جستجوی مقالات مرتبط با کلیدواژه « yield prediction » در نشریات گروه « کشاورزی »

One of the common methods to estimate the yield of crops before their harvest is the use of satellite images and remote sensing technology. One of the efficient methods extracted from remote sensing data to monitor crop yield is the use of spectral indices. This study was carried out to estimate maize yield before its harvest using Landsat 8 satellite images in a part of the lands of Shahid Beheshti agro-industry, Dezful. Eight spectral indices including NDVI, TNDVI, GNDVI, SAVI, OSAVI, NDWI, MNDWI, and NDMI were used in this study. A correlation relationship was formed between the indices and yield with 70% of the data. Then the relationships were evaluated with 30% of the data. MNDWI index had the highest correlation with the field data in December and January, respectively, with a coefficient of determination of 30% and 13%. R2 and RMSE between the observed and predicted yield values by this index for the mentioned months were 72%, 1/71 tons per hectare, and 48%, 1/54 tons per hectare, respectively. Multivariable equations produced for both dates in January and December were obtained with R2 and RMSE, 31%, 0/06 tons per hectare, and 55%, 1/19 tons per hectare, respectively. According to the results, by using the average MNDWI ، SAVI, and NDVI indices in the produced relationships, the errors caused by the spatial changes of yield at the farm level are minimized, and it is possible to estimate the yield before harvest.

In recent years, the use of machine learning algorithms has been a promising way to improve crop yield predictions, especially when using non-linear relationships. This research was conducted with the aim of evaluating the yield estimation of irrigated and rainfed barley as well as the total yield of barley produced in the provincial centers of Iran using remote sensing data and machine learning methods including XGBoost and SVM. The results showed that by using climatic data, drought indices and plant indices of remote sensing and also XGBoost and SVM algorithms, it is possible to reliably estimate barley yield in different regions of the country with different climates. In general, the RMSE error obtained for both models was acceptable (0.41 and 0.77 t/ha). The R2 determination coefficient values for XGBoost and SVR algorithms in modeling rainfed barley cultivation performance were equal to 0.2 and 0.22 respectively, in irrigated barley performance were equal to 0.52 and 0.55 and for total barley were equal to 0.66 and 0.65, indicating that the rainfed yield modeling were not as suitable as irrigated and total barely yield modeling. The RBF kernel was chosen as the best kernel to use for the SVM algorithm. Also, in this research, while examining the effects of change in train and test data dividing, the parameters of precipitation, temperature, and evapotranspiration were determined as the most important parameters affecting the performance of the barley yield for both algorithms in different evaluated conditions.

Early prediction of wheat yield is major challenge in agrictultural management and food security. This requires regional estimation of yield which is costly and time consuming. Remote sensing (RS) is a fundamental approach for achieving practical and effective solutions for this problem. The existing differences in the spectral reflectance characteristic of agiven phenomena such as crop canopies, can serve as an identifier attribute . Wheat farms can have different yields based on their various growth and management conditions. These variations can affect the spectral reflectance. In this research, the yields data for year 2017 of 200 wheat farms located in Gonbad-Kavus and Voshmgir Dam regions, Golestan province north of Iran have been studied. After correcting the Landsat 8 satellite images of month of May, spectral bands data were extracted, and 15 spectral indices were calculated. The relation between spectral indices and wheat yield were worked out using multivariate linear regression and M5 regression tree model. According to the nonlinear relation between wheat yield and spectral reflectance, multivariate linear regression did not perform well. The correlation coefficient was about 63%, and the mean error was about 425 kg ha-1. The M5 regression tree model predicted wheat yield with an accuracy of 89 % and root mean square error (RMSE) lower than 325 kg ha-1. This accuracy was obtained using bands 1, 3, and NMDI and NDWI indices and band 4 to band 3 ratios (B43). Inaccurate yield data the non-uniformity and small area of the farms are the main sources of error.

Understanding the factors that limit prediction of regional performance of sugarcane crop and improve management practices is essential. The purpose of this study was to select a model that could estimate the potential of sugarcane production by considering the climatic, plant, and soil and terrain characteristics of the region. The study areas in the south of Khuzestan Province included the agro-industries of Amir Kabir, Miza Kuchak Khan, Dabal Khazaei, Salman Farsi, and Farabi. For this purpose, 100 sites in farms of different areas under sugarcane cultivation were studied based on diversity in soil characteristics. The method of this research was done in two hierarchies. In the first stage, estimation of sugarcane production potential using FAO growth model and, in the second stage, estimation of land production potential for sugarcane according to the effect of limiting factors in soil (as calculated using soil index with parametric method) that reduced yield in the first stage. To analyze the data, regression by standard methods, stepwise, and estimation curve were used. In standard and stepwise regression, soil properties were considered as an independent variable and the observed yield was considered as a dependent variable. In the linear, second and third degree estimation curve methods, the observed yield of sugarcane was considered as an independent variable and predicted yield was selected as dependent variable. The results of the first phase showed that the production potential of the FAO growth model was 95.8 tons/ha while the results of the second phase showed that the production potential of sugarcane in the region at different management levels was estimated from 18 to 69.3 tons/ha. Factors of yield reduction included limiting factors such as lime content, heavy and very heavy soil texture, drainage, salinity and sodicity, and lack of proper management. The results of standard and stepwise regression methods showed that the coefficient of determination was 0.52 and 0.49 and the standard error (ME) was 10.13 and 9.77 tons/ha, respectively. Soil properties could predict yield by standard method up to 52% and stepwise method up to 49%. In the estimation curve method for linear model, second degree, and third degree, the coefficients of determination were 0.74, 0.85, and 0.87, respectively, and standard errors were 7.8, 5.8 and 5.3 tons/ha. Therefore, the third degree estimation curve method, which uses the FAO growth model to predict crop yield, has higher accuracy and less error than the standard and stepwise regression model, which use only the effect of land characteristics on the observed yields.

Remote sensing and using satellite images have been widely considered due to the high speed of measurement and great area of coverage. Canola is a source of edible oil and its cultivation in Iran and the world is developing. Comparing with other crops, canola, because of its yellow flowers, has a different canopy color, and only a few researches have been carried out in order to assess the spectral indices for prediction of its yield. Therefore, the main objective of this research is to evaluate some spectral vegetation indices to estimate the yield of canola in different growth stages.

The study was performed in 2016-2017 in Karaj, Iran. Three canola farms were chosen for the evaluation of the relationship between yield and some vegetation indices derived from the Sentinel-2 sensor. The sensor data were processed in five stages: before flowering, early flowering, peak of flowering, green and dry maturity, and the vegetation indices were extracted for each of them. This research was pixel-based and the pixels network of each studied farm was determined by RTKGPS. During harvesting time, for measurement of grain yield, five samples, four from the corners and one from the center of the pixel, were taken and their average was considered as the representative amount of the pixel. Totally, 112 pixels from three studied farms were used to calibrate the predictive models. By using Simple Linear Regression (SLR) models, ten new and conventional vegetation indices were assessed. Also, Multivariate Linear Regression (MLR) models and Artificial Neural Net (ANN) models with four bands, three visible bands and NIR band, as inputs, were used to predict the canola yield. In order to validate the SLR and MLR models, the "K-Fold" method of cross-validation was used, and for the validation of ANN models, 15% of data were used; 70% for the train, 15% for validation, and 15% for the test.

Results and Discussion:

 The results showed that, on the basis of SLR models, among the growth stages, the highest coefficient of determination (R2) in each of the vegetation indices belonged to one of the two stages: the peak of flowering and green maturity. According to SLR models, among the vegetation indices in different stages, the NDYI in the peak of the flowering stage had the highest correlation with yield (R2 = 73%). Also, the RVI with 29%, BNDVI with 52%, NDVI with 56%, and GNDVI with 35% had the highest R2 in the before flowering, early flowering, peak of flowering, green and dry maturity stages, respectively. MLR models resulted to the best yield predictive model at the peak of flowering stage (R2 = 76% for the calibration and R2 = 73% and RMSE = 0.641 for the validation). For ANN models, the strongest model achieved at peak of flowering stage (R2 = 92% for the calibration (train) and R2 = 77% and RMSE = 0.612 for the validation (test)). It seems that the results are affected by yellow flowers of canola, and absorption of blue light by their petals. Therefore, in the peak of the flowering stage, the reflection of the blue light is more likely to belong to green leaves and stems. Therefore, any index such as NDYI, which the blue reflection is subtracted in its equation, represents better the number of flowers, and since the density of flowers is directly related to the yield, the yield will be predicted with more precision.

 Conclusions:

 The results of the analysis of the indices by SLR models showed that the correlation of each of the vegetation indices with the canola yield, in different stages of growth, has a considerable difference. Based on this model, the highest R2 in each of these indices happened in the peak of flowering or green maturity stage, and among these indices in different stages, the NDYI in the peak of the flowering stage had the highest R2. Finally, in both of the MLR and ANN models, with four bands, three visible bands and near-infrared band, as inputs, the best yield predictive model resulted in the peak of the flowering stage.

This study was conducted to predict alfalfa forage yield based on some climatic, soil and vegetative indices (PVI) derived from Sentinel-2 sensor in Sarayan (Southern Khorasan) of Iran in July 2018. Alfalfa yield data were collected from 52 points (10 of those experimental points) in two consecutive harvests to predict alfalfa performance, a stepwise multivariate linear regression was used. Results showed that the alfalfa performance map, in both consecutive harvests, with the mean of rainfall, PVI index and soil class was significant at 1% probability level. Validation showed that R2, RMSE and GMER were 0.82, 0.88 and 0.91, respectively, indicating the high compliance of the estimated performance model with the actual yield of alfalfa. Also, the results of chi-square test (P = 0.99) showed non-significant difference between actual values and estimated hay yield during two harvests. Therefore, due to the high reliability of terrestrial observations and climatic data in the region, these variables can be used to provide proper utilization pattern for forage plants.

Topography by affecting soil quality can significantly affect the production of agricultural products. This study was conducted in Toshan area of Golestan province to evaluate the spatial variation of wheat production and to create regression models between the crop, soil, and topography characteristics. Wheat and soil sampling from different slope positions were collected randomly from 100 points. Soil physical-chemical analyses and wheat yield parameter measurements were carried out. The digital elevation model was prepared, and topographic attributes were calculated. The results showed that the highest total yield (14.53 t/ha) and grain yield (4.41 t/ha) were observed in <10% slope class, which showed a significant difference compared to steeper slopes. The highest (15.82 t/ha) and the lowest (5.68 t/ha) total mean yields were corresponded to paw slope and shoulder slope, respectively. The highest grain yield was obtained from the foot slope and paw slope. The amounts of mean grain yield were 4.61 t/ha in foot slope and 4.66 t/ha in paw slope. Curvature and wetness index had a remarkable relationship with wheat yield. According to the regression analyses, topography indexes have been able to explain the spatial variability of yield, indicating the importance of these factors' impact on water distribution in yield production process in the studied area. Increasing yield productions in low slope positions in this study was probably due to increasing in soil depth and plant available water, as well as accumulation of organic matter and nutrient elements such as nitrogen and potassium due to deposition of eroded materials in these positions.

In order to improve food security as one of the most crucial needs of the society, prediction and estimation of wheat yield should be considered by decision makers in the country. On the other hand, agricultural production is always at risk of climate and international markets changes, however, this risk is never completely eliminated, but we can estimate the yield before the harvest season to minimize them. Nowadays, one of the methods of yield estimating is using satellite- imagery. Remote sensing data allow estimating crop yield based on vegetation indices. The present study aimed to find a fast way with acceptable accuracy for predicting wheat yield in the field in Shahrekord County, Chaharmahal and Bakhtiari province, Iran by using Landsat 8 data.

In order to fulfill the goal three sets of Landsat 8 imagery data dated on June 4th, 20th and July 6th 2016 were downloaded from http://earthexplorer.usgs.gov. The acquisition dates were corresponded with milk, dough and ripening stages of wheat growth cycle. Concurrently three wheat cultivated farms were selected in the Shahrekord County, based upon surface area, homogeneity in wheat cultivated farms and satisfaction of landlords with sampling. Coordinates of samples were recorded using GPS device (Garmin etrex). In the whole, the yield of 60 plots (0.25 m2) from selected fields were used for yield prediction. The density of plants per plots was considered and the yield in each plot was estimated. Nine introduced vegetation indices by literature were considered and the correlation coefficients between indices' valued and estimated yields were calculated. The models were evaluated using the coefficient of determination (R2) and standard error of estimation (SEE), reduced variance (RV) and mean estimation error (MEE). Image processing and statistical analysis were carried out using ILWIS 3.3 and Sigma Plot 10.0 software, respectively.

The results showed that among the imagery data, the highest correlation coefficients existed between wheat yield and vegetation indices developed by images dated on 20th June 2016 corresponded to wheat dough stage. The coefficients of determinations (R2) of models with vegetation indices NDVI, NRVI, OSAVI, RVI, SAVI, RDVI, DVI, EVI and GNDVI were 0.86, 0.86, 0.86, 0.86, 0.86, 0.83, 0.81, 0.80, 0.78, respectively. The results indicated that the most appropriate models were Polynomial and quadratic. The results also showed the potential of satellite images for yield prediction at pre-harvest stage with an accuracy above 80 percents.

According to this study appropriate utilizing of satellite images and field observation at dough stage aided yield estimation of wheat in semi-arid regions. The most appropriate indices for yield estimation were NDVI, NRVI and OSAVI. Among the statistical models, polynomial and quadratic models with the highest coefficients of determination were introduced as the best models. Based upon the results Landsat 8 data and field samples could be used for wheat yield forecasting which help decision makers managing the market for this strategic crop. The results should be tested in other similar climates.

  Potential production condition is defined as situation where water and nutrients are supplied in ample with no damage from weeds, pests and diseases. Under this situation potential yield (Yp) which is defined by solar radiation, temperature, CO2 concentration and genotypic characteristics of the crop species could be achieved. Closing the gap between Yp and actual yield (Ya) harvested by farmers is considered as the main challenge of agronomic sciences all around the world. Crop simulation models provide a powerful tool for prediction of Yp at regional and national scales. However, the accuracy of these models is highly dependent on the quality of input data which are usually not fully available even in developed countries. Simplified models designed based on few simple equations may be considered as an alternative where accurate data sets are lacking. In this research, Yp of wheat and sugar beet were estimated using some simple methods and the results were compared with those of complex simulation models under the same climatic and management conditions in Khorasan province, north east of Iran. Potential yields of wheat and sugar beet were estimated by three simple calculation methods including FAO method (FAO, 1981), modified FAO, FAO-M (Versteeg and van Keulen, 1986) and RUE-based method (Nonhebel, 1997) at three different regions (Torbat, Mashhad and Neishabor) in Khorasan province. In these methods total above ground dry matter (DM) is calculated using two equations and Yp is estimated as the product of DM and harvest index with minimum weather data requirements. In addition, Yp for both crops and at the same locations was predicted using LINTUL and SUCROS simulation models which were previously calibrated for yield estimation of wheat and sugar beet at regional level. The accuracy of predicted potential yields by five methods was compared statistically against the measured yields obtained from the field experiments and high yielding farms at the studied regions. Mean observed yield of wheat over the three studied regions was 7.18 t ha-1 and the estimated potential yield in the same regions using two simulation models and three simplified models was ranged between 6.92 and 7.63 t ha-1. Prediction error for the simulation models was 1.39 and for simple methods less than 5% (4.64%). Relative root mean squared error (RMSEn) for the simple methods was higher than that of complex models. However, it was between 7.11 and 10.16 % of the mean measured wheat yield which could be statistically considered as reasonable. Calculated values of modeling efficiency (ME) were positive and higher than 0.60 except for FAO-M method (ME=0.48). Measured sugar beet yield averaged over regions was 82.5 t ha-1 and estimated potential yield by different methods was between 89 and 91 t ha-1. Simple calculation methods had lower accuracy in predicting sugar beet yield compared to simulation models but RMSEnof simple methods never exceeded 11.5% showing statistically acceptable prediction. Cross validation indicated significant correlation between the results of three simple methods and that of complex simulation models. Results of this study showed that potential yields of different crops can be estimated accurately using simple calculation methods with minimum weather data requirements. Such methods may be used as an alternative for agroecological zoning and yield gap analysis at regional level where calibrated simulation models and complete daily weather data sets are not available

  Predicting the yield of strategic crops such as wheat and barley can provide liquidity to purchase products from farmers, suitable space for storage, sufficient pricing, and estimating the import and export needs of the agricultural products. Crop yield forecasting can be estimated using different methods. Artificial neural network is one of the methods recently has been considered and has high accuracy in crop yield estimation. In this approach the relationship between independent and dependent variables as well as their complex interactions will be studied and hidden correlatioins between the variables is discovered. Crop yield, particularly in case of rain-fed crops, severely affected by climatic factors such as rainfall, humidity, temperature fluctuations and solar radiation during the growth season. On this basis, it seems that in case of relationships between climatic data (as independent variables) and crop yield (as dependent variable) it would be possible to predict agricultural production. Considering the importance of forecasting agricultural production, an experiment was conducted in Kermanshah province, with the aim of estimating rainfed and irrigated barley yield using artificial neural network approach,. Barley yield data of 25 years (1991 to 2015) as well as raw and corresponding standardized meteorological data (total annual rainfall, average annual temperature, humidity annual average total sunshine hours, average annual evaporation and the number of frost days) were used as input data networks. To find the best network, performance of different types of neural networks was tested to barley yield evaluation. To evaluate models the statistics indices of correlation coefficient (R), coefficient of determination (R2), mean squared error (MSE) and root mean square error (RMSE) were used. The results showed that the best neural network built for rainfed barley was Modular Neural Network with Momentum learning law. The network had a correlation coefficient of 0.96 and 0.92, and coefficient of determination of 0.85 and 0.92 for raw and standard data, respectively. The best neural network built for irrigated barley was also Modular Neural Network with Momentum learning law. The correlation coefficient for the raw and standard input data was 0.78 and 0.72 with coefficient of determination of 0.60 and 0.51, respectively. The results showed the less efficiency of artificial Neural Network in predicting irrigated compared to rainfed barley yield. Comparison of MSE and RMSE between the models also revealed that networks related to rainfed barely with more accuracy had more correlation coefficient compared to irrigated barely. It seems that the managing operations such as watering in irrigated barley has reduced the effect of climate factors on barley yield. So that the sensitivity of irrigated barley yield was much less than the rainfed barley. In fact, in irrigated cultivations, depending on the amount of available water, crops were irrigated regularly. This would be in addition to supplying the needed water of crops, also reduce the impact of other climatic factors such as temperature. Generally based on the results of this study, the accuracy of Artificial Neural Network to predict the yield of rainfed barley was more acceptable than irrigated barley in Kermanshah province. Rainfed cultivation are more affected by climatic factors such as rainfall and temperature, which would be the reason of the achieved results. Hence, the accuracy of neural network for rainfed barley was more than irrigated barely, which represents more relationship between yield of rainfed barley with climatic factors as inputs of the model. So the sensitivity test of the yield to climatic factors revealed more sensitivity in rainfed than irrigated barley. Moreover, the accuracy of neural network showed that neural network was built for barley than wheat.

Considering the importance of sunflower as one of the most important plants in the production of edible oils، present study was developed in order to determine the best nonlinear regression function which can quantify growth of diameter of sunflower head to time. At the present study in order to fit the best regression model explaining relationship between increasing of sunflower head diameter of Lakomka and Progress varieties and time، an experiment was performed at 60 kilometers of Amlash city in rainfed condition. At first، the head diameter measurement since the beginning of growth maturity and harvest was recorded. Then different regression models including exponential، power، logarithmic، logistic، schnute and gompertz were used. Then to determine the most efficient model، different parameters of evaluation of model fitting were used. The results revealed that schnute and logistic models are the best model for explaining of the head diameter variation of Lakomka and Progress varieties to time. Schnute model with highest adjusted coefficient of determination (0. 99 and 0. 98) and lowest root error mean of squares was determined as the best model in explaining of growth pattern of head diameter to time. It is expected that using these models could be used for prediction of head diameter of Lakomka and Progress varieties with high precision.

Remote sensing is one of the techniques recently used for the purpose of monitoring and predicting crop yield. The objective of this study was predicting wheat yield in Mashhad during 2009 and 2010 by Landsat (TM) and (ETM+) images with ground based data. For this purpose, three images dated 6, 14 and 22 Jun 2009 and 8 and 24 May and 9 Jun 2010 were analysed and vegetation indices were computed. Field data included wheat growth stage, yield and geographical position. Results showed that the best relation between yield and indices were obtained in 14 Jun 2009 and 14 May 2010. These dates coincided with milky and maturity stages of wheat. Cumulative indices estimated yield prediction more precisely than others indices. Red band had the highest correlation (r2=0.76) with yield among the other band. Through the other indices, PD311 and PD312 had the best correlation with the yield. Correlation between NDVI and yield was not significant. The best correlation with the yield was obtained by applying the multi linear regression method (r2=0.83).

Crop yield, soil properties and erosion are strongly affected by terrain parameters. Therefore, knowledge about the effects of terrain parameters on strategic crops such as wheat production will help us with sustainable management of landscape. This study was conducted in 900ha, of Ardal district, Charmahal and Bakhtiari Province to develop regression models on wheat yield components vs. terrain parameters. Wheat yield and its components were measured in 100 points. Points were distributed randomly in stratified geomorphic surfaces. Yield components were measured by harvesting of 1 m2 plots. Terrain parameters were calculated by a 3×3 m spacing from digital elevation model. The result of descriptive statistics showed that all variables followed a normal distribution. The highest and lowest coefficient of variance (CV) was related to grain yield (0.36) and thousand seeds weight (0.13), respectively. Multiple regression models were established between yield components and terrain parameters attributes. The predictive models were validated using validation data set (20% of all data). The regression analysis revealed that wetness index and curvature were the most important attributes which explained about 45-78% of total yield components variability within the study area. The overall results indicated that topographic attributes may control a significant variability of rain-fed wheat yield. The result of validation analysis confirmed the above-stated conclusion with low RMSE and ME measures.