جستجوی مقالات مرتبط با کلیدواژه « Regression Model » در نشریات گروه « آبخیزداری، بیابان، محیط زیست، مرتع »

The correct estimation of flood flow in rivers is an important issue and plays a significant role in the optimal use of water resources, operation of dam reservoirs, and the design and planning of water projects.

In this research, a simple and conceptual method based on Manning's formula in real flow conditions is used to estimate the flood flow discharge. In this method, firstly, for the combined effect of friction slope and Manning's roughness coefficient, the alpha parameter (α) was defined and calculated for 12 hydrometric stations located in three main rivers of Golestan province (including Gorganrood, Atrak, and Qarasoo).

The results showed that the value of this parameter decreases continuously with the increase of the flow depth and finally asymptotically reaches a constant value. This behavior shows that the value of α is nearly constant for the upper flow depths which indicate the occurrence of floods, and hence using this constant value and the Manning formula, the river flood discharge can be estimated. In the next step, we tried to provide a regression model between the Alpha parameter and the flow depth. The regression modeling results showed that for most of the hydrometric stations, the coefficients of determination (R2) of the presented equations are smaller than 0.3 which demonstrates its low efficiency. For this reason, machine learning models were used and the parameter was modeled by the Artificial Neural Networks (ANN), Decision Tree (M5tree), and Support Vector Regression (SVR) models.

The modeling results showed that the decision tree model with a mean absolute error of 0.35, determination coefficient of 0.88, and root mean square error of 0.86 has the best accuracy in the test phase. After determining the parameter α, the amount of flood discharge was predicted. The best performance among the models was the decision tree in predicting the flow rate in rivers. After comparing the observed values, the decision tree model has an average absolute error of 1.32, a determination coefficient of 0.89, and an average square root error of 63. 3. It has the best accuracy in the test phase.

In recent years, the increases in population and expansion of settlements in hazardous areas have greatly increased the impact of natural disasters in industrialized and developing countries. Landslide risk zoning helps to identify strategic points and geographically prone critical areas. Therefore, measures for rapid, safe mitigation and strategic planning for the future are important. In fact, landslide risk assessment may be a suitable and cost-effective help for land use planning, so in this regard, the aim of the current research is to model the risk of landslides using non-parametric data mining in watersheds of Hyrcanian forests.

For this purpose, the map of landslide risk according to the Mora and vahrson method with the effect of factors affecting the occurrence of landslides including topographical factors, hydrological and climatic factors, geological factors, land cover factors, human factors, hydrographic network from digital elevation model, geological map It was prepared using the map of the Geological Organization of the country, and to obtain the elevation index and the relative height, first, using the curve of height levels taken from the 1:25000 topographic maps of the region, the map of the elevation classes was prepared. After that, the study area was divided into one-square-kilometer grids, and maps with the lowest and highest elevations in one-square-kilometer grids were obtained; in the last step, by subtracting these two maps, a map was obtained whose information shows the value of the postal index and relative height. Monthly rainfall was also used to obtain the soil moisture index. Finally, algorithms of three random forest models, an artificial neural network, and a decision tree algorithm were used to model the risk of landslides in the STATISTICA 12.0 software environment.

According to the results, the highest distribution of landslide areas belongs to the low-risk class (76%). TP, SL, SR, and TS variables were considered the essential factors of landslide occurrence based on their importance. The results of validation with three algorithms of RF, CART, and ANN showed; According to the coefficient of explanation obtained in modeling the risk of landslides, the artificial neural network model with (R2=0.99) is more accurate than other methods.

The results of the present study showed that data mining methods have a high capability in predicting the risk of landslides. Therefore, the use of the mentioned methods can be considered in reducing the risks associated with landslides and planning for land use.

The present study was conducted to investigate the differences between several remote sensing indices, four different plot sizes and two different sampling methods to estimate the percentage of plant cover and production in three plant communities in 2013. Ground sampling was performed in three communities with different dominant vegetation in two forms of six and three quadrats. Four different dimensions of the quadrats were used nested to estimate production and coverage percentage. Sampling was performed in each community within 30 pixels along three transects with different heights (sampling method and testing were performed according to the researcher opinion). Dominant plant density was measured by counting the bases per plot, plant cover percentage as an estimate and production in the form of double sampling in relation to cover percentage. The results showed that with increasing plot area, the degree of correlation of plant indexes of Landsat image and their significance in relation to production and percentage of plant cover will increase. But this increase in community 2 is more dramatic in most respects with predominantly shrubs. As in this community, most of the studied indicators in 3 × 3 plot have a correlation and a reliable model, and in the three sizes of 1×1, 2×1 and 2×2 plots, the resulting models are not valid enough and have high RMSE. In community one with dominant broadleaf plants, only the model obtained in 1×1 plot and in community two with dominant plants, the model obtained in 1×1 and 1×2 plots are not statistically valid, although sometimes they have a significant correlation. The results of two different models are statistically different in communites 1 and 3 and there was no significant difference between them in community 2. According to the results of the study of indicators in terms of correlation and model validity, the two indicators NDVI and CTVI can be used in community 1, NDVI and TSAVI1 in community 2 and NDVI, NRVI and TSAVI1 in community 3 recommended for estimating production and percentage of plant cover using satellite images.

In this study, land use change trend in coastal areas of south of Iran, including Sistan Baluchestan, Hormozgan, Bushehr were studied during 31 years (1988-2019). Landsat satellite images including TM, ETM +, OLI were used for this purpose. The maximum likelihood method was also used to classify satellite images. Vegetation indices (SAVI) and salinity index (SI1) were used to influence land use classes on land degradation and desertification. The results of land use change trend showed that Changes in land use classes of man-made areas and saline lands as the most dynamic land use classes followed an increasing trend of 2.72% and 1.14%, respectively. The results of multivariate regression analysis showed that three vegetation, saline and man-made variables, and four vegetation, saline, man-made and sandy areas explain 22.3% and 37.8% of the dependent variables of vegetation and salinity indices, respectively. Therefore, in more than 60% of changes, other environmental variables are affected. Hence, with respect to the results of land use area and regression analysis and the mean trend of SAVI and SI1 changes, where vegetation and SAVI classes show decreasing trend and saline land and SI1 classes have increasing trend, it can be concluded that land use changes is phenomena that changes ecosystem services with almost irreversible impacts. This will have nothing to do with disturbing the ecosystem balance and intensifying land degradation in coastal areas.

According to the importance of loss areas, which have embarrassed a wide region of Golestan province and its important role in formation of the sediment of Gorgan river, more accurate recognition of types of loss erosion seems essential in order to providing necessary function for erosion and sediment problems in them, specially gully erosion. So, in this study we are going to providing regression model through investigation of effective earth environmental factors on gully erosion to identify gully initiation mechanism on loss land. For this, locational situation of the gully (Temer Ghareh Ghozi region) was identified after recognizing airy photos in the scale of 1:20000 and basic study as climate, geology, soil, herbal coverage, land function, and topography began after providing unit topography by using geographical information system. In order to control the effect of herbal coverage on initiation of gully erosion, we work to remove density of herbal coverage by 1*1 square meter plots. Finally, a proper regression model was offered to develop gully erosion, after recognition and computation the relationship between environmental earth factors with initiation, growth and development of gully erosion through multi-variable analysis by using SPSS. The result shows that earth environmental factors such as height, slope, and aspect, height of hillside, annual rainfall, stone units, land type and density of herbal coverage have an effective role in initiation and expansion of gully erosion in loss land.

Suspended sediment (SS) flux in a river is an important parameter for the watershed management. The large extend of watersheds areas and limited of sediment stations measurement have been caused that different methods have been developed to SS estimation. In this study, 42 sediment measurement stations existed in south and southeast of the Caspian Sea with over 20 year period was chosen. The sediment rating curve (SRC) was drawled using discharge and SS load measurements for 42 stations, then daily SS load was estimated by using the daily discharge (Q) and the average of annual SS was computed. By using the gamma test (GT), 14 effective variables on SS was reduced to 5 main variables and the mentioned stations were recognized in four homogeneous groups by using Andrew c urves (AC). For each homogeneous group and all existed stations using five selected main variables was developed regression models for average of annual SS load. Relative error (RE), r elative bias (RBIAS) and r oot relative mean squared error (RRMSE) indices statistics are used for evaluating the accuracy of the models. The results indicated that a model was not provided for group 1. The models of g roups 2 to 4 have a good accuracy and the model of g roup 4 with RE(29%), RBIAS(12%) and RRMSE(35%) in the calibration stage, and RE(19%), RBIAS(12%) and RRMSE(25%) in the validation stage has the best performance and One model for all stations has high error. Peak discharge with two return periods (Qp) has the most impact on the average of annual SS change.

Artificial neural network is a new method to estimate phenomena changes that have wide application in various branches of science. Reservoir sediment traps efficiency and sediment volume is an issue that can be investigated with this method. The purpose of this study is the determinate of sediment trapping efficiency in delayed dams and a comparison between neural network methods with regression models. So in this study, was used the physical model of delayed dams in the soil conservation and watershed management research center. In order to simulate a flood hydrograph, was used the idea of a linear reservoir model. Then with the release of flood and sediment trap sediments, examined the function of delayed dam. Then, with identify the affecting parameters on reservoirs sediment trapping; was developed neural network model on based back propagation of error. Also was applied the regression models for investigation the relationship between the parameters and compare the estimation results with observation. Finally, was applied the statistical indexes R2, RMSE and MAPE to assess the accuracy and precision of the model. According to results, the average values of R2, RMSE and MAPE in regression models are equal to 0.456, 26.6 and 62.1 respectively; but are the values in the neural network model 0.982, 4.6 and 6.1. So the artificial neural network has more ability in compared to regression equation to predict trap sediment in delayed dams. Also the results showed estimation of sediment trapping efficiency has depend on the number of parameters in equation and must be determined the optimized equation on based number of parameters.

Bromus kopetdaghensis is one of perennial plants of Poaceae family and it is important in biodiversity of arid and semi arid grasslands of Iran. This experiment was conducted as a completely randomized design with 4 replications in laboratory of faculty of Agriculture، Ferdowsi University of Mashhad، Iran. Treatments were different constant temperature of 5، 10، 15، 20، 25، 30، 35، and 40°C with 4 replications and in each replication 25 seeds were used. The results showed significant effects on different germination features of Bromus kopetdaghensis. The lowest germination rate (with 2. 26 seed per day) was obtained at 5°C and the highest (with 11. 39 seed per day) at 30°C. The maximum and minimum length of caulicle was obtained at 35°C (with 9. 41cm) and 10°C (with 2. 48cm)، respectively. Maximum length of radical was also observed at 35°C (with 9. 62cm) and minimum length of radical was observed at 10 and 20°C، without significant different with 3. 63 and 3. 48 cm. Based on the regression analysis between rate of germination and temperature the cardinal temperature of maximum، optimum، and minimum were obtained in the range of 0. 71-4. 25، 24-30. 36 and 41. 01-45. 48، respectively.

In this study the efficiency of statistical regression models and, also, the relationships between rainfall characteristics with runoff and sediment yield were investigated in the plot scale. For this purpose, data from runoff and sediment yield of 12 storms during October 2010 to October 2011 were used in the plot scale, which, installed in representative basin of Khamsan located at south of Kurdistan province. Hence, after each storm-runoff event, amounts of runoff and sediment were collected and measured after transporting to the laboratory. The storms data were also acquired from meteorological station located in the basin. Results showed that, sediment yield of plots have the most correlation with runoff volume, maximum intensities of storm in 60, 55, 65, 25 and 20 minute with correlation coefficient of (0.887, 0.821, 0.818, 0.814, 0.813 and 0.810). in the other hand, the runoff production of plot has the most correlation with rainfall height and 3th & 4th quartile of rainfall in significant level of 99 percent. It has been also revealed that Logarithmic model with the highest determination coefficient (0.825), efficiency coefficient (0.84) and the least value of relative error (11.57) and Root mean square (0.058) is the best model for predicting plot’s runoff production. It is Followed by power and inverse models with values of efficiency coefficient (0.8 & 0.794), determination coefficient (0.769 & 0.768) and relative error of verification (4.73 & 3.84). Results, therefore, demonstrated that multi-variation models is better to explain sediment yields of storm followed by S curve, cubic, quadratic, exponential, power and linear models.  

Soil water retention curve (SWRC) is an important hydraulic property in the studies of water and solute movement in soil and its direct measurement is time consuming and expensiveness. Therefore many indirect procedures like pedotransfer function and empirical models have proposed by soil scientists. In this study، a model based on fractal theory، was used to estimate water retention curve. For estimating Fractal dimension (D) that used in this model، it was tried to find out a simple relation between this parameter and feasible soil properties such as، soil particle size distribution (PSD) fractal dimension، clay، silt and sand content، ρb، dg، by applying stepwise regression analysis. The SWRC was measured for 40 soil samples، which included all soil texture classes، by pressure plate. the measured DSWRC for 33 soil samples، used for regression analysis and 7 soil samples was used for model validation. The regression analysis showed a linear relationship between DSWRC and DPSD and dg with goodness of fit، R2 = 0. 95. Estimation SWRC for 7 soil sample whit replacement estimated D by linear model، showed that the model had a good estimation، especially for light texture soils. Also there is an overestimate that can be modified by model calibration.

Some contaminants associate with soil particles and, thus, their transport and fate in the environment is determined by the soil erosion processes. Eutrophication, low oxygen levels and high nutrient (nitrogen and phosphorus) concentrations in reservoirs, canals and other water courses, is a common water pollution feature. Phosphorus (P) is one of the major plant nutrients and also the major nutrients controlling eutrophication of surface water. P load pulsed by heavy rainfall may damage the ecological quality of downstream. The present study was conducted in Kojour Watershed located in Mazandaran province.The results showed that P loss varied from 0.26 gr lit-1 to 1860.64 gr lit-1. It also proved that P concentration could not estimated by flow discharge, while sediment concentration can estimate loss of P with determination coefficient and estimation error of 94% and 23%, respectively. The results could facilitate the application of given methods obtained in the present study to other ungauged watersheds with similar conditions and leading to the suitable soil and water management and planning.

Complication of initiation process, expansion of gully erosion and unfamiliarity of the function of soil physicochemical characteristics, effective in this type of erosion, is one of the most important challenges to achieve functional approaches in order to prevent it. According to the importance of loss areas in Golestan Province which is embarrassed a wide area and has an important role in production of sediment in Gorganrood river, exact configuration of types of erosion in loss is seemed essential for offering a proper function to control the erosion and sediment, especially gully erosion. So in this research, in order to identify initiation mechanism and gully erosion in loss areas, we are going to recognize airy photos with the scale of 1:20000 and the soil physicochemical characteristics in Temer Ghareh Ghozi region. Through direct revision, 10 gullies were identified, topography and merfometry, after recognizing the gully situation on activity units map in GIS environment. Then the sample from the soil of each gully, in the scale of 25, 50 and 75% of the length of the gully were done, in order to specify the soil physicochemical characteristics. At the end, by using SPSS software, the relationship between the soil physicochemical agents and development and expansion of gully erosion were offered through regression model. The outcome of analysis and corrosion of the data demonstrate the effective role of the soil physicochemical characteristics, specially the scale of silt, dissolvable and the soil saturation degree in initiation and expansion of gully erosion in loss areas.

Direct biomass measurements such as clipping method is difficult, destructive and time consuming. This research was performed to study possibility of Sphaerocoma aucheri forage production using some vegetative traits (such as cover characteristics including its area mean diameter, small diameter, large diameter and height of the plants). Dry forage production and some independent variables were randomly collected, from 50 plants of S. aucheri in a 4 years enclosure sites in Hormozgan province. After eliminating the outlier data and conducting normal test, multi variables regression in SPSS package was used to analyses data. The results showed that 77% changes in dry biomass of the species can satisfactory be estimated using canopy cover data The coefficient of cover area and constant index in regression model were 3.192 and 6.283 respectively. Validated model revealed that Sphaerocoma aucheri dry forage can be estimated properly measuring canopy cover of the species.

One of the effective factors in soil erosion is geological formations of the drainage basinfrom the view points of their erodibility and the amount of runoff and sediment yield. Inthis investigation a part of Segzi- Kuhpayeh subdrainage, located in Zayandeh-RoodDrainage Basin with 67091.7 ha area was chosen for studying erodibility and sedimentyield of Quaternary Formations. For doing this, the map of Quaternary Formations wasprepared. Then in each unit a field rainfall simulator was used and runoff and sedimentwere collected from the plots and erodibility and sediment yield factors were obtained.Soil samples were taken adjacent to each micro plot for laboratory tests and the datawere analyzed statistically. The results of regression analyses have shown thaterodibility and sediment yield are significantly different in various QuaternaryFormations. The most important physical and chemical parameters in sediment yield indifferent formations are Silt, Sand, very Fine Sand, Wetness, Caco3 and in erodibility ofdifferent formations are Silt+ very Fine Sand, Caco3, Silt, Clay, very Fine Sand and EC.