An efficient estimation of crop performance in sheep fescue (Festuca ovina L.) using artificial neural network and regression models.

Festuca ovina L. (sheep fescue), a perennial grass plant found in mountainous regions, is important from both an ecological and economic viewpoint. However, the variability of biological yield of sheep fescue due to its reliance on different characteristics makes it difficult to accurately prediction using classic modeling techniques. In this study, machine learning methods and multiple regression models (linear and non-linear) are used to investigate the interdependence of various morphological and physiological characteristics on accurate prediction of the biological yield (BY) of sheep fescue. Principal components analysis and stepwise regression were used to select six agronomic parameters i.e. thousand seed weight (TSW), relative water content (RWC), canopy cover (CC), leaf area index, number of florescence, and viability (VA), while the output variable was BY. To optimized the artificial neural network (ANN) structure, different transfer functions and training algorithms, different number of neurons in each layer, different number of hidden layers and training iteration were tested. The accuracy of the models and algorithms is analyzed by root mean square error (RMSE), mean absolute error (MAE), and determination coefficient (R2). According to the findings, ANN models were more accurate than regression models. The ANN model with two hidden layers (i.e. structure of 6-4-8-1) which had RMSE, MAE and R2 scores of 0.087, 0.065 and 0.96, respectively, was discovered as the best model for predicting the BY. In addition, result of the sensitivity analysis showed TSW, RWC and CC, in that order, were the variables most important for high-quality BY estimation in both models regardless of input combination. Finally, the paper concludes that early flowering sheep fescue genotypes with long maturation and great TSW must be regarded as the most suitable model for increasing BY in breeding projects.

Modeling potential habitats and predicting habitat connectivity for Leucanthemum vulgare Lam. in northwestern rangelands of Iran.

Invasive plants can alter the function and structure of ecosystems resulting in social, economic, and ecological damage. Effective methods to reduce the dominance of invasive plants are needed. The present study was aimed at modeling the invasive species Leucanthemum vulgare Lam. in the rangelands of the Namin region in northwest Iran, as well as predicting the habitat connectivity of this species to detect areas with high habitat connectivity. Modeling of potential habitats was performed using logistic regression (LR) and maximum entropy (MaxEnt); the ensemble map which resulted from these was used to predict habitat connectivity using the electrical circuit method. Topography (elevation, slope, and aspect), climate (precipitation and temperature), and soil (acidity, electrical conductivity, soil texture, calcium, magnesium, sodium, phosphorus, potassium, organic carbon, organic matter, saturation percentage, and total neutralizing value) were used in this study. The presence and absence points of the L. vulgare were recorded using a stratified-random sampling method by means of a global positioning system. Soil samples were collected at a depth of 0 to 30 cm where L. vulgare was present and also where it was absent. According to the results, in LR, the variables of temperature, phosphorus, organic matter, and sand and in the MaxEnt, the variables of sand, total neutralizing value (TNV), and silt were the most influential factors on the distribution of L. vulgare. The appraisal of the MaxEnt performance and the precision of the model prediction were 0.97. The Kappa indices for the predicted map obtained from the LR and MaxEnt models were 0.80 and 0.73, respectively. The models' evaluation indicated that both models were able to predict the distribution of L. vulgare habitats with a high level of accuracy; however, LR was more reliable. According to the LR prediction, 9.91% (10,556.25 ha) of the Namin region was attacked by L. vulgare. Connectivity assessment showed that the current density spread of L. vulgare continued from the northeast of the Namin region toward the southeast. On the other hand, the higher current density spread was demonstrated in the eastern region (rangelands adjacent to Fandoghlu forests), and other rangelands which are more threatened by the invasion of L. vulgare. Identifying sites exposed to invasive species can help implement programs to prevent invasive species from invading areas where management and prevention should be implanted to prevent and/or reduce the spread.

Comparison of spectral and spatial-based approaches for mapping the local variation of soil moisture in a semi-arid mountainous area.

Accurate information on soil moisture (SM) is critical in various applications including agriculture, climate, hydrology, soil and drought. In this paper, various predictive relationships including regression (Multiple Linear Regression, MLR), machine learning (Random Forest, RF; Triangular regression, Tr) and spatial modeling (Inverse Distance Weighing, IDW and Ordinary kriging, OK) approaches were compared to estimate SM in a semi-arid mountainous watershed. In developing predictive relationship, Remote Sensing datasets including Landsat 8 satellite imagery derived surface biophysical characteristic, ASTER digital elevation model (DEM) derived surface topographical characteristic, climatic data recorded at the synoptic station and in situ SM data measured at Landsat 8 overpass time were utilized, while in spatial modeling, point-based SM measurements were interpolated. While 70%(calibration set) of the measured SM data were used for modeling, 30%(validation set) were used to evaluate modeling accuracy. Finally, the SM uncertainty maps were created for different models based on a bootstrapping approach. Among the environmental parameter sets, land surface temperature (LST) showed the highest impact on the spatial distribution of SM in the region at all dates. Mean R2(RMSE) between measured and modeled SM on three dates obtained from the MLR, RF, IDW, OK, and Tr models were 0.70(1.97%), 0.72(1.92%), 0.59(2.38%), 0.59(2.27%) and 0.71(1.99%), respectively. The results showed that RF and IDW produced the highest and lowest performance in SM modeling, respectively. Generally, the performance of RS-based models was higher than interpolation models for estimating SM due to the influence from combination of topographic parameters and surface biophysical characteristics. Modeled SM uncertainty with different models varies in the study area. The highest uncertainty in SM modeling was observed at the north part of the study area where the surface heterogeneity is high. Using RS data increased the accuracy of SM modeling because they can capture the surface biophysical characteristics and topographical properties heterogeneity.

Improved digital soil mapping with multitemporal remotely sensed satellite data fusion: A case study in Iran.

Modeling and mapping of soil properties are critical in many environmental, climatic, ecological and hydrological applications. Digital soil mapping (DSM) techniques are now commonly applied to predict soil properties with limited data by developing predictive relationships with environmental covariates. Most studies derive covariates from a digital elevation model (named static covariates). Many works also include single-day remotely sensed satellite imagery. However, multitemporal satellite images can capture information about soil properties over time and bring additional information in predicting soil properties in DSM. We refer to covariates derived from multitemporal satellite images as dynamic covariates. The objective of this study was to assess the performance of DSM when using terrain derivatives (static covariates), single-date remotely sensed satellite indices (limited dynamic covariates), multitemporal satellite indices (dynamic covariates), and combinations of terrain derivatives and satellite indices (covariate fusion) as covariates in predicting soil properties and estimating uncertainty. Three soil properties are considered in this study: organic carbon (OC), sand content, and calcium carbonate equivalent (CCE). Inclusion of single and/or multitemporal remotely sensed satellite indices improved the prediction of soil properties over traditionally used terrain indices. Significant improvements were observed in the prediction of soil properties using two models, Cubist and random forest (RF). The increase in the R2 values for Cubist and RF were 126% and 78% for OC, 110% and 54% for sand, and 87% and 32% for CCE. The RMSE decreased by 34% and 27% for OC, 25% and 12% for sand, and 39% and 19% for CCE, when compared to the terrain indices only model. This also reduced the uncertainty of estimation and mapping. These clearly showed the advantage of using multitemporal satellite data fusion rather than simply using static terrain indices for DSM of soil properties to deliver a great potential in improving soil modeling and mapping for many applications.

Varying the vegetative and morphological traits of Thymus kotschyanus L. submitted to potassium silicate nanoparticles, superabsorbent hydrogel, effective microorganisms, and animal manure / Variação das características vegetativas e morfológ icas de thymus kotschyanus l. submetida a nanopartículas de silicato de potássio, o hidrogel superabsorvente, os microorgan ismos efetivos e o esterco animal

To study the effects of some new facilitators on the vegetative and morphological traits of Thymus kotschyanus, nine treatments were tested in the experimental rangeland field at the University of Mohaghegh Ardabili, Ardabil, Iran. Treatments included control, potassium silicate nanoparticles (PSN) with two levels of 500 and 1000 mg/lit, superabsorbent hydrogel (SH) with two levels of 10 and 30 g/kg, animal manure (AM) with two levels of 100, 200 g/kg, and effective microorganisms (EM) with two levels of 1 and 2%. Data were subjected to one-way analysis of variance (ANOVA). Results of mean comparisons of treatments for Thymus kotschyanus characteristics showed that the highest amount of studied traits were observed in the treated SH 30 g/kg. These traits include plant height (19.44 cm), basal area (4.66 cm), canopy cover (99.11%), number of secondary branches (9.44) and depth of rooting (16.49 cm), aerial parts volume (26.77 cm3), root volume (17.66 cm3), aerial parts fresh weight (14.40 g), aerial parts dry weight (7.18 g), root fresh weight (3.98 g), and root dry weight (2.07 g). In general, the impact of treatments on Thymus growth traits were ranked as follows: SH 30 g/kg, PSN 1000 mg/lit, AM 200 g/kg, SH 10 g/kg, PSN 500 mg/lit, AM 100 g/kg, EM 2%, and EM 1%. In addition,the lowest plant traits were found in control. Overall, it is recommended extending the cultivation of this native medicinal plant by considering ecological conditions in each region. In addition, to promote the establishment and facilitate the growth of planted species, it is recommended using the facilitators utilized in the present work.

Para estudar os efeitos de alguns novos facilitadores sobre as características vegetativas e morfológicas de Thymus kotschyanus L., nove tratamentos foram testados no campo experimental de pastagens na Universidade de Mohaghegh Ardabili, Ardabil, Irã. Os tratamentos incluíram controle, nanopartículas de silicato de potássio (PSN) com dois níveis de 500 e 1000 mg/L, hidrogel superabsorvente (SH) com dois níveis de 10 e 30 g/kg, esterco animal (AM) com dois níveis de 100 e 200 g/kg e microorganismos efetivos (EM) com dois níveis de 1 e 2%. Os dados foram submetidos a uma análise de variância unidirecional (ANOVA). Os resultados das comparações médias dos tratamentos para as características de Thymus kotschyanus L. mostraram que a maior quantidade das características estudadas foi observada no tratamento com SH 30 g/kg. Essas características incluem altura de plantas (19,44 cm), área basal (4,66 cm), cobertura de dossel (99,11%), número de ramos secundários (9,44) e profundidade de enraizamento (16,49 cm), volume de partes aéreas (26,77 cm3), volume de raiz (17,66 cm3), peso fresco de partes aéreas (14,40 g), peso seco de partes aéreas (7,18 g), peso fresco de raiz (3,98 g) e peso seco de raiz (2,07 g). Em geral, o impacto dos tratamentos nas características de crescimento de Thymus foi classificado da seguinte forma: SH 30 g/kg, PSN 1000 mg/L, AM 200 g/kg, SH 10 g/kg, PSN 500 mg/L, AM 100 g/kg, EM 2% e EM 1%. Além disso, as características mais baixas da planta foram encontradas no controle. Em geral, recomenda-se estender o cultivo desta plantamedicinal nativa considerando as condições ecológicas em cada região. Além disso, para promover o estabelecimento e facilitar o crescimento de espécies plantadas, recomenda-se utilizar os facilitadores utilizados no presente trabalho.

Spatial and temporal variability of soil moisture in relation with topographic and meteorological factors in south of Ardabil Province, Iran.

This research has been tried to evaluate the spatial and temporal variability of surface soil moisture (SM) in a semi-arid and cold region of Ardabil Province in Iran with an area of about 10,000 km2. The used SM data is the SMAP Enhanced L2 Radiometer Half-Orbit 9 km Soil Moisture, provided by NASA. The study area was subdivided into 120 locations consisting 10 × 12 grids, matching with the pixels of the SMAP images. In order to evaluate the spatial variations of SM, the relation of mean SM with coefficient of variation and standard deviation has been evaluated and, then, the representative location for mean SM of the area has been identified using the index of temporal stability. Moreover, the effect of topographic factors (elevation, slope, and aspect) on spatial variations of SM, and the effect of meteorological factors (rainfall, sunshine hours, temperature, relative humidity, wind speed, and number of dry days) on temporal variations of SM have been investigated. The relation of mean SM with the coefficient of variation and standard deviation represented an exponentially negative and upper convex shape, respectively. The SM content of the representative location had a correlation with the mean SM of the area with the coefficient of determination value of 0.91. Of the three topographic factors, only the slope factor, and of the meteorological factors all of them except the wind speed have showed a significant relationship with SM spatial and temporal variations respectively.

Object-based land-use/land-cover change detection using Landsat imagery: a case study of Ardabil, Namin, and Nir counties in northwest Iran.

In this study, land-use/land-cover (LULC) change in the Ardabil, Namin, and Nir counties, in the Ardabil province in the northwest of Iran, was detected using an object-based method. Landsat images including Thematic Mapper (TM), Landsat Enhanced Thematic Mapper Plus (ETM+), and Operational Land Imager (OLI) were used. Preprocessing methods, including geometric and radiometric correction, and topographic normalization were performed. Image processing was conducted according to object-based image analysis using the nearest neighbor algorithm. An accuracy assessment was conducted using overall accuracy and Kappa statistics. Results show that maps obtained from images for 1987, 2002, and 2013 had an overall accuracy of 91.76, 91.06, and 93.00%, and a Kappa coefficient of 0.90, 0.83, and 0.91, respectively. Change detection between 1987 and 2013 shows that most of the rangelands (97,156.6 ha) have been converted to dry farming; moreover, residential and other urban land uses have also increased. The largest change in land use has occurred for irrigated farming, rangelands, and dry farming, of which approximately 3539.8, 3086.9, and 2271.9 ha, respectively, have given way to urban land use for each of the studied years.