DOI:
https://doi.org/10.14483/23448350.22614Published:
08/30/2024Issue:
Vol. 50 No. 2 (2024): May-August 2024Section:
Research ArticlesAplicación de métodos de aprendizaje profundo para la imputación de niveles de concentración de clorofila-a en la Costa Pacífica colombiana
Applying Deep Learning Methods for the Imputation of Chlorophyll-a Concentration Levels in the Colombian Pacific Coast
Keywords:
Chlorophyll-a, aprendizaje profundo, cobertura nubosa, MODIS, predicción de valores perdidos, sector pesquero, series temporales (es).Keywords:
chlorophyll-a, cloud cover, deep learning, fishing sector, missing value prediction, MODIS, time series (en).Downloads
Abstract (es)
El sector pesquero en Colombia, que aporta el 0.3 % del Producto Interno Bruto (PIB) y genera exportaciones
por USD 45.1 millones (equivalente al 3.3 % del PIB agropecuario), enfrenta desafíos significativos debido a la
falta de precisión en la medición de la clorofila-a, un indicador crucial de la salud de los ecosistemas marinos.
El uso de imágenes satelitales, particularmente aquellas obtenidas por el sensor MODIS, es esencial para obtener
datos precisos. Sin embargo, la alta cobertura nubosa, común en la geografía colombiana, afecta la calidad y
la disponibilidad de estas imágenes durante gran parte del año, creando lagunas en los datos críticos para la
evaluación del estado de los ecosistemas marinos. Este trabajo propone un algoritmo de aprendizaje profundo
basado en series temporales para la predicción de valores perdidos de clorofila-a. La metodología presentada
supera las limitaciones impuestas por la cobertura nubosa, alcanzando una precisión R2 superior a 0.8 en uno de
los modelos. En este contexto específico, la implementación y la evaluación de diversos modelos de aprendizaje
profundo han demostrado ser alternativas efectivas para proporcionar una evaluación más precisa y continua de
las áreas pesqueras. Esto ofrece información valiosa para mejorar la gestión y sostenibilidad del sector pesquero
en Colombia al añadir un componente temporal a la predicción de valores de clorofila-a. Esto, mediante datos de
hasta tres meses previos a la característica objetivo.
Abstract (en)
The fishing sector in Colombia, which contributes 0.3% of the Gross Domestic Product (GDP) and generates USD 45.1 million in exports (equivalent to 3.3% of the agricultural GDP), faces significant challenges due to the lack of precision in measuring chlorophyll-a, a crucial indicator of marine ecosystem health. The use of satellite images, particularly those obtained by the MODIS sensor, is essential for obtaining accurate data. However, the high cloud cover, which is common in Colombian geography, affects the quality and availability of these images for much of the year, creating gaps in critical data for assessing the state of marine ecosystems. This work proposes a deep learning algorithm based on time series for predicting missing chlorophyll-a values. The presented methodology overcomes the limitations imposed by cloud cover, achieving an R2 accuracy above 0.8 in one of the models. In this specific context, the implementation and evaluation of various deep learning models have proven to be effective alternatives in providing a more accurate and continuous assessment of fishing areas. This offers valuable information to improve the management and sustainability of the fishing sector in Colombia by adding a temporal component to the prediction of chlorophyll-a values, using data from up to three months prior to the target feature.
References
Abbas, M. M., Melesse, A. M., Scinto, L. J., Rehage, J. S. (2019). Satellite estimation of chlorophyll-a using moderate resolution imaging spectroradiometer (MODIS) sensor in shallow coastal water bodies: Validation and improvement. Water, 11(8), 1621. https://doi.org/10.3390/W11081621
Barazzetti, L., Scaioni, M., Gianinetto, M. (2014). Automatic co-registration of satellite time series via least squares adjustment. European Journal of Remote Sensing, 47(1), 55-74. https://doi.org/10.5721/EuJRS20144705
Baudena, A., Ser-Giacomi, E., D’Onofrio, D., Capet, X., Cotté, C., Cherel, Y., D’Ovidio, F. (2021). Finescale structures as spots of increased fish concentration in the open ocean. Scientific Reports, 11(1), 1-13. https://doi.org/10.1038/s41598-021-94368-1
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5-32. https://doi.org/10.1023/A:1010933404324/METRICS
Chen, S., Deng, L., Zhao, J. (2024). Enhanced reconstruction of satellite-derived monthly chlorophyll a concentration with fourier transform convolutional-LSTM. IEEE Transactions on Geoscience and Remote Sensing, 62, 1-14. https://doi.org/10.1109/TGRS.2024.3394399
Cho, H., Park, H. (2019). Merged-LSTM and multistep prediction of daily chlorophyll-a concentration for algal bloom forecast. IOP Conference Series: Earth and Environmental Science, 351(1), 012020. https://doi.org/10.1088/1755-1315/351/1/012020
Chuvieco Salinero, E. (2002). Teledection ambiental: la observación de la tierra desde el espacio. Ariel.
Dash, A., Jetley, S., Rege, A., Chopra, S., Sawant, R. (2022). Drought prediction and water quality estimation using satellite images and machine learning [Artículo de conferencia]. 7th International Conference on Communication and Electronics Systems, ICCES 2022, India. https://doi.org/10.1109/ICCES54183.2022.9835727
Du, Z., Wu, S., Kwan, M. P., Zhang, C., Zhang, F., Liu, R. (2018). A spatiotemporal regression-kriging model for spacetime interpolation: A case study of chlorophyll-a prediction in the coastal areas of Zhejiang, China. International Journal of Geographical Information Science, 32(10), 1927-1947. https://doi.org/10.1080/13658816.2018.1471607
Essien, P., Figueiredo, C. A. O. B., Takahashi, H., Klutse, N. A. B., Wrasse, C. M., Afonso, J. M. de S., Quispe, D. P., Lomotey, S. O., Ayorinde, T. T., Sobral, J. H. A., Eghan, M. J., Sackey, S. S., Barros, D., Bilibio, A. V., Nkrumah, F., Quagraine, K. A. (2022). Intertropical convergence zone as the possible source mechanism for southward propagating medium-scale traveling ionospheric disturbances over South American low-latitude and equatorial region. Atmosphere, 13(11), e1836. https://doi.org/10.3390/ATMOS13111836
Eze, E., Kirby, S., Attridge, J., Ajmal, T. (2021). Time series chlorophyll-a concentration data analysis: A novel forecasting model for aquaculture industry. Engineering Proceedings, 5(1), e27. https://doi.org/10.3390/ENGPROC2021005027
Hänninen, J., Mäkinen, K., Nordhausen, K., Laaksonlaita, J., Loisa, O., Virta, J. (2022). The “Seili-index” for the prediction of chlorophyll-α levels in the Archipelago Sea of the northern Baltic Sea, southwest Finland. Environmental Modeling & Assessment, 27, 571-584. https://doi.org/10.1007/s10666-022-09822-9
Hu, H., Fu, X., Li, H., Wang, F., Duan, W., Zhang, L., Liu, M. (2023). Prediction of lake chlorophyll concentration using the BP neural network and Sentinel-2 images based on time features. Water Science and Technology, 87(3), 539-554. https://doi.org/10.2166/WST.2023.019
Li, X., Sha, J., Wang, Z. L. (2018). Application of feature selection and regression models for chlorophyll-a prediction in a shallow lake. Environmental Science and Pollution Research, 25(20), 19488-19498. https://doi.org/10.1007/S11356-018-2147-3/METRICS
Luis-xmart/framework (s.f.). https://gitfront.io/r/Luis-xmart/3nDgRRysh6j3/framework-chl-temp/
Ma, L., Liu, Y., Zhang, X., Ye, Y., Yin, G., Johnson, B. A. (2019). Deep learning in remote sensing applications: A meta-analysis and review. ISPRS Journal of Photogrammetry and Remote Sensing, 152, 166-177. https://doi.org/10.1016/J.ISPRSJPRS.2019.04.015
Nascimento Silva, H. A., Rosato, A., Altilio, R., Panella, M. (2018). Water quality prediction based on wavelet neural networks and remote sensing [Artículo de conferencia]. International Joint Conference on Neural Networks, Rio de Janeiro, Brasil. https://doi.org/10.1109/IJCNN.2018.8489662
Papenfus, M., Schaeffer, B., Pollard, A. I., Loftin, K. (2020). Exploring the potential value of satellite remote sensing to monitor chlorophyll-a for US lakes and reservoirs. Environmental Monitoring and Assessment, 192(12), 1-22. https://doi.org/10.1007/S10661-020-08631-5
Park, Y., Cho, K. H., Park, J., Cha, S. M., Kim, J. H. (2015). Development of early-warning protocol for predicting chlorophyll-a concentration using machine learning models in freshwater and estuarine reservoirs, Korea. Science of the Total Environment, 502, 31-41. https://doi.org/10.1016/J.SCITOTENV.2014.09.005
Poddar, S., Chacko, N., Swain, D. (2019). Estimation of chlorophyll-a in northern coastal bay of Bengal using Landsat-8 OLI and Sentinel-2 MSI sensors. Frontiers in Marine Science, 6, e00598. https://doi.org/10.3389/FMARS.2019.00598
Salomonson, V. V., Barnes, W., Masuoka, E. J. (2006). Introduction to MODIS and an overview of associated activities. Earth Science Satellite Remote Sensing: Science and Instruments, 1, 12-32. https://doi.org/10.1007/978-3-540-37293-6_2
Shin, Y., Kim, T., Hong, S., Lee, S., Lee, E., Hong, S. W., Lee, C. S., Kim, T. Y., Park, M. S., Park, J., Heo, T. Y. (2020). Prediction of chlorophyll-a concentrations in the Nakdong River using machine learning methods. Water, 12(6), 1822. https://doi.org/10.3390/W12061822
Silveira Kupssinskü, L., Thomassim Guimarães, T., Menezes de Souza, E., C. Zanotta, D., Roberto Veronez, M., Gonzaga, L., Mauad, F. F. (2020). A method for chlorophyll-a and suspended solids prediction through remote sensing and machine learning. Sensors, 20(7), e2125. https://doi.org/10.3390/s20072125
Stramska, M., Jakacki, J. (2024). Variability of chlorophyll a concentration in surface waters of the open Baltic Sea. Oceanologia, 66(2), 365-380. https://doi.org/10.1016/J.OCEANO.2024.02.003
Wang, G., Hao, J., Ma, J., Jiang, H. (2011). A comparative assessment of ensemble learning for credit scoring. Expert Systems with Applications, 38(1), 223-230. https://doi.org/10.1016/J.ESWA.2010.06.048
Wang, H., Yan, X., Chen, H., Chen, C., Guo, M. (2015). Chlorophyll-a predicting model based on dynamic neural network. Applied Artificial Intelligence, 29(10), 962-978. https://doi.org/10.1080/08839514.2015.1097142
Wilson, A. M., Jetz, W. (2016). Remotely sensed high-resolution global cloud dynamics for predicting ecosystem and biodiversity distributions. PLOS Biology, 14(3), e1002415. https://doi.org/10.1371/JOURNAL.PBIO.1002415
Yajima, H., Derot, J. (2018). Application of the Random Forest model for chlorophyll-a forecasts in fresh and brackish water bodies in Japan, using multivariate long-term databases. Journal of Hydroinformatics, 20(1), 191-205. https://doi.org/10.2166/HYDRO.2017.010
Yu, X., Shen, J., Zheng, G., Du, J. (2022). Chlorophyll-a in Chesapeake Bay based on VIIRS satellite data: Spatiotemporal variability and prediction with machine learning. Ocean Modelling, 180, e102119. https://doi.org/10.1016/j.ocemod.2022.102119
How to Cite
APA
ACM
ACS
ABNT
Chicago
Harvard
IEEE
MLA
Turabian
Vancouver
Download Citation
License
Copyright (c) 2024 Luis-Miguel Martínez-Vargas, Julián-Fernando Muñoz-Ordóñez, Yady-Tatiana Solano-Correa
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
When submitting their article to the Scientific Journal, the author(s) certifies that their manuscript has not been, nor will it be, presented or published in any other scientific journal.
Within the editorial policies established for the Scientific Journal, costs are not established at any stage of the editorial process, the submission of articles, the editing, publication and subsequent downloading of the contents is free of charge, since the journal is a non-profit academic publication. profit.
