The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY (한국해양학회지:바다)

The Korean Society of Oceanography (한국해양학회)
권호 표지
DOI QR코드

DOI QR Code

Estimation of the Spring and Summer Net Community Production in the Ulleung Basin using Machine Learning Methods

기계학습법을 이용한 동해 울릉분지의 봄과 여름 순군집생산 추정

  • DOSHIK HAHM (Department of Oceanography and Marine Research Institute, Pusan National University) ;
  • INHEE LEE (Department of Oceanography, Pusan National University) ;
  • MINKI CHOO (Department of Oceanography, Pusan National University)
  • 함도식 (부산대학교 해양학과 및 해양연구소) ;
  • 이인희 (부산대학교 해양학과) ;
  • 추민기 (부산대학교 해양학과)
  • Received : 2023.11.23
  • Accepted : 2024.01.06
  • Published : 2024.02.29
Download PDF
⟨ Previous Next ⟩

Abstract

The southwestern part of the East Sea is known to have a high primary productivity compared to those in the northern and eastern parts, which is attributed to nutrients supplies either by Tsushima Warm Current or by coastal upwelling. However, research on the biological pump in this area is limited. We developed machine learning models to estimate net community production (NCP), a measure of biological pump, with high spatial and time scales of 4 km and 8 days, respectively. The models were fed with the input parameters of sea surface temperature, chlorophyll-a, mixed layer depths, and photosynthetically active radiation and trained with observed NCP derived from high resolution measurements of surface O2/Ar. The root mean square error between the predicted values by the best performing machine model and the observed NCP was 6 mmol O2 m-2 d-1, corresponding to 15% of the average of observed NCP. The NCP in the central part of the Ulleung Basin was highest in March at 49 mmol O2 m-2 d-1 and lowest in June and July at 18 mmol O2 m-2 d-1. These seasonal variations were similar to the vertical nitrate flux based on the 3He gas exchange rate and to the particulate organic carbon flux estimated by the 234Th disequilibrium method. To expand this method, which produces NCP estimate for spring and summer, to autumn and winter, it is necessary to devise a way to correct bias in NCP by the entrainment of subsurface waters during the seasons.

동해 남서부해역은 대마난류나 연안 용승에 의한 영양염 공급 등으로 동해 북부나 동부에 비해 일차생산력이 높은 것으로 알려져 있지만, 이 해역의 생물 펌프에 관한 연구는 제한적이다. 본 연구에서는 O2/Ar 측정으로 산출한 고해상도 순군집생산 현장 관측 결과와 기계학습 모형을 결합하여 시공간 해상도가 8일 간격, 4 km인 봄과 여름 순군집생산 시계열 자료를 추정하였다. 기계 모형의 예측과 실측의 평균 제곱근 오차는 6 mmol O2 m-2 d-1로 관측값 평균의 15%에 해당했다. 울릉분지 중앙부의 순군집생산은 3월에 49 mmol O2 m-2 d-1로 가장 높았고, 6월과 7월에 18 mmol O2 m-2 d-1로 가장 낮았다. 이 같은 계절 변화는 3He 기체교환율로 추정한 질산염 공급률이나 234Th 비평형법으로 추정한 입자유기탄소 방출률과 유사하였다. 봄과 여름의 순군집생산 추정으로 한정된 이 연구방법을 가을과 겨울로 확대하기 위해서는 아표층수의 표층 혼입에 따른 O2/Ar 순군집생산의 오차를 보정하는 연구가 필요하다.

Keywords

Acknowledgement

이 과제는 부산대학교 기본연구지원사업(2년)에 의하여 연구되었음.

References

  1. Anderson, L., A. Urence and J.L. Sarmiento, 1994. Redfield ratios of remineralization determined by nutrient data analysis. Global Biogeochemical Cycles, 8(1): 65-80. https://doi.org/10.1029/93GB03318
  2. Behrenfeld, M.J. and P.G. Falkowski, 1997. Photosynthetic rates derived from satellite-based chlorophyll concentration, Limnology And Oceanography, 42(1): 1-20. https://doi.org/10.4319/lo.1997.42.1.0001
  3. Breiman, L., 2001. Random Forests. Machine Learning, 45(1): 5-32. https://doi.org/10.1023/A:1010933404324
  4. Cassar, N., B.A. Barnett, M.L. Bender, J. Kaiser, R.C. Hamme and B. Tilbrook, 2009. Continuous High-Frequency Dissolved O2/Ar Measurements by Equilibrator Inlet Mass Spectrometry. Analytical Chemistry, 81: 1855-1864. https://doi.org/10.1021/ac802300u
  5. Chang, C.H., N.C. Johnson and N. Cassar, 2014. Neural network-based estimates of Southern Ocean net community production from in-situ O2 / Ar and satellite observation: a methodological study. Biogeosciences, 11: 3279-3297. https://doi.org/10.5194/bg-11-3279-2014
  6. Chen, T. and C. Guestrin, 2016. Xgboost: A scalable tree boosting system. Proceedings of the 22nd Acm Sigkdd International Conference on Knowledge Discovery and Data Mining, 785-794.
  7. Craig, H. and T. Hayward, 1987. Oxygen supersaturation in the ocean: Biological versus physical contributions. Science, 235(4785), 199-202. https://doi.org/10.1126/science.235.4785.199
  8. Duarte, C.M., A. Regaudie-de Gioux, J.M. Arrieta, A. Delgado-Huertas, and S. Agust ́i, 2013. The Oligotrophic Ocean is Heterotrophic. Annual Review Of Marine Science, 5(1): 551-569. https://doi.org/10.1146/annurev-marine-121211-172337
  9. European Centre for Medium-Range Weather Forecasts (ECMWF), 2023. Available at: https://www.ecmwf.int/en/forecasts/dataset/ecmwf-reanalysis-v5. [last accessed: October 10, 2023].
  10. Ferron, S., D.A. Valle, K.M. del, Bjorkman, P.D. Quay, M.J. Church and D.M. Karl, 2016. Application of membrane inlet mass spectrometry to measure aquatic gross primary production by the 18O in vitro method. Limnology and Oceanography: Methods, 14(9): 610-622. DOI: https://doi.org/10.1002/lom3.10116.
  11. Geurts, P., D. Ernst and L. Wehenkel, 2006. Extremely randomized trees. Machine Learning, 63(1): 3-42. DOI: https://doi.org/10.1007/s10994-006-6226-1.
  12. Gregor, L., S. Kok and P. Monteiro, 2017. Empirical methods for the estimation of Southern Ocean CO2: support vector and random forest regression. Biogeosciences, 14.
  13. Gregor, L., T.J. Ryan-Keogh, S.-A. Nicholson, M. Plessis, I. du, Giddy and S. Swart, 2019. GliderTools: A Python Toolbox for Processing Underwater Glider Data. Frontiers In Marine Science, 6.
  14. Hahm, D. and I. Lee, 2018. Estimation of Net Community Production Based on O2/Ar Measurements. Journal of The Korean Society of Oceanography, 23(1): 49-62.
  15. Hahm, D. and K.-R. Kim, 2001. An estimation of the new production in the southern East Sea using helium isotopes. Journal Of The Korean Society Of Oceanography, 36(1): 19-26.
  16. Hahm, D., S. Park, S.-H. Choi, D.-J. Kang, T. Rho and T. Lee, 2019b. Estimation of surface fCO2 in the southwest east sea using machine learning techniques. The Sea, 24(3): 375-388. https://doi.org/10.7850/JKSO.2019.24.3.375
  17. Hahm, D., T.S. Rhee, H.-C. Kim, C.J. Jang, Y.S. Kim and J.-H. Park, 2019a. An observation of primary production enhanced by coastal upwelling in the southwest east/japan sea. Journal of Marine Systems, 195: 30-37. https://doi.org/10.1016/j.jmarsys.2019.03.005
  18. Hamme, R. and S. Emerson, 2006. Constraining bubble dynamics and mixing with dissolved gases: Implications for productivity measurements by oxygen mass balance. Journal of Marine Research, 64(1): 73.
  19. Haskell, W.Z., M.G. Prokopenko, R.H.R. Stanley and A.N. Knapp, 2016. Estimates of vertical turbulent mixing used to determine a vertical gradient in net and gross oxygen production in the oligotrophic South Pacific Gyre. Geophysical Research Letters, 43(14): 7590-7599. https://doi.org/10.1002/2016GL069523
  20. Hyun, J., D. Kim, C. Shin, J. Noh, E. Yang, J. Mok, S. Kim, H. Kim and S. Yoo, 2009. Enhanced phytoplankton and bacterioplankton production coupled to coastal upwelling and an anticyclonic eddy in the Ulleung basin, East Sea. Aquatic Microbial Ecology, 54: 45-54. https://doi.org/10.3354/ame01280
  21. Joo, H., D. Lee, S.H. Son and S.H. Lee, 2018. Annual new production of phytoplankton estimated from MODIS-derived nitrate concentration in the East/Japan Sea. Remote Sensing, 10(5), DOI: 10.3390/rs10050806.
  22. Joo, H.T., S.H. Son, J.W. Park, J.J. Kang, J.-Y. Jeong, C.I. Lee, C.-K. Kang and S.H. Lee, 2016. Long-term pattern of primary productivity in the East/Japan sea based on ocean color data derived from MODIS-Aqua. Remote Sensing, 8(1): 25.
  23. Kaiser, J., M.K. Reuer, B. Barnett and M.L. Bender, 2005. Marine productivity estimates from continuous O2/Ar ratio measurements by membrane inlet mass spectrometry. Geophysical Research Letters, 32(19).
  24. Kim, D., E.J. Yang, K.H. Kim, C.W. Shin, J. Park, S. Yoo and J.H. Hyun, 2012. Impact of an anticyclonic eddy on the summer nutrient and chlorophyll a distributions in the Ulleung Basin, East Sea (Japan Sea). Ices Journal of Marine Science, 69(1): 23-29. https://doi.org/10.1093/icesjms/fsr178
  25. Kim, D., M.-S. Choi, H.-Y. Oh, Y.-H. Song, J.-H. Noh and K.H. Kim, 2011. Seasonal export fluxes of particulate organic carbon from 234Th/238U disequilibrium measurements in the ulleung Basin1 (tsushima basin) of the east Sea1 (sea of japan). Journal of Oceanography, 67(5): 577-588. DOI: https://doi.org/10.1007/s10872-011-0058-8.
  26. Kim, S.-K., K.-I. Chang, B. Kim and Y.-K. Cho, 2013. Contribution of ocean current to the increase in N abundance in the Northwestern Pacific marginal seas. Geophysical Research Letters, 40(1): 143-148. https://doi.org/10.1029/2012GL054545
  27. Kwak, J.H., J. Hwang, E.J. Choy, H.J. Park, D.-J. Kang, T. Lee, K.-I. Chang, K.-R. Kim and C.-K. Kang, 2013a. High primary productivity and f-ratio in summer in the Ulleung basin of the East/Japan Sea. Deep Sea Research I, 79: 74-85. https://doi.org/10.1016/j.dsr.2013.05.011
  28. Kwak, J.H., S.H. Lee, H.J. Park, E.J. Choy, H.D. Jeong, K.R. Kim and C.K. Kang, 2013b. Monthly measured primary and new productivities in the Ulleung Basin as a biological "hot spot" in the East/Japan Sea. Biogeosciences, 10(7): 4405-4417. https://doi.org/10.5194/bg-10-4405-2013
  29. Le, T.T., W. Fu and J.H. Moore, 2020. Scaling tree-based automated machine learning to biomedical big data with a feature set selector. Bioinformatics, 36(1): 250-256. DOI: https://doi.org/10.1093/bioinformatics/btz470.
  30. Lee, I., D. Hahm, D. Shin, C.-S. Hong, S. Nam, G. Kim and T. Lee, 2021. Determination and uncertainty of spring net community production estimated from O2/Ar measurements in the northern East China Sea and southern Yellow Sea. Continental Shelf Research, 230: 104570.
  31. Lee, S. and S. Yoo, 2016. Interannual variability of the phytoplankton community by the changes in vertical mixing and atmospheric deposition in the ulleung basin, east sea: A modelling study. Ecological Modelling, 322: 31-47. DOI: https://doi. org/https://doi.org/10.1016/j.ecolmodel.2015.11.012.
  32. Li, Z. and N. Cassar, 2016. Satellite estimates of net community production based on O 2/Ar observations and comparison to other estimates. Global Biogeochemical Cycles, 30(5): 735-752. https://doi.org/10.1002/2015GB005314
  33. Lipschultz, F., N.R. Bates, C.A. Carlson and D.A. Hansell, 2002. New production in the Sargasso Sea: History and current status. Global Biogeochemical Cycles, 16(1): 1-1-1-17, DOI: https://doi.org/10.1029/2000GB001319.
  34. Nakaoka, S., M. Telszewski, Y. Nojiri, S. Yasunaka, C. Miyazaki, H. Mukai and N. Usui, 2013. Estimating temporal and spatial variation of ocean surface pCO2 in the North Pacific using a self-organizing map neural network technique. Biogeosciences, 10(9): 6093-6106. https://doi.org/10.5194/bg-10-6093-2013
  35. NASA, 2023. Available at: https://oceancolor.gsfc.nasa.gov/l3/. [last accessed: October 10, 2023].
  36. Ocean Productivity, 2023. Available at: https://sites.science.oregonstate.edu/ocean.productivity/index.php. [last accessed: October 10, 2023].
  37. Onitsuka, G., I. Uno, T. Yanagi and J.-H. Yoon, 2009. Modeling the effects of atmospheric nitrogen input on biological production in the japan sea. Journal of Oceanography, 65(3): 433-438. DOI: https://doi.org/10.1007/s10872-009-0038-4.
  38. Onitsuka, G., T. Yanagi and J.-H. Yoon, 2007. A numerical study on nutrient sources in the surface layer of the Japan Sea using a coupled physical-ecosystem model. Journal Of Geophysical Research-Oceans, 112(C5): C05042.
  39. Park, K., D. Hahm, J.O. Choi, S. Xu, H.-C. Kim and S. Lee, 2019. Spatiotemporal variation in summer net community production in the amundsen sea polynya: A self-organizing map analysis approach. Continental Shelf Research, 184: 21-29. https://doi.org/10.1016/j.csr.2019.07.001
  40. Scikit Learn, 2023. Available at: https://scikit-learn.org. [last accessed: October 10, 2023].
  41. Stanley, R.H.R., W.J. Jenkins, S.C. Doney and D.E. Lott III, 2015. The 3 He flux gauge in the Sargasso Sea: a determination of physical nutrient fluxes to the euphotic zone at the Bermuda Atlantic Time-series Site. Biogeosciences, 12(17): 5199-5210. https://doi.org/10.5194/bg-12-5199-2015
  42. Teeter, L., R.C. Hamme, D. Ianson and L. Bianucci, 2018. Accurate estimation of net community production from O2/ar measurements. Global Biogeochemical Cycles, 32(8): 1163-1181. https://doi.org/10.1029/2017GB005874
  43. Volk, T. and M.I. Hoffert, 1985. Ocean Carbon Pumps: Analysis of Relative Strengths and Efficiencies in Ocean-Driven Atmospheric CO2 Changes, in The Carbon Cycle and Atmospheric CO2: Natural Variations Archean to Present, pp. 99-110, American Geophysical Union, Washington, D.C.
  44. Wanninkhof, R. 2014. Relationship between wind speed and gas exchange over the ocean revisited. Limnol. Oceanogr.: Methods, 12(6): 351-362. https://doi.org/10.4319/lom.2014.12.351
  45. XGBoost, 2023. Available at: https://xgboost.readthedocs.io. [last accessed: October 10, 2023].
  46. Yamada, K., J. Ishizaka and H. Nagata, 2005. Spatial and Temporal Variability of Satellite Primary Production in the Japan Sea from 1998 to 2002. Journal Of Oceanography, 61(5): 857-869. https://doi.org/10.1007/s10872-006-0005-2
  47. Yang, S.R., 1997. Primary production in the ocean-waste disposal area in: Development of monitoring technology for the wastes disposal sea areas. 2nd KORDI report. Korea Ocean Research; Development Institute.
  48. Yang, S.R., 1998. Primary production in the ocean-waste disposal area in: Development of monitoring technology for the wastes disposal sea areas. 3rd KORDI report,. Korea Ocean Research; Development Institute.
  49. Yoo, S. and J. Park, 2009. Why is the southwest the most productive region of the East Sea/Sea of Japan? Journal Of Marine Systems, 78(2): 15-15. https://doi.org/10.1016/j.jmarsys.2009.02.014
  50. Zeng, J., Y. Nojiri, P. Landschutzer, M. Telszewski and S. Nakaoka, 2014. A Global Surface Ocean fCO2 Climatology Based on a Feed-Forward Neural Network. Journal Of Atmospheric And Oceanic Technology, 31(8): 1838-1849. https://doi.org/10.1175/JTECH-D-13-00137.1