Machine learning ensembles, neural network, hybrid and sparse regression approaches for weather based rainfed cotton yield forecast.

Cotton is a major economic crop predominantly cultivated under rainfed situations. The accurate prediction of cotton yield invariably helps farmers, industries, and policy makers. The final cotton yield is mostly determined by the weather patterns that prevail during the crop growing phase. Crop yield prediction with greater accuracy is possible due to the development of innovative technologies which analyses the bigdata with its high-performance computing abilities. Machine learning technologies can make yield prediction reasonable and faster and with greater flexibility than process based complex crop simulation models. The present study demonstrates the usability of ML algorithms for yield forecasting and facilitates the comparison of different models. The cotton yield was simulated by employing the weekly weather indices as inputs and the model performance was assessed by nRMSE, MAPE and EF values. Results show that stacked generalised ensemble model and artificial neural networks predicted the cotton yield with lower nRMSE, MAPE and higher efficiency compared to other models. Variable importance studies in LASSO and ENET model found minimum temperature and relative humidity as the main determinates of cotton yield in all districts. The models were ranked based these performance metrics in the order of Stacked generalised ensemble > ANN > PCA ANN > SMLR ANN > LASSO> ENET > SVM > PCA SMLR > SMLR SVM > SMLR. This study shows that stacked generalised ensembling and ANN method can be used for reliable yield forecasting at district or county level and helps stakeholders in timely decision-making.

Timing of oomycete-specific fungicide application impacts the efficacy against fruit rot disease in arecanut.

Fungicidal application has been the common and prime option to combat fruit rot disease (FRD) of arecanut (Areca catechu L.) under field conditions. However, the existence of virulent pathotypes, rapid spreading ability, and improper time of fungicide application has become a serious challenge. In the present investigation, we assessed the efficacy of oomycete-specific fungicides under two approaches: (i) three fixed timings of fungicidal applications, i.e., pre-, mid-, and post-monsoon periods (EXPT1), and (ii) predefined different fruit stages, i.e., button, marble, and premature stages (EXPT2). Fungicidal efficacy in managing FRD was determined from evaluations of FRD severity, FRD incidence, and cumulative fallen nut rate (CFNR) by employing generalized linear mixed models (GLMMs). In EXPT1, all the tested fungicides reduced FRD disease levels by >65% when applied at pre- or mid-monsoon compared with untreated control, with statistical differences among fungicides and timings of application relative to infection. In EXPT2, the efficacy of fungicides was comparatively reduced when applied at predefined fruit/nut stages, with statistically non-significant differences among tested fungicides and fruit stages. A comprehensive analysis of both experiments recommends that the fungicidal application can be performed before the onset of monsoon for effective management of arecanut FRD. In conclusion, the timing of fungicidal application based on the monsoon period provides better control of FRD of arecanut than an application based on the developmental stages of fruit under field conditions.

Identification of Potential Phytochemical/Antimicrobial Agents against Pseudoperonospora cubensis Causing Downy Mildew in Cucumber through In-Silico Docking.

Compatibility interactions between the host and the fungal proteins are necessary to successfully establish a disease in plants by fungi or other diseases. Photochemical and antimicrobial substances are generally known to increase plant resilience, which is essential for eradicating fungus infections. Through homology modeling and in silico docking analysis, we assessed 50 phytochemicals from cucumber (Cucumis sativus), 15 antimicrobial compounds from botanical sources, and six compounds from chemical sources against two proteins of Pseudoperonospora cubensis linked to cucumber downy mildew. Alpha and beta sheets made up the 3D structures of the two protein models. According to Ramachandran plot analysis, the QNE 4 effector protein model was considered high quality because it had 86.8% of its residues in the preferred region. The results of the molecular docking analysis showed that the QNE4 and cytochrome oxidase subunit 1 proteins of P. cubensis showed good binding affinities with glucosyl flavones, terpenoids and flavonoids from phytochemicals, antimicrobial compounds from botanicals (garlic and clove), and chemically synthesized compounds, indicating the potential for antifungal activity.

Evaluation of machine learning approaches for prediction of pigeon pea yield based on weather parameters in India.

Pigeon pea is the second most important grain legume in India, primarily grown under rainfed conditions. Any changes in agro-climatic conditions will have a profound influence on the productivity of pigeon pea (Cajanus cajan) yield and, as a result, the total pulse production of the country. In this context, weather-based crop yield prediction will enable farmers, decision-makers, and administrators in dealing with hardships. The current study examines the application of the stepwise linear regression method, supervised machine learning algorithms (support vector machines (SVM) and random forest (RF)), shrinkage regression approaches (least absolute shrinkage and selection operator (LASSO) or elastic net (ENET)), and artificial neural network (ANN) model for pigeon pea yield prediction using long-term weather data. Among the approaches, ANN resulted in a higher coefficient of determination (R2 = 0.88-0.99), model efficiency (0.88-1.00) with subsequent lower normalised root mean square error (nRMSE) during calibration (1.13-12.55%), and validation (0.33-21.20%) over others. The temperature alone or its interaction with other weather parameters was identified as the most influencing variables in the study area. The Pearson correlation coefficients were also determined for the observed and predicted yield. Those values also showed ANN as the best model with correlation values ranging from 0.939 to 0.999 followed by RF (0.955-0.982) and LASSO (0.880-0.982). However, all the approaches adopted in the study were outperformed the statistical method, i.e. stepwise linear regression with lower error values and higher model efficiency. Thus, these approaches can be effectively used for precise yield prediction of pigeon pea over different districts of Karnataka in India.

Molecular Characterization and Genetic Variation in Ceratocystis fimbriata Ell. and Halst. on Pomegranate

Fifteen isolates of Ceratocystis fimbriata collected from different locations in Karnataka were characterized using ITS gene technology. It produced an amplification size of 600−650 bp, which indicated that all the isolates belong to the genus Ceratocystis, thus confirming the identity of the pathogenic isolates. To test genetic variability, isolates were analyzed using microsatellite markers. An UPGMA dendrogram for genetic variation among the isolates showed that all the isolates fell into two major clusters. The first cluster consisted of isolate Cf-10 and the second cluster was further divided into two sub-clusters. Sub-cluster one consisted of isolate Cf-2. Sub-cluster two was again divided into five groups. The first group included isolate Cf-13, the second group consisted of isolate Cf-14, the third group included isolates Cf-1, Cf-4, Cf-6, Cf-7, Cf-8 and Cf-9, the fourth group included Cf-5 and Cf-11, and the fifth group consisted of Cf-3, Cf-12 and Cf-15. The dissimilarity coefficient ranged from 0.00 to 0.20 among the isolates. Isolates Cf-1, Cf-3, Cf-4, Cf-5 Cf-6, Cf-7, Cf-8, Cf-9, Cf-11, Cf-12 and Cf-15 were found to be highly similar, as their dissimilarity coefficient was zero. Maximum dissimilarity (0.20) was found between isolate Cf-10 and all the other isolates, suggesting they were genetically distinct.

Control Efficiency and Yield Response of Chemical and Biological Treatments against Fruit Rot of Arecanut: A Network Meta-Analysis.

Fruit rot disease (FRD) in arecanut has appeared in most of the arecanut growing regions of India in the last few decades. A few comprehensive studies on the management of FRD under field conditions have examined various treatment combinations for disease control and yield response analysis. This study aimed to compare the control efficiencies and yield responses of treatments applied over multiple locations and compute the probable returns of investment (ROIs) for treatment costs. Data were gathered from 21 field trials conducted across five main arecanut growing regions of India in the period 2012−2019. The collected data were subjected to analysis with a multivariate (network) meta-analytical model, following standard statistical protocols. The quantitative, synthesized data were evaluated for the estimated effects of disease pressure (DPLow ≤ 35% of FRDInc in the treatments > DPHigh), mean disease control efficiencies (treatment mean, C), and yield responses (R) corresponding to the tested treatments. Based on disease control efficacy, the evaluated treatments were grouped into three efficacy groups (EGs): higher EGs were observed for the Bordeaux mixture (C, 81.94%) and its stabilized formulation (C, 74.99%), Metalaxyl + Mancozeb (C, 70.66%), while lower EGs were observed in plots treated with Biofight (C, 29.91%), Biopot (C, 25.66%), and Suraksha (C, 29.74%) and intermediate EGs were observed in plots to which microbial consortia (bio-agents) had been applied. Disease pressure acted as a significant moderator variable, influencing yield response and gain. At DPLow, the Bordeaux fungicide mixture (102%, 22% of increased yield) and Metalaxyl + Mancozeb (77.5%, +15.5%) exhibited higher yield responses, with absolute arecanut yield gains of 916.5 kg ha−1 and 884 kg ha−1, while, under DPHigh, Fosetyl-AL (819.6 kg ha−1) showed a yield response of 90.5%. To ensure maximum yield sustainability, arecanut growers should focus on the spraying of fungicides (a mixture of different active ingredients or formulations or products) as a preventative measure, followed by treating palms with either soil microbial consortia or commercial formulations of organic fungicides.

Multigene phylogeny and haplotype analysis reveals predominance of oomycetous fungus, Phytophthora meadii (McRae) associated with fruit rot disease of arecanut in India.

An oomycetous fungus Phytophthora causing fruit rot is the most devastating disease of arecanut in different agro-climatic zones of Karnataka with varied climatic profiles. The main aim of this investigation was to characterize the geo-distant Phytophthora populations infecting arecanut using robust morphological, multi-gene phylogeny and haplotype analysis. A total of 48 geo-distant fruit rot infected samples were collected during the South-West monsoon of 2017-19. Pure culture of the suspected pathogen was isolated from the infected nuts and pathogenic ability was confirmed and characterized. Colony morphology revealed typical whitish mycelium with stellate or petalloid pattern and appearance with torulose hyphae. Sporangia were caducous, semipapillate or papillate, globose, ellipsoid or ovoid-obpyriform in shape and sporangiophores were irregularly branched or simple sympodial in nature. Subsequent multi-gene phylogeny (ITS, ß-tub, TEF-1α and Cox-II) and sequence analysis confirmed the identity of oomycete as Phytophthora meadii which is predominant across the regions studied. We identified 49 haplotypes representing the higher haplotype diversity with varying relative haplotype frequency. Comprehensive study confirmed the existence of substantial variability among geo-distant populations (n = 48) of P. meadii. The knowledge on population dynamics of the pathogen causing fruit rot of arecanut generated from this investigation would aid in developing appropriate disease management strategies to curtail its further occurrence and spread in arecanut ecosystem.

Exploring the Impact of Climatic Variables on Arecanut Fruit Rot Epidemic by Understanding the Disease Dynamics in Relation to Space and Time.

To understand the spatio-temporal dynamics and the effect of climate on fruit rot occurrence in arecanut plantations, we evaluated the intensity of fruit rot in three major growing regions of Karnataka, India for two consecutive years (2018 and 2019). A total of 27 sampling sites from the selected regions were monitored and the percentage disease intensity (PDI) was assessed on 50 randomly selected palms. Spatial interpolation technique, ordinary kriging (OK) was employed to predict the disease occurrence at unsampled locations. OK resulted in aggregated spatial maps, where the disease intensity was substantial (40.25-72.45%) at sampling sites of the Malnad and coastal regions. Further, Moran's I spatial autocorrelation test confirmed the presence of significant spatial clusters (p ≤ 0.01) across the regions studied. Temporal analysis indicated the initiation of disease on different weeks dependent on the sampling sites and evaluated years with significant variation in PDI, which ranged from 9.25% to 72.45%. The occurrence of disease over time revealed that the epidemic was initiated early in the season (July) at the Malnad and coastal regions in contrary to the Maidan region where the occurrence was delayed up to the end of the season (September). Correlations between environmental variables and PDI revealed that, the estimated temperature (T), relative humidity (RH) and total rainfall (TRF) significantly positively associated (p = 0.01) with disease occurrence. Regression model analysis revealed that the association between Tmax, RH1 and TRF with PDI statistically significant and the coefficients for the predictors Tmax, RH1 and TRF are 1.731, 1.330 and 0.541, respectively. The information generated in the present study will provide a scientific decision support system, to generate forecasting models and a better surveillance system to develop adequate strategies to curtail the fruit rot of arecanut.

Spatial distribution and identification of potential risk regions to rice blast disease in different rice ecosystems of Karnataka.

Rice is a globally important crop and highly vulnerable to rice blast disease (RBD). We studied the spatial distribution of RBD by considering the 2-year exploratory data from 120 sampling sites over varied rice ecosystems of Karnataka, India. Point pattern and surface interpolation analyses were performed to identify the spatial distribution of RBD. The spatial clusters of RBD were generated by spatial autocorrelation and Ripley's K function. Further, inverse distance weighting (IDW), ordinary kriging (OK), and indicator kriging (IK) approaches were utilized to generate spatial maps by predicting the values at unvisited locations using neighboring observations. Hierarchical cluster analysis using the average linkage method identified two main clusters of RBD severity. From the Local Moran's I, most of the districts were clustered together (at I > 0), except the coastal and interior districts (at I < 0). Positive spatial dependency was observed in the Coastal, Hilly, Bhadra, and Upper Krishna Project ecosystems (p > 0.05), while Tungabhadra and Kaveri ecosystem districts were clustered together at p < 0.05. From the kriging, Hilly ecosystem, middle and southern parts of Karnataka were found vulnerable to RBD. This is the first intensive study in India on understanding the spatial distribution of RBD using geostatistical approaches, and the findings from this study help in setting up ecosystem-specific management strategies against RBD.

Immunoadjuvant and Humoral Immune Responses of Garlic (Allium sativum L.) Lectins upon Systemic and Mucosal Administration in BALB/c Mice.

Dietary food components have the ability to affect immune function; following absorption, specifically orally ingested dietary food containing lectins can systemically modulate the immune cells and affect the response to self- and co-administered food antigens. The mannose-binding lectins from garlic (Allium sativum agglutinins; ASAs) were identified as immunodulatory proteins in vitro. The objective of the present study was to assess the immunogenicity and adjuvanticity of garlic agglutinins and to evaluate whether they have adjuvant properties in vivo for a weak antigen ovalbumin (OVA). Garlic lectins (ASA I and ASA II) were administered by intranasal (50 days duration) and intradermal (14 days duration) routes, and the anti-lectin and anti-OVA immune (IgG) responses in the control and test groups of the BALB/c mice were assessed for humoral immunogenicity. Lectins, co-administered with OVA, were examined for lectin-induced anti-OVA IgG response to assess their adjuvant properties. The splenic and thymic indices were evaluated as a measure of immunomodulatory functions. Intradermal administration of ASA I and ASA II had showed a four-fold and two-fold increase in anti-lectin IgG response, respectively, vs. the control on day 14. In the intranasal route, the increases were 3-fold and 2.4-fold for ASA I and ASA II, respectively, on day 50. No decrease in the body weights of animals was noticed; the increases in the spleen and thymus weights, as well as their indices, were significant in the lectin groups. In the adjuvanticity study by intranasal administration, ASA I co-administered with ovalbumin (OVA) induced a remarkable increase in anti-OVA IgG response (~six-fold; p < 0.001) compared to the control, and ASA II induced a four-fold increase vs. the control on day 50. The results indicated that ASA was a potent immunogen which induced mucosal immunogenicity to the antigens that were administered intranasally in BALB/c mice. The observations made of the in vivo study indicate that ASA I has the potential use as an oral and mucosal adjuvant to deliver candidate weak antigens. Further clinical studies in humans are required to confirm its applicability.

Defensive Mechanisms in Cucurbits against Melon Fly (Bactrocera cucurbitae) Infestation through Excessive Production of Defensive Enzymes and Antioxidants.

Melon fly (Bactrocera cucurbitae) is the most common pest of cucurbits, and it directly causes damage to cucurbit fruits in the early developmental stage. The infection of fruit tissues induces oxidative damage through increased generation of cellular reactive oxygen species. The effects of melon fly infestation on the production of defensive enzymes and antioxidant capabilities in five cucurbit species, namely, bottle gourd, chayote, cucumber, snake gourd, and bitter gourd, were investigated in this study. The total phenolic and flavonoid content was considerably higher in melon fly infestation tissues compared to healthy and apparently healthy tissues. The chayote and bottle gourd tissues expressed almost 1.5- to 2-fold higher phenolic and flavonoid contents compared to the tissues of bitter gourd, snake gourd, and cucumber upon infestation. Defensive enzymes, such as peroxidase (POD), superoxide dismutase (SOD), polyphenol oxidase (PPO), and catalase (CAT), were high in healthy and infected tissues of chayote and bottle gourd compared to bitter gourd, snake gourd, and cucumber. The activity of POD (60-80%), SOD (30-35%), PPO (70-75%), and CAT (40-50%) were high in infected chayote and bottle gourd tissue, representing resistance against infestation, while bitter gourd, snake gourd, and cucumber exhibited comparatively lower activity suggesting susceptibility to melon fly infection. The antioxidant properties were also high in the resistant cucurbits compared to the susceptible cucurbits. The current research has enlightened the importance of redox-regulatory pathways involving ROS neutralization through infection-induced antioxidative enzymes in host cucurbit resistance. The melon fly infestation depicts the possible induction of pathways that upregulate the production of defensive enzymes and antioxidants as a defensive strategy against melon fly infestation in resistant cucurbits.

Assessment of the Spatial Distribution and Risk Associated with Fruit Rot Disease in Areca catechu L.

Phytophthora meadii (McRae) is a hemibiotrophic oomycete fungus that infects tender nuts, growing buds, and crown regions, resulting in fruit, bud, and crown rot diseases in arecanut (Areca catechu L.), respectively. Among them, fruit rot disease (FRD) causes serious economic losses that are borne by the growers, making it the greatest yield-limiting factor in arecanut crops. FRD has been known to occur in traditional growing areas since 1910, particularly in Malnad and coastal tracts of Karnataka. Systemic surveys were conducted on the disease several decades ago. The design of appropriate management approaches to curtail the impacts of the disease requires information on the spatial distribution of the risks posed by the disease. In this study, we used exploratory survey data to determine areas that are most at risk. Point pattern (spatial autocorrelation and Ripley's K function) analyses confirmed the existence of moderate clustering across sampling points and optimized hotspots of FRD were determined. Geospatial techniques such as inverse distance weighting (IDW), ordinary kriging (OK), and indicator kriging (IK) were performed to predict the percent severity rates at unsampled sites. IDW and OK generated identical maps, whereby the FRD severity rates were higher in areas adjacent to the Western Ghats and the seashore. Additionally, IK was used to identify both disease-prone and disease-free areas in Karnataka. After fitting the semivariograms with different models, the exponential model showed the best fit with the semivariogram. Using this model information, OK and IK maps were generated. The identified FRD risk areas in our study, which showed higher disease probability rates (>20%) exceeding the threshold level, need to be monitored with the utmost care to contain and reduce the further spread of the disease in Karnataka.

Application of homobrassinolide enhances growth, yield and quality of tomato.

Brassinosteroids (BRs) have emerged as pleiotropic phytohormone owing to their wide function in crop growth and metabolism. Homobrassinolide (HBR) being an analogue of BRs is known to improve the growth, yield and quality parameters in many crop plants. Thus, an evaluation study was conducted for two years (2018 and 2019) to elucidate the performance of tomato plants (Solanum lycopersicum L.) to a novel group of phytohormone,HBR. The field experiment comprised of seven treatments with homobrassinolide 0.04% (Emulsifiable Concentrate) EC at four different concentrations (0.06, 0.08, 0.10 and 0.12 g active ingredient (a.i.) ha-1) and two well-known growth promoters viz., Gibberellic acid (GA), Naphthalene Acetic Acid (NAA) along with the untreated control. Plant height and chlorophyll concentration were found significantly different in both years of experiment as well as among the different treatments. HBR at 0.12 g a.i. ha-1 was found better with maximum number of fruits (77.36 plant-1), fruit length (6.72 cm), fruit breadth (6.45 cm) and fruit weight (80.52 g) over other concentrations and treatments. Fruit yield was more pronounced in the plots treated with plant growth regulators compared to untreated control. However, significantly higher fruit yield of 91.07 t ha-1 (62.58 t ha-1 with untreated control) along with improved quality traits viz., fruit firmness (4.11 kg cm-2), ascorbic acid content (24.09 mg 100 g-1), total soluble solids (4.43°Brix) and keeping quality (12.50 days) was recorded in 0.12 g a.i. ha-1 HBR treated plots. Thus, it can be inferred that HBRapplication would be a better option to enhance growth, yield as well as quality traits in tomato.

Influence of sowing windows and genotypes on growth, radiation interception, conversion efficiency and yield of guar.

Crop growth largely depends on radiation. Radiation is the main impetus for photosynthesis and movement of photosynthates from source to sink. Therefore, identification of the optimum sowing windows and suitable cultivars for efficient utilization of radiation is of prime importance. A field study was conducted in red clay soil during 2014 and 2015 Kharif season and the treatments consisted of three genotypes and three sowing windows by using randomized complete block design with three replications. The effect of genotypes and sowing windows was found significant with respect to number of trifoliate leaves, leaf area ratio, dry matter production, grain numbers, pod length, test weight, grain yield, and stover yield of guar during 2014 as compared to 2015 sown crop. Statistically significant plant height, number of trifoliate leaves, number of branches, leaf area ratio, absolute growth rate, leaf area index, dry matter, grain number, pod length, grain yield, stover yield and a higher cumulative radiation interception were recorded with 15th August sown crop as compared to other sowing windows. The plant height, number of trifoliate leaves, number of branches, leaf area ratio, absolute growth rate, leaf area index, dry matter, grain number, pod length, grain yield, stover yield and maximum cumulative interception of radiation were significant with RGC-1003 as compared to RGC-936 and HG-365. It is observed that the incident PAR to dry matter accumulation conversion efficiency was varied with cultivars and different sowing windows which ranges from 0.74 g MJ-1 to 0.79 g MJ-1.

Assessing the Impact of Higher Levels of CO2 and Temperature and Their Interactions on Tomato (Solanumlycopersicum L.).

Climate change has increasing effects on horticultural crops. To investigate the impact of CO2 and temperature at elevated levels on tomato production and quality of fruits an experiment was conducted by growing plants in open top chambers. The tomato plants were raised at EC550 (elevated CO2 at 550 ppm) and EC700 (elevated CO2 at 700 ppm) alone and in combination with elevated temperature (ET) + 2 °C in the open top chambers. These elevate CO2 and temperature treatment effects were compared with plants grown under ambient conditions. Outcome of the experiment indicated that growth parameters namely plant stature in terms of height (152.20 cm), leaf number (158.67), canopy spread (6127.70 cm2), leaf area (9110.68 cm2) and total dry matter (223.0 g/plant) were found to be high at EC700 compared to plants grown at ambient conditions in open field. The plants grown at EC700 also exhibited significantly higher number of flowers (273.80) and fruits (261.13), more fruit weight (90.46 g) and yield (5.09 kg plant-1) compared to plants grown at ambient conditions in open field. The percent increase in fruit yield due to EC varied from 18.37 (EC550) to 21.41 (EC700) percent respectively compared to open field and the ET by 2 °C has reduced the fruit yield by 20.01 percent. Quality traits like Total Soluble Solids (3.67 °Brix), reducing sugars (2.48%), total sugars (4.41%) and ascorbic acid (18.18 mg/100 g) were found maximum in EC700 treated tomato than other elevated conditions. Keeping quality was also improved in tomato cultivated under EC700 (25.60 days) than the open field (17.80 days). These findings reveal that CO2 at 700 ppm would be a better option to improve both quantitative as well as qualitative traits in tomato. Among the combinations, EC550 + 2 °C proved better than EC700 + 2 °C with respect to yield as well as for the quality traits. The tomato grown under ET (+2 °C) alone recorded lowest growth and yield attributes compared to open field conditions and rest of the treatments. The positive influence of EC700 is negated to an extent of 14.35 % when the EC700 combined with elevated temperature of + 2 °C. The present study clearly demonstrates that the climate change in terms of increased temperature and CO2 will have a positive effect on tomato by way of increase in production and quality of fruits. Meanwhile the increase in EC beyond 700 ppm along with ET may reduce the positive effects on yield and quality of tomato.