A.e. Bonyad, T. Hajyghaderi,
Volume 11, Issue 42 (winter 2008)
Abstract
The natural forest and range stands of Zanjan province are located in mountainous areas. Inventorying and mapping of natural forest and range stands in mountainous areas are difficult and costly. Satellite data are suitable for this purpose. The Landsat ETM+ image data of 2002 are used for classification and mapping of natural forest stands in Zanjan province. For the purpose of data reduction and principal components extraction, the principal components analysis (PCA) was used. Just the scores of the first three PCs (PCA1، PCA2 and PCA3 (that accounted for 76.67 percent of the total variance were considered as new images for future analysis. A raster geographic information system (RGIS) database file was prepared and involved 7 ETM+ bands, 3 principle component analysis, 9 factor analysis and 8 vegetation indexes of image data. The correlation coefficients of 27 image layers and optimum index factors (OIF) of selected images were computed and 12 groups were found suitable for natural forest and range stands. Maximum liklelihood classification (MLC) method was used in this study. In order to test the accuracy of map, kappa index of agreement was calculated. The highest KIP belonged to three λ3, λ4, λ5 Landsat image bands with KIP = 0.86. The highest OIF belonged to three PCA3, FA2 and MIR with value of 233.44 and lower OIF belonged to three λ4, λ5, λ7 with value of 83.63. The overall, user’s and producer’s accuracy rates were 88.45, 73.69 and 70.23 percent respectively. The results of the study show that the Landsat ETM+ image data were appropriate for classification and mapping of natural forest and range stands in Zanjan province.
K. Ghaderi, B. Motamedvaziri, M. Vafakhah, A.a. Dehghani,
Volume 25, Issue 4 (Winiter 2022)
Abstract
Proper flood discharge forecasting is significant for the design of hydraulic structures, reducing the risk of failure, and minimizing downstream environmental damage. The objective of this study was to investigate the application of machine learning methods in Regional Flood Frequency Analysis (RFFA). To achieve this goal, 18 physiographic, climatic, lithological, and land use parameters were considered for the upstream basins of the hydrometric stations located in Karkheh and Karun watersheds (46 stations with a statistical length of 21 years). The best Probability Distribution Function (pdf) was then determined using the Kolmogorov-Smirnov test at each station to estimate the flood discharge with a return period of 50-year using maximum likelihood methods and L-moments. Finally, RFFA was performed using a decision tree, Bayesian network, and artificial neural network. The results showed that the log Pearson type 3 distribution in the maximum likelihood method and the generalized normal distribution in the L moment method are the best possible regional pdfs. Based on the gamma test, the parameters of the perimeter, basin length, shape factor, and mainstream length were selected as the best input structure. The results of regional flood frequency analysis showed that the Bayesian model with the L moment method (R2 = 0.7) has the best estimate compared to other methods. Decision tree and artificial neural network were in the following ranks.
