Askı Maddesi Boyutunun Askı Maddesi Konsantrasyonu Üzerindeki Etkisinin Doğrusal Karma Regresyon Modeli ile Analizi

Öz

Açık kanal akımlarında askı maddesi miktarının artması hidrolik ve çevre mühendisliği açısından önemli bir problemdir. Bu artış, rezervuar kapasitelerinin azalmasına, su yapılarının zamanla aşınmasına, su kalitesinin giderek bozulmasına ve sucul yaşam alanlarının zarara uğramasına yol açabilir. Bu nedenle, askı maddesi konsantrasyonunun doğru bir şekilde tahmin edilmesi, etkili sediment yönetimi, erozyon kontrolü ve su kaynakları sistemlerinin sürdürülebilirliği için kritik öneme sahiptir. Bu çalışmada, birden fazla laboratuvar deneyinden derlenmiş bir veri kümesi kullanarak, dane çapının askı maddesi konsantrasyonu üzerindeki etkisi araştırılmıştır. Geleneksel doğrusal regresyon (LM) temel tahminler sağlarken, deneysel gruplardaki farklılıklardan kaynaklanan değişkenliği yakalamada başarısız olur. Bunu ele almak için, dane çapı için rastgele etkilere sahip doğrusal karma etkili modeller (LMM) kullanıldı ve diğer hidrolik ve geometrik parametreleri (örneğin, debi, genişlik, derinlik, sıcaklık, eğim) sabit etkiler olarak ele alındı. LMM, LM modeli için 0,24 olan koşullu R² değeriyle karşılaştırıldığında yaklaşık 0,74’lik bir değer elde etmiş ve önemli ölçüde daha düşük Akaike bilgi kriteri ve Bayesçi bilgi kriteri değerleri göstererek üstün bir model uyumu göstermiştir. Sonuçlar, debi, eğim, derinlik ve dane çapının askı maddesi konsantrasyonunu önemli ölçüde etkilediğini, ancak yatak genişliği ve sıcaklığın etkilemediğini ortaya koymaktadır. Bu bulgular, birden fazla deneysel kaynaktan gelen sediment konsantrasyonu veri kümelerini analiz ederken grup düzeyinde rastgele etkilerin dahil edilmesinin gerekliliğini vurgulamaktadır.

Anahtar Kelimeler

• askı maddesi
• karma etki modeli
• dane çapı
• hidrolik

Analysis of the Effect of Suspended Sediment Particle Size on Suspended Sediment Concentration Using Linear Mixed-Effects Regression Model

Abstract

The increase in suspended sediment concentration in open channel flows poses significant challenges in hydraulic and environmental engineering. Such increases can lead to reduced reservoir capacities, gradual abrasion of hydraulic structures, deterioration of water quality, and damage to aquatic habitats. Therefore, accurately predicting suspended sediment concentration is critical for effective sediment management, erosion control, and the sustainability of water resource systems. In this study, a dataset compiled from multiple laboratory experiments was utilized to investigate the effect of grain size on suspended sediment concentration. While traditional linear regression (LM) provided baseline estimates, it failed to account for the variability arising from differences among experimental groups. To address this, linear mixed-effects models (LMM) with random effects for grain size and fixed effects for other hydraulic and geometric parameters (e.g., discharge, width, depth, temperature, slope) were applied. The LMM achieved a conditional R2 of approximately 0.74, compared to 0.24 for the LM model, and exhibited substantially lower Akaike and Bayesian information criterion values, indicating superior model performance. The results revealed that discharge, slope, depth, and grain size significantly influence suspended sediment concentration, whereas channel width and temperature showed no significant effects. These findings highlight the necessity of incorporating group-level random effects when analyzing sediment concentration datasets derived from multiple experimental sources.

Keywords

• Suspended Sediment
• Mixed Model
• Particle Size
• Hydraulics

Kaynakça

1. [1] E. Shamaei and M. Kaedi, “Suspended sediment concentration estimation by stacking the genetic programming and neuro-fuzzy predictions,” Appl Soft Comput, vol. 45, pp. 187–196, Aug. 2016, doi: 10.1016/j.asoc.2016.03.009.
2. [2] J. Lepesqueur, R. Hostache, N. Martínez-Carreras, E. Montargès-Pelletier, and C. Hissler, “Sediment transport modelling in riverine environments: on the importance of grain-size distribution, sediment density, and suspended sediment concentrations at the upstream boundary,” Hydrol Earth Syst Sci, vol. 23, no. 9, pp. 3901–3915, Sep. 2019, doi: 10.5194/hess-23-3901-2019.
3. [3] M. Cieśla and R. Gruca-Rokosz, “Implications of suspended sediment in the migration of nutrients at the water-sediment interface in retention reservoirs,” Sci Rep, vol. 14, no. 1, p. 24924, Oct. 2024, doi: 10.1038/s41598-024-76556-x.
4. [4] J. Park and J. R. Hunt, “Coupling fine particle and bedload transport in gravel-bedded streams,” J Hydrol (Amst), vol. 552, pp. 532–543, Sep. 2017, doi: 10.1016/j.jhydrol.2017.07.023.
5. [5] J. Deng, B. Camenen, C. Legout, and G. Nord, “Estimation of fine sediment stocks in gravel bed rivers including the sand fraction,” Sedimentology, vol. 71, no. 1, pp. 152–172, Jan. 2024, doi: 10.1111/sed.13132.
6. [6] A. Kumar, S. Hossain, S. Sen, S. Mohan, and K. Ghoshal, “Grain-size distribution in suspension under non-equilibrium conditions,” International Journal of Sediment Research, vol. 39, no. 5, pp. 774–794, Oct. 2024, doi: 10.1016/j.ijsrc.2024.06.003.
7. [7] A. van Hamel, P. Molnar, J. Janzing, and M. I. Brunner, “Suspended sediment concentrations in Alpine rivers: from annual regimes to sub-daily extreme events,” Hydrol Earth Syst Sci, vol. 29, no. 13, pp. 2975–2995, Jul. 2025, doi: 10.5194/hess-29-2975-2025.
8. [8] Š. Zezulka, B. Maršálek, E. Maršálková, K. Odehnalová, M. Pavlíková, and A. Lamaczová, “Suspended Particles in Water and Energetically Sustainable Solutions of Their Removal—A Review,” Processes, vol. 12, no. 12, p. 2627, Nov. 2024, doi: 10.3390/pr12122627.

9. [9] C. T. Yang and J. B. Stall, “Unit Stream Power for Sediment Transport in Natural Rivers,” Illinois State Water Survey, University of Illinois Water Resources Center, Research Report No. 88, July 1974. [Online]. Available: https://www.researchgate.net/publication/237395916_Unit_Stream_Power_for_Sediment_Transport_in_Alluvial_Rivers
10. [10] A. Molinas and B. Wu, “Transport of sediment in large sand-bed rivers,” Journal of Hydraulic Research, vol. 39, no. 2, pp. 135–146, 2001, doi: 10.1080/00221680109499814.
11. [11] S.Q. Yang, “Sediment transport capacity in rivers,” Journal of Hydraulic Research, vol. 43, no. 2, pp. 131–138, Jan. 2005, doi: 10.1080/00221686.2005.9641229.
12. [12] E. Doğan, “Katı Madde Konsantrasyonunun Yapay Sinir Ağlarını Kullanarak Tahmin Edilmesi,” Teknik Dergi, vol. 20, no. 96, pp. 4567–4582, 2009.
13. [13] E. Dogan, “Suspended Sediment Load Estimation in Lower Sakarya River By Using Artificial Neural Networks, Fuzzy Logic and Neuro-Fuzzy Models,” 2005,
14. [14] G. Çeribaşı and E. Doğan, “Aşağı Sakarya Nehrindeki Askı Maddesi Miktarının Esnek Yöntemler ile Tahmini,” Karaelmas Fen ve Mühendislik Dergisi, vol. 6, no. 2, pp. 351–358, 2016
15. [15] Ö. Terzi and T. Baykal, “Dalgacık- GEP Modeli ile Akarsularda Askıda Katı Madde Miktarı Tahmini,” Süleyman Demirel Üniversitesi Fen Bilimleri Enstitüsü Dergisi, vol. 19, no. 2, pp. 39–45, 2015, doi: 10.19113/sdufbed.89396.
16. [16] A. Ülke, S. Özkul, and G. Tayfur, “Ampirik Yöntemlerle Gediz Nehri İçin Askıda Katı Madde Yükü Tahmini,” Teknik Dergi, vol. 22, no. 107, pp. 5387–5407, 2011,
17. [17] J. G. Ibrahim and G. Molenberghs, “Missing data methods in longitudinal studies: a review,” TEST, vol. 18, no. 1, pp. 1-43, May. 2009, doi.org/10.1007/s11749-009-0138-x
18. [18] A. J. Carnoli, P. oude Lohuis, Lutgarde M.C. Buydens, G. H. Tinnevelt, and J. J. Jansen, “Linear Mixed-Effects models in chemistry: A tutorial,” Analytica Chimica Acta, vol. 1304, pp. 342444–342444, Mar. 2024, doi: https://doi.org/10.1016/j.aca.2024.342444.
19. [19] S. Nakagawa and H. Schielzeth, “A General and Simple Method for Obtaining R2 from Generalized Linear Mixed-Effects Models,” Methods in Ecology and Evolution, vol. 4, no. 2, pp. 133–142, Dec. 2012, doi: https://doi.org/10.1111/j.2041-210x.2012.00261.x.
20. [20] T.-L. Liu, B. Flückiger, and K. de Hoogh, “A comparison of statistical and machine-learning approaches for spatiotemporal modeling of nitrogen dioxide across Switzerland,” Atmospheric Pollution Research, vol. 13, no. 12, p. 101611, Dec. 2022, doi: https://doi.org/10.1016/j.apr.2022.101611.
21. [21] S. Yang, G. Yang, B. Li, and R. Wan, “Water quality improves with increased spatially surface hydrological connectivity in plain river network areas,” Journal of Environmental Management, vol. 377, p. 124703, Mar. 2025, doi: https://doi.org/10.1016/j.jenvman.2025.124703.
22. [22] A. A. Gili, E. J. Noellemeyer, and M. Balzarini, “Hierarchical linear mixed models in multi-stage sampling soil studies,” Environmental and Ecological Statistics, vol. 20, no. 2, pp. 237–252, Aug. 2012, doi: https://doi.org/10.1007/s10651-012-0217-0.
23. [23] J. S. Lessels and T. F. A. Bishop, “Estimating water quality using linear mixed models with stream discharge and turbidity,” Journal of Hydrology, vol. 498, pp. 13–22, Aug. 2013, doi: https://doi.org/10.1016/j.jhydrol.2013.06.006.
24. [24] F. Belhaj et al., “Predicting precipitation and NDVI utilization of the multi-level linear mixed-effects model and the CA-markov simulation model,” Clim Serv, vol. 38, p. 100554, Apr. 2025, doi: 10.1016/j.cliser.2025.100554.
25. [25] E. Barca, D. De Benedetto, and A. M. Stellacci, “Contribution of EMI and GPR proximal sensing data in soil water content assessment by using linear mixed effects models and geostatistical approaches,” Geoderma, vol. 343, pp. 280–293, Jun. 2019, doi: 10.1016/j.geoderma.2019.01.030.
26. [26] K. RK. Dilrukshi et al., “Determining bad actors: A linear mixed effects model approach to elucidate behavioral toxicity of metal mixtures in drinking water,” Ecotoxicol Environ Saf, vol. 287, p. 117296, Nov. 2024, doi: 10.1016/j.ecoenv.2024.117296.
27. [27] Government of West Bengal, “Study on the Critical Tractive Force Various Grades of Sand,” Annual Report of the River Research Institute, West Bengal, Publication No. 26, Part I, 1965. [Online]. Available: https://www.aloki.hu/pdf/1704_98379863.pdf
28. [28] T. R. Daves, “Summary of experimental data for flume tests over fine sand,” Department of Civil Engineering, University of Southampton, 1971. [Online]. Available: https://conservancy.umn.edu/bitstreams/a279845d-26e8-4409-ab42-d3b7b17be30f/download
29. [29] P.- Y. Ho, “Abhängigkeit der Geschiebebewegung von der Kornform und der Temperatur,” Preussische Versuchsanstalt für Wasser-, Erd- und Schiffbau, Berlin, Mitt., Heft 37 (43 pp.), 1939. [Online]. Available: https://public.ucrlib.aspace.cdlib.org/repositories/5/archival_objects/607416
30. [30] F. T. Mavis, T.-Y. Liu, and E. Soucek, “The transportation of detritus by flowing water—II,” University of Iowa Studies in Engineering, Bulletin 11 (New Series No. 341), University of Iowa, Iowa City, IA, 1937. [Online]. Available: https://iro.uiowa.edu/view/pdfCoverPage?download=true&filePid=13811796880002771&instCode=01IOWA_INST
31. [31] W. R. Brownlie, “Compilation of alluvial channel data: Laboratory and field,” W. M. Keck Laboratory of Hydraulics and Water Resources, California Institute of Technology, Pasadena, CA, Report No. KH-R-43B, November 1981. [Online]. Available: https://authors.library.caltech.edu/records/deqcj-g7748/latest
32. [32] E. Doğan, “Akarsularda Taşınan Toplam Katı Madde Miktarının Yapay Zekâ Metotları ile Tahmin Edilmesi,” Doktora Tezi, Sakarya Üniversitesi Fen Bilimleri Enstitüsü, 2008.
33. [33] N. M. Laird and J. H. Ware, “Random-Effects Models for Longitudinal Data,” Biometrics, vol. 38, no. 4, p. 963, Dec. 1982, doi: 10.2307/2529876.
34. [34] J. Zhang, Y. Yang, and J. Ding, “Information criteria for model selection,” WIREs Computational Statistics, vol. 15, no. 5, Sep. 2023, doi: 10.1002/wics.1607.
35. [35] Q. Liu, M. A. Charleston, S. A. Richards, and B. R. Holland, “Performance of Akaike Information Criterion and Bayesian Information Criterion in Selecting Partition Models and Mixture Models,” Syst Biol, vol. 72, no. 1, pp. 92–105, May 2023, doi: 10.1093/sysbio/syac081.
36. [36] J. E. Nash and J. V. Sutcliffe, “River flow forecasting through conceptual models part I — A discussion of principles,” Journal of Hydrology, vol. 10, no. 3, pp. 282–290, Apr. 1970, doi: https://doi.org/10.1016/0022-1694(70)90255-6.
37. [37] H. V. Gupta, H. Kling, K. K. Yilmaz, and G. F. Martinez, “Decomposition of the mean squared error and NSE performance criteria: Implications for improving hydrological modelling,” Journal of Hydrology, vol. 377, no. 1–2, pp. 80–91, Oct. 2009, doi: https://doi.org/10.1016/j.jhydrol.2009.08.003.
38. [38] A. Gelman and J. Hill, Data Analysis Using Regression and Multilevel/Hierarchical Models. Cambridge: Cambridge University Press, 2006. doi: 10.1017/CBO9780511790942.
39. [39] D. J. Barr, R. Levy, C. Scheepers, and H. J. Tily, “Random effects structure for confirmatory hypothesis testing: Keep it maximal,” J Mem Lang, vol. 68, no. 3, pp. 255–278, Apr. 2013, doi: 10.1016/j.jml.2012.11.001.
40. [40] S. Nakagawa and H. Schielzeth, “Repeatability for Gaussian and non‐Gaussian data: a practical guide for biologists,” Biological Reviews, vol. 85, no. 4, pp. 935–956, Nov. 2010, doi: 10.1111/j.1469-185X.2010.00141.x.
41. [41] José C. Pinheiro and Douglas M. Bates, Mixed-Effects Models in S and S-PLUS. in Statistics and Computing. New York: Springer-Verlag, 2000. doi: 10.1007/b98882.
42. [42] C. R. Henderson, “Best Linear Unbiased Estimation and Prediction under a Selection Model,” Biometrics, vol. 31, no. 2, p. 423, Jun. 1975, doi: 10.2307/2529430.
43. [43] M. P. Lamb, W. E. Dietrich, and J. G. Venditti, “Is the critical Shields stress for incipient sediment motion dependent on channel-bed slope?,” Journal of Geophysical Research, vol. 113, no. F2, May 2008, doi: https://doi.org/10.1029/2007jf000831.
44. [44] K. Khosravi, Amir, L. Mao, J. F. Rodriguez, P. M. Saco, and A. D. Binns, “Experimental Analysis of Incipient Motion for Uniform and Graded Sediments,” Water, vol. 13, no. 13, pp. 1874–1874, Jul. 2021, doi: https://doi.org/10.3390/w13131874.