Yuksel, KivancGulci, NeseAkay, Abdullah EminGulci, Sercan2026-02-082026-02-0820251001-62792589-7284https://doi.org/10.1016/j.ijsrc.2025.01.011https://hdl.handle.net/20.500.12885/5705In this study, the effectiveness of different stabilization techniques implemented on the forest road cut slopes was investigated in terms of controlling erosion and runoff. Wood production residues, hydro-seeding, and jute geotextile treatments were applied on study plots located on the example road. The amount of erosion and runoff were measured on the study plots which were established for different slope grades of 20 degrees, 30 degrees, and 40 degrees. Then, the amount of erosion and runoff measured from the plots were compared to determine the performance of stabilization techniques on the cut slope. In the solution process, an Artificial Neural Network (ANN) model, which is one of the machine learning algorithms, was used to predict sediment yield from forest road cut slopes. The sediment yields averaged over the three slope grades from highest to lowest were measured as 6.41, 1.16, 0.65, and 0.45 g/m2 in the control plot with no treatment, jute geotextile, hydroseeding, and wood production residues, respectively. The averaged over the three runoff amounts slope grades from the highest to the lowest were determined as 6.82, 3.71, 1.64, and 1.30 mm/m2 in control the plot, jute geotextile, hydroseeding, and wood production residues, respectively. Comparing to the control plot, wood production residues, hydroseeding, and jute geotextile treatments reduced the sediment yields by 14,10, and 5 times, respectively. On the other hand, wood production residues, hydroseeding, and jute geotextile applications reduced the runoff amount by 5, 4, and 2 times, respectively. As a result, it was found that wood production residues and hydroseeding treatment can be more efficient in reducing the amount of runoff and sediment yield compared to the jute geotextile treatment. The ANN method achieved high accuracy in predicting sediment yield and it was concluded that the ANN can be used as an effective method to evaluate soil slope stabilization techniques. (c) 2025 International Research and Training Centre on Erosion and Sedimentation. Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).eninfo:eu-repo/semantics/openAccessForest roadsSoil slope stabilizationSediment yieldRunoffCut slopeArtificial Neural Network (ANN)Article10.1016/j.ijsrc.2025.01.011403476488WOS:0015150246000012-s2.0-85219580902Q2Q1