INPE) has been a partner in significant projects that conduct atmospheric investigations impacting various sectors, such as the Amazon Tall Tower Observatory (ATTO) project. Since 2009, the project has conducted studies on the interactions between climate and the Amazon forest. ATTO has played an essential role in providing large volumes of data obtained by meteorological sensors, contributing to a deeper understanding of the atmospheric dynamics of the region. In a landscape where Artificial Intelligence-based rainfall forecast models gain prominence, this study explores the imbalance of data from the ATTO Campina field experiment and its influence on short-term rainfall forecasts using Artificial Neural Networks (ANNs). Metrics such as MAE, RMSE, and POD, as well as FAR indices, were applied in the assessment and revealed the connection between data balance and forecast results. More balanced data or data with greater weights for different rainfall ranges yield better results. The study emphasizes the importance of reliable data for training rain forecast models, aiming to improve the dexterity of these models. This approach is fundamental to increase the reliability of these models in real environments.

Downloads

Download data is not yet available.

Author Biographies

Lourenço José Cavalcante Neto, National Institute for Space Research

Master's student in Applied Computing at the National Institute for Space Research, INPE. Holds a specialization in Informatics in Education from the Cultural Union College of the State of São Paulo (UCESP), completed in 2016, with a degree in Computing from the Federal Institute of Tocantins (IFTO), Araguatins campus, obtained in 2015. Also has technical training in Internet Informatics from IFTO, Palmas campus, obtained in 2014. Between March 2016 and May 2019, was part of the faculty at the Federal Institute of Mato Grosso (IFMT), Guarantã do Norte Advanced Campus, and is currently a permanent member of the faculty at the Federal Institute of Tocantins (IFTO), Araguatins campus. The main areas of expertise are primarily focused on Artificial Intelligence, Database, and System Development in PHP and Python languages.

Alan James Peixoto Calheiros, National Institute for Space Research (INPE)

Possui graduação em Meteorologia pela Universidade Federal de Alagoas (2006), Mestrado (2008) e Doutorado (2013) em Meteorologia pelo Instituto Nacional de Pesquisas Espaciais (INPE). Desde 2015 é Tecnologista do INPE. Entre 2015 e 2017 desenvolveu atividades na Divisão de Operações do Centro de Previsão de Tempo e Estudos Climáticos (CPTEC/INPE). Atualmente faz parte do grupo do Laboratório Associado de Computação e Matemática Aplicada (LabAC/CoCTE/INPE) e membro permanente do programa de pós-graduação do INPE em Computação Aplicada (CAP). Tem experiência na área de Meteorologia, com ênfase em Sensoriamento Remoto da Atmosfera, atuando principalmente nos seguintes temas: Sistemas Automáticos de Previsão a Curto Prazo de Tempestades, Estimativa de Precipitação por Satélites e Radares Meteorológicos, Microfísica de Nuvens e Precipitação e Modelagem dos Processos Radiativos para Sensoriamento Remoto da Atmosfera

References

L. Pires, K. Sušelj, L. Rossato, and J. Teixeira, “Analyses of Shallow Convection over the Amazon Coastal Region Using Satellite Images, Data Observations and Modeling,” Rev. Bras. Meteorol., vol. 33, pp. 366–379, Jun. 2018, doi: 10.1590/0102-7786332009.

M. O. Andreae et al., “The Amazon Tall Tower Observatory (ATTO): overview of pilot measurements on ecosystem ecology, meteorology, trace gases, and aerosols,” Atmospheric Chem. Phys., vol. 15, no. 18, pp. 10723–10776, Sep. 2015, doi: 10.5194/acp-15-10723-2015.

V. A. Costa, “A narrative review of papers developed on the Amazon Tall Tower Observatory experimental site,” Res. Soc. Dev., vol. 10, no. 14, p. e73101421749, Oct. 2021, doi: 10.33448/rsd-v10i14.21749.

W. R. Moninger et al., “Summary of the First Conference on Artificial Intelligence Research in Environmental Sciences (AIRIES),” Bull. Am. Meteorol. Soc., vol. 68, no. 7, pp. 793–800, Jul. 1987.

D. W. McCann, “A Neural Network Short-Term Forecast of Significant Thunderstorms,” Weather Forecast., vol. 7, no. 3, pp. 525–534, Sep. 1992, doi: 10.1175/1520-0434(1992)007<0525:ANNSTF>2.0.CO;2.

A. P. Almeida, “Uso de redes neurais para previsão de descargas elétricas nuvem-solo a partir de dados de radar: Um estudo de caso na Amazônia”, Dissertation (Masters in Applied Computing), National Institute for Space Research, São José dos Campos-SP, Brazil, Sep. 2021.

S. Haykin, Neural Networks: A Comprehensive Foundation (2nd Edition) Neural Networks: A Comprehensive Foundation. 1998.

Z. Basha, N. Bhavana, P. Bhavya, and S. V, Rainfall Prediction using Machine Learning & Deep Learning Techniques. Aug. 2020, p. 97. doi: 10.1109/ICESC48915.2020.9155896.

M. Akkisetti, M. Rajeevan, V. R. Madineni, J. Bhate, and c. v Naidu, “Nowcasting severe convective activity over southeast India using ground-based microwave radiometer observations,” J. Geophys. Res. Atmospheres, vol. 118, pp. 1–13, Jan. 2013, doi: 10.1029/2012JD018174.

M. Darji, V. Dabhi, and H. Prajapati, Rainfall forecasting using neural network: A survey. Jul. 2015. doi: 10.1109/ICACEA.2015.7164782.

A. Maitra and R. Chakraborty, “Rain prediction using radiometric observations at a tropical location,” in 2017 XXXIInd General Assembly and Scientific Symposium of the International Union of Radio Science (URSI GASS), Aug. 2017, pp. 1–4. doi: 10.23919/URSIGASS.2017.8105124.

J. Joss and A. Waldvogel, “Ein Spektrograph fur Niedersch Pagsmessungen mit automatischer Auswertung,” Pure Appl. Geophys., vol. 68, pp. 240–246, Dec. 1967, doi: 10.1007/BF00874898.

C. Caracciolo, F. Prodi, and R. Uijlenhoet, “Comparison between Pludix and impact/optical disdrometers during rainfall measurement campaigns,” Atmospheric Res., vol. 82, pp. 137–163, Nov. 2006, doi: 10.1016/j.atmosres.2005.09.007.

M. Löffler-Mang and J. Joss, “An Optical Disdrometer for Measuring Size and Velocity of Hydrometeors,” J. Atmospheric Ocean. Technol. - J ATMOS OCEAN TECHNOL, vol. 17, Feb. 2000, doi: 10.1175/1520-0426(2000)017<0130:AODFMS>2.0.CO;2.

E. Habib, W. Krajewski, and A. Kruger, “Sampling Errors of Tipping-Bucket Rain Gauge Measurements,” J. Hydrol. Eng., vol. 6, Apr. 2001, doi: 10.1061/(ASCE)1084-0699(2001)6:2(159).

G. Upton and A. R. Rahimi, “On-line detection of errors in tipping-bucket raingauges,” J. Hydrol., vol. 278, pp. 197–212, Jul. 2003, doi: 10.1016/S0022-1694(03)00142-2.

S. Ochoa-Rodriguez, L.-P. Wang, P. Willems, and C. Onof, “A Review of Radar‐Rain Gauge Data Merging Methods and Their Potential for Urban Hydrological Applications,” Water Resour. Res., vol. 55, Aug. 2019, doi: 10.1029/2018WR023332.

Radiometrics Corporation, “MP3000A Profiler Operator’s Manua.” Boulder, CO, USA, 2008.

K. Bhattacharyya and M. Islam, “SHORT TERM RAIN FORECASTING FROM RADIOMETRIC BRIGHTNESS TEMPERATURE DATA,” J. Mech. Contin. Math. Sci., vol. 15, Feb. 2020, doi: 10.26782/jmcms.2020.02.00007.

E. R. Westwater, “The accuracy of water vapor and cloud liquid determination by dual-frequency ground-based microwave radiometry,” Radio Sci., vol. 13, no. 4, pp. 677–685, Aug. 1978, doi: 10.1029/RS013i004p00677.

Y. Chen et al., “Improving the heavy rainfall forecasting using a weighted deep learning model,” Front. Environ. Sci., vol. 11, p. 1116672, Feb. 2023, doi: 10.3389/fenvs.2023.1116672.

L. C. P. Velasco, R. P. Serquiña, M. S. A. Abdul Zamad, B. F. Juanico, and J. C. Lomocso, “Week-ahead Rainfall Forecasting Using Multilayer Perceptron Neural Network” Procedia Comput. Sci., vol. 161, pp. 386–397, Nov. 2019, doi: 10.1016/j.procs.2019.11.137.

R. Levine, “Statistical Methods in the Atmospheric Sciences by Daniel S. Wilks,” J. Am. Stat. Assoc., vol. 95, pp. 344–345, Mar. 2000, doi: 10.2307/2669579.