Applications of machine learning in pine nuts classification.
Pine nuts are not only the important agent of pine reproduction and afforestation, but also the commonly consumed nut with high nutritive values. However, it is difficult to distinguish among pine nuts due to the morphological similarity among species. Therefore, it is important to improve the quality of pine nuts and solve the adulteration problem quickly and non-destructively. In this study, seven pine nuts (Pinus bungeana, Pinus yunnanensis, Pinus thunbergii, Pinus armandii, Pinus massoniana, Pinus elliottii and Pinus taiwanensis) were used as study species. 210 near-infrared (NIR) spectra were collected from the seven species of pine nuts, five machine learning methods (Decision Tree (DT), Random Forest (RF), Multilayer Perceptron (MLP), Support Vector Machine (SVM) and Naive Bayes (NB)) were used to identify species of pine nuts. 303 images were used to collect morphological data to construct a classification model based on five convolutional neural network (CNN) models (VGG16, VGG19, Xception, InceptionV3 and ResNet50). The experimental results of NIR spectroscopy show the best classification model is MLP and the accuracy is closed to 0.99. Another experimental result of images shows the best classification model is InceptionV3 and the accuracy is closed to 0.964. Four important range of wavebands, 951-957 nm, 1,147-1,154 nm, 1,907-1,927 nm, 2,227-2,254 nm, were found to be highly related to the classification of pine nuts. This study shows that machine learning is effective for the classification of pine nuts, providing solutions and scientific methods for rapid, non-destructive and accurate classification of different species of pine nuts. https://doi.org/10.1038/s41598-022-12754-9
Chemical composition and bioactive compounds of cashew (Anacardium occidentale) apple juice and bagasse from Colombian varieties.
Cashew nut production generates large amounts of cashew apple as residue. In Colombia, cashew cultivation is increasing together with the concerns on residue management. The objective of this study was to provide the first chemical, physical and thermal decomposition characterization of cashew apple from Colombian varieties harvested in Vichada, Colombia. This characterization was focused to identify the important bioactive and natural compounds that can be further valorized in the formulation of food, nutraceuticals, and pharmacological products. The results obtained in this study are helpful to portray the cashew apple as a potential by-product due to its renewable nature and valuable composition, instead of seeing it just as an agricultural residue. For that, cashew apples of Regional 8315 and Mapiria varieties were studied. The natural juice (cashew apple juice) that was extracted from the cashew apples and the remanent solids (cashew apple bagasse) were separately analyzed. The HPLC analytical technique was used to determine the concentration of bioactive compounds, structural carbohydrates, and soluble sugars that constitute this biomass. Spectrophotometric techniques were used to determine the concentration of tannins, carotenoids, and total polyphenols. Mineral content and antioxidant activity (DPPH and ABTS assays) were determined in the biomass. Also, the thermal decomposition under an inert atmosphere or pyrolysis was performed on cashew apple bagasse. The varieties of cashew apple studied in this work showed similar content of bioactive compounds, total phenolic content, and structural carbohydrates. However, the Mapiria variety showed values slightly higher than the Regional 8315. Regarding cashew apple juice, it is rich in tannins and ascorbic acid with values of 191 mg/100 mL and 70 mg/100 mL, respectively, for Mapiria variety. Additionally, the principal reservoir of bioactive compounds and constitutive carbohydrates was the cashew apple bagasse. About 50 wt.% of it was composed of cellulose and hemicellulose. Also, in the bagasse, the ascorbic acid content was in a range of 180-200 mg/100 g, which is higher than other fruits and vegetables. Moreover, alkaloids were identified in cashew apples. The maximum value of antioxidant activity (DPPH assay: 405 TEs/g) was observed in the bagasse of Mapiria variety. The bagasse thermal decomposition started around 150 °C when the structural carbohydrates and other constitutive substances started to degrade. After thermogravimetric analysis, a remanent of 20% of the initial weight suggested the formation of a rich-carbon solid, which could correspond to biochar. Therefore, the cashew apple harvested in Vichada is a valuable reservoir of a wide range of biomolecules that potentially could be valorized into energy, foods, and pharmacologic applications. Nevertheless, future work is necessary to describe the complex compounds of this residual biomass that are still unknown. https://doi.org/10.1016/j.heliyon.2022.e09528
High-resolution weather network reveals a high spatial variability in air temperature in the Central valley of California with implications for crop and pest management
Weather is the most important driver of crop development. However, spatial variability in weather makes it hard to obtain reliable high resolution datasets across large areas. Most growers rely on data from a single station that can be up to 50km away to make decisions about irrigation, pest management and penology-associated cultural practices at the block level. In this regard, we hypothesize that kriging a large network of weather stations can improve thermal time data quality compared to using the closest station. This study aims to explore the spatial variability in California's Central Valley and what is the relationship between the density of weather stations used and the error in the measurement of temperature related metrics and derived models. For this purpose, we used temperature records from January 1st 2020 to March 1st 2021 collected by the California Irrigation Management Information System (CIMIS) and a system of 731 weather stations placed above the canopy of trees in commercial orchards (in-orchard). We observed large discrepancies (>300 GDDTb0) in thermal time accumulation between using an interpolation of all stations available and just using the closest CIMIS station. Our data suggests these differences are not systematic bias but true differences in mesoclimate. Similar results were observed for chill accumulation in areas especially prone to not meeting pistachio chill requirements where the discrepancies between using the site-specific in-orchard weather station network and not using them were up to 10 CP. The use of this high resolution network of weather stations revealed spatial patterns in grape, almond, pistachio and pests phenology not reported before. Whereas previous studies have been focused on predictions at the county or state or regional level, our data suggests that a finer resolution can result in major improvements in the quality of data crucial for crop decision making. https://doi.org/10.1371/journal.pone.0267607
Optimization of hazelnut husk medium for pullulan production by a domestic A. pullulans strain.
Hazelnut husk is one of the most abundant agricultural residue in Turkey. Valorization of this lignocellulosic biomass would provide a promoting alternative for economical production of pullulan. In this study, dried hazelnut husk hydrolysate was used directly as fermentation medium for pullulan production by a domestic strain of Aureobasidium pullulans. The aim of this work was the optimization of some fermentation medium parameters by central composite design using response surface methodology (RSM). The effects of (NH4)2SO4 concentration, the volume of concentrated H2SO4 and the amount of ground hazelnut husk on pullulan production were optimized by RSM. The optimum levels of the fermentation parameters defined as 7.2 gL-1, 2.5 mL and 20 g, respectively. The maximum pullulan and exopolysaccharide concentrations were determined as 74.39 and 75.95 gL-1, respectively in the optimum conditions. Specific growth rate of the strain was found as 0.097 h-1. FTIR spectral attributes confirmed the structure of pullulan. Thermal decomposition temperature of synthesized pullulan was found as 247.15 °C. This study showed that hazelnut husk was one of the novel substrate for production of the pullulan by A. pullulans AZ-6. No previous work was found to utilize dried hazelnut husk as fermentation medium for pullulan production by A. pullulans. https://doi.org/10.1080/10826068.2022.2084625
Adsorption of brilliant green dye onto activated carbon prepared from cashew nut shell by KOH activation: Studies on equilibrium isotherm.
Activated carbon from cashew nut shell via a potassium hydroxide (KOH) at 600 °C in an N2 atmosphere and their characteristics using FT-IR, XRD, SEM with EDS, and BET analysis was investigated. The cashew nut shell activated carbon obtained by KOH activation with a CNS/KOH ratio of 1:1 at 600 °C (N2 atmosphere) for 2 h had the highest surface area (407.80 m2/g) as compared to other ratio samples. Amongst, CNS/KOH ratios of 1:1 sample are used for the adsorbent, they are effects of contact time, pH, adsorbent dose, and initial dye concentration on brilliant green (BG) removal efficiency were studied. Moreover, the Langmuir and Freundlich adsorption models consisted utilized to affirm the adsorption isotherms. They are, best fitting for BG experimental equilibrium data was achieved with the Langmuir isotherm, giving a maximum BG adsorption capacity of 243.90 mg/g. https://doi.org/10.1016/j.envres.2022.113497
The Use of Waste Hazelnut Shells as a Reinforcement in the Development of Green Biocomposites
Biodegradable Mater-Bi (MB) composites reinforced with hazelnut shell (HS) powder were prepared in a co-rotating twin-screw extruder followed by compression molding and injection molding. The effects of reinforcement on the morphology, static and dynamic mechanical properties, and thermal and rheological properties of MB/HS biocomposites were studied. Rheological tests showed that the incorporation of HS significantly increased the viscosity of composites with non-Newtonian behavior at low frequencies. On the other hand, a scanning electron microscope (SEM) examination revealed poor interfacial adhesion between the matrix and the filler. The thermal property results indicated that HS could act as a nucleating agent to promote the crystallization properties of biocomposites. Furthermore, the experimental results indicated that the addition of HS led to a significant improvement in the thermomechanical stability of the composites. This paper demonstrates that the incorporation of a low-cost waste product, such as hazelnut shells, is a practical way to produce low-cost biocomposites with good properties. With a content of HS of 10%, a remarkable improvement in the elastic modulus and impact strength was observed in both compression and injection-molded samples. With a higher content of HS, however, the processability in injection molding was strongly worsened. https://doi.org/10.3390/polym14112151
Novel combination artificial neural network models could not outperform individual models for weather-based cashew yield prediction.
Cashew is an important cash crop which is ecologically sensitive, making it vulnerable to climate change. So, the present study compares the performance of stepwise linear regression (SLR), least absolute shrinkage and selection operator (LASSO), elastic net, and artificial neural network (ANN) individually against the ANN model combined with SLR, LASSO, elastic net, and principal components analysis (PCA) for prediction of cashew yield based on weather parameters. The model performances were evaluated using three approaches: (1) Taylor plot; (2) statistical metrics like coefficient of determination (R2), root mean square error (RMSE), and normalized RMSE (nRMSE); and (3) ranking followed by Kruskal-Wallis and Dunn's post hoc test. The results revealed that during calibration, the R2 and RMSE ranged from 0.486 to 0.999 and 2.184 to 88.040 kg ha-1, respectively, while RMSE and nRMSE varied from 3.561 to 242.704 kg ha-1 and 0.799 to 89.949%, respectively, during validation. Kruskal-Wallis and Dunn's post hoc test revealed LASSO as the best model which was at par with ELNET, SLR, and ELNET-ANN. So, these models can be used for cashew yield prediction for the study area well in advance. https://doi.org/10.1007/s00484-022-02306-1
State-led agricultural subsidies drive monoculture cultivar cashew expansion in northern Western Ghats, India.
Agricultural commodity production constitutes an important livelihood source for farmers but significantly contributes to tropical deforestation and biodiversity loss. While the socioecological effects of agricultural commodities such as palm oil, cocoa and coffee are well studied, the effects for commodities such as cashew (Anacardium occidentale) have received less attention. Global cultivated area for cashew increased rapidly from 526,250 ha in 1980 to ~5.9 million ha in 2018. India is the world's second largest cashew producer, with cashew farms often occurring adjacent to remnant forests. To mitigate deforestation for cashew expansion, it is necessary to understand present-day land use policies and management practices that drive this expansion. Through semi-structured interviews (n = 65) and a literature review on agricultural policies in India, we evaluated the role of state-led land use policies in cashew expansion and characterised present-day cashew farming systems in the Sawantwadi-Dodamarg landscape in India. Agricultural subsidies introduced from 1980s to 1990s encouraged cultivar cashew expansion and influenced land use conversion from rice and privately owned forest to cashew. Farmers preferred cultivar cashew as they produced higher yields faster, although they required more agrochemical inputs and were susceptible to pests and wildlife depredation. About 80% of farmers had planted cashew farms by clearing forests in the past 30 years and expressed interest to continue the same. Farmers avoided applying for government-sponsored compensation for crop losses due to wildlife depredation and chose instead to expand cultivar cashew into forested areas. Our study deepens the understanding of how government-led agricultural subsidies drive farmers' uptake of cashew cultivars, farmers' cashew management practices, and how these factors drive deforestation in this landscape at the state and farm level. We recommend further research with equitable stakeholder participation in cashew farming systems to devise sound planning for forest conservation and sustainability standards for the cashew industry. https://doi.org/10.1371/journal.pone.0269092
Assessing the sustainability index of part-time and full-time hazelnut farms in Giresun and Ordu Province, Turkey
The study's primary purposes were to assess the sustainability of hazelnut farms and explore the effects of part-time and full-time farming types on sustainability in hazelnut production in the Giresun and Ordu Province of Turkey. One hundred fifty-two hazelnut farms were selected using the stratified sampling method, and data were collected by using face-to-face questionnaires. Several steps were taken, including using factor analysis after standardizing the variables to determine their weights to calculate the composite hazelnut farms' sustainability index. The research findings showed that overall hazelnut sustainability scores of farms varied from 0.28 to 0.59, and the average score was 0.44 at sampled farms. The composite hazelnut sustainability index was at an unsatisfactory level. The social and economic sustainability index values of farms were equal, and they were higher than the environmental index value. The values were 0.50 and 0.30, respectively. The economic sustainability index score of full-time farms was higher than that of part-time farms, and part-time farms had higher environmental sustainability index scores than that of full-time farms. Social sustainability scores were not different in terms of farm type. It was recommended that when designing and regulating support policies, policy-makers should differentiate part-time and full-time farming. Training and extension programs must be planned to increase the level of knowledge of every willing farmer. To increase sustainability, specific policies are developed according to the farming type. https://doi.org/10.1007/s11356-022-20966-9
Pecan agroforestry systems improve soil quality by stimulating enzyme activity.
Background: Forest and plantation intercropping are efficient agroforestry systems that optimize land use and promote agroforestry around the world. However, diverse agroforestry systems on the same upper-plantation differently affect the physical and chemical properties of the soil. Methods: The treatments for this study included a single cultivation (CK) pecan control and three agroforestry systems (pecan + Paeonia suffruticosa + Hemerocallis citrina, pecan + Paeonia suffruticosa, and pecan + Paeonia lactiflora). Soil samples were categorized according to the sampling depth (0-20 cm, 20-40 cm, 40-60 cm). Results: The results demonstrated that the bulk density (BD) of soil under the pecan agroforestry system (PPH and PPL) was reduced by 16.13% and 7.10%, respectively, and the soil moisture content (MC) and total soil porosity (TPO) increased. Improvements in the physical properties of the soil under the PPS agroforestry system were not obvious when compared with the pecan monoculture. The soil total phosphorus (TP), total nitrogen (TN), available potassium (AK), and total carbon (TC) increased significantly, while the soil urease (S-UE), alkaline phosphatase (S-AKP), and 1,4-β-N-acetylglucosamines (S-NAG) enzyme activity also increased significantly, following agroforestry. Overall, the pecan agroforestry system significantly improved the physical properties of the pecan plantation soil, enriched the soil nutrients, and increased the activity of soil enzymes related to TC, TN, and TP cycles. https://doi.org/10.7717/peerj.12663