Precise forecasting of reference evapotranspiration (ET₀) is one of the critical initial steps in determining crop water requirements, which contributes to the reliable management and long-term planning of the world’s scarce water sources. This study provides daily prediction and multi-step forward forecasting of ET₀ utilizing a long short-term memory network (LSTM) and a bi-directional LSTM (Bi-LSTM) model. For daily predictions, the LSTM model’s accuracy was compared to that of other artificial intelligence-based models commonly used in ET0 forecasting, including support vector regression (SVR), M5 model tree (M5Tree), multivariate adaptive regression spline (MARS), probabilistic linear regression (PLR), adaptive neuro-fuzzy inference system (ANFIS), and Gaussian process regression (GPR). The LSTM model outperformed the other models in a comparison based on Shannon’s entropy-based decision theory, while the M5 tree and PLR models proved to be the lowest performers. Prior to performing a multi-step-ahead forecasting, ANFIS, sequence-to-sequence regression LSTM network (SSR-LSTM), LSTM, and Bi-LSTM approaches were used for one-step-ahead forecasting utilizing the past values of the ET₀ time series. The results showed that the Bi-LSTM model outperformed other models and that the sequence of models in ascending order in terms of accuracies was Bi-LSTM > SSR-LSTM > ANFIS > LSTM. The Bi-LSTM model provided multi-step (5 day)-ahead ET₀ forecasting in the next step. According to the results, the Bi-LSTM provided reasonably accurate and acceptable forecasting of multi-step-forward ET₀ with relatively lower levels of forecasting errors. In the final step, the generalization capability of the proposed best models (LSTM for daily predictions and Bi-LSTM for multi-step-ahead forecasting) was evaluated on new unseen data obtained from a test station, Ishurdi. The model’s performance was assessed on three distinct datasets (the entire dataset and the first and the second halves of the entire dataset) derived from the test dataset between 1 January 2015 and 31 December 2020. The results indicated that the deep learning techniques (LSTM and Bi-LSTM) achieved equally good performances as the training station dataset, for which the models were developed. The research outcomes demonstrated the ability of the developed deep learning models to generalize the prediction capabilities outside the training station.

Grain legumes are commonly used for food and feed all over the world and are the main source of protein for over a billion people worldwide, but their production is at risk from climate change. Water deficit and heat stress both significantly reduce the yield of grain legumes, and the faba bean is considered particularly susceptible. The genetic improvement of faba bean for drought adaptation (water deficit tolerance) by conventional methods and molecular breeding is time-consuming and laborious, since it depends mainly on selection and adaptation in multiple sites. The lack of high-throughput screening methodology and low heritability of advantageous traits under environmental stress challenge breeding progress. Alternatively, selection based on secondary characters in a controlled environment followed by field trials is successful in some crops, including faba beans. In general, measured features related to drought adaptation are shoot and root morphology, stomatal characteristics, osmotic adjustment and the efficiency of water use. Here, we focus on the current knowledge of biochemical and physiological markers for legume improvement that can be incorporated into faba bean breeding programs for drought adaptation.

Cadmium (Cd), a potent heavy metal, causes a significant reduction in plant growth and its yield by interfering with the plant’s mineral nutrition and, primarily, by inducing Cd-induced oxidative damage. Cd mobilization at the soil–root interface is also very important in context of its bioavailability to plants. Therefore, an experiment was carried out to evaluate the mitigating role of iron-enriched biochar (Fe-BC) on Cd accumulation in soil and Cd toxicity in radish plants. Radish seeds were sown in pots, and two levels of Cd (0 and 0.75 mg kg⁻¹) and two levels of Fe-BC (0 and 0.5%) were applied. Cd stress significantly reduced radish fresh and dry biomass production, which was due to high production of malondialdehyde (36%) and increase in cell membrane permeability (twofold) relative to control. Moreover, Cd stress considerably reduced chlorophyll concentrations and uptake of some essential nutrients, such as Ca, K, and Fe. Contrarily, Fe-BC application ameliorated Cd toxicity by triggering the activation of antioxidant enzymes (catalase and ascorbate peroxidase), primary and secondary metabolite accumulation (protein and phenolics concentrations), and by improving plant mineral nutrition under Cd treatment, compared with Cd treatment only. The ability of biosorbent material (Fe-BC) to adsorb the Cd ion on its surface and its immobilization from Cd-polluted soil to plant root was determined by using Langmuir and Freundlich isotherm models. Interestingly, Cd concentration was found in soil as diethylenetriamine (DTPA)-extractable soil Cd on radish root, but not reported in radish shoot with Cd+Fe-BC treatment, compared to Cd treatment; suggesting that Fe-BC treatment has a potential to provide extra strength to the root and shoot, and plays an important role in regulation ionic and redox homeostasis under Cd stress.

Cultivating multiple crops together can provide numerous benefits, including improved soil health and crop yield. The objective of our study was to determine the optimum planting techniques in intercropping systems, and to maximize their benefits by mitigating competition for resources such as land, space, light interception, and nutrition. The performance of successively planted maize (Zea mays L.) grown with cowpea (Vigna unguiculata L.) was evaluated with a field trial in Bangladesh. The treatments in our study were: (a) sole maize, (b) sole cowpea, (c) crops sown simultaneously, and (d) crops sown with different time lags (1, 2, and 3 weeks) between the maize-sowing and cowpea-sowing dates. Data on the crops’ physiological parameters were recorded. These included light interception, leaf area index (LAI), Soil Plant Analysis Development (SPAD), harvest index, and yield. Simultaneously, canopy coverage was measured using camera-based photo analysis. In addition, an economic analysis of intercropping maize with soybean or cowpea was conducted using gross margin analysis and benefit-cost ratio. In our results, the below-canopy photosynthetically active radiation (PAR) was significantly higher in intercropping treatments when maize was sown three weeks after cowpea. In contrast, the LAI value of the maize and cowpea was significantly greater when sown on the same day than in other intercropping treatments. As a result, the maize yield reduced when intercropped with cowpea. This reduction maximized when both species were sown simultaneously due to higher competition for resources, including nutrients and light. Intercropping was more beneficial in terms of land equivalent ratio than both sole cropping of maize and cowpea, especially when maize was planted three weeks later. However, this benefit was not retained when calculated as maize equivalent yield since the contribution of cowpea was small in the overall maize yield, suggesting the importance of the relative economic value of the component species. Among all treatments, the lowest maize equivalent yield (6.03 ± 0.14 t ha⁻¹) was obtained from sole cowpea, and the largest land equivalent ratio (1.67 ± 0.05) was obtained from intercropping with maize sown three weeks after cowpea. This treatment provided a net income of USD 786.32 ± 25.08 ha⁻¹. This study has shown that together, maize–cowpea intercropping with a temporal niche difference of three weeks may be a better option for sustainable crop production in Bangladesh, maximizing land use. However, it may not provide a significantly greater maize equivalent yield and economic return.

Several experiments were carried out to establish an efficient regenerating protocol for cultivated eggplant varieties. Among the five varieties cultured on Murashige and Skoog (MS) medium with free plant growth regulator (PGR), Nayantara performed better considering the number of shoots/explant (2.48). Considering explant types and culture conditions, better performance was observed (3.68 shoots/explant) when seed germination in the dark was proceeded by bottom hypocotyl segments cultured under dark conditions. A higher rate of shoot regeneration was observed in Nayantara when cultured in Zeatin Riboside (ZR) and Thidizuron (TDZ) supplemented MS medium. The highest number of shoots per explant was produced on MS medium supplemented with 2.0 mg/L ZR and 0.1 mg/L indole acetic acid (6.65 shoots/explant). Proliferation and elongation of the regenerated shoots were obtained in the MS medium with free PGR. The best rooting performance was observed in MS medium supplemented with 2.0 mg/L indole butyric acid. Plantlets with well developed roots and shoots were successfully transferred to soil