Cultivation of rainfed maize in Bangladesh is not possible in the winter season because of the erratic distribution of rainfall and high evaporation. So the production of maize inevitably relys on irrigation water availability. In order to evaluate the performance of the AquaCrop model, data were obtained from experiments conducted at the IWM research field of Bangladesh Agricultural Research Institute, Gazipur (24°0´ N, 90°25´ E, and 8 m above mean sea level) during the winter season of 2015-2016. The experimental site is characterized by a tropical monsoon climate with a cool and dry winter from October through early March with temperatures ranging from 11° C to 27° C, and an average total winter rainfall of about 15.9 cm. A variety of hybrid maize (BARI Maize-9) was cultivated under full and deficit irrigation conditions. The experiment was set in a randomized complete block design (RCBD) with three replications. Seeds of maize were planted on 6 December 2015 at a spacing of 60 cm x 20 cm in field plots and each plot size was 6 m x 4.2 m.. The following treatments were arranged in the experiment:
T1 = Rainfed
T2 = Irrigation at seedling, vegetative, silking and grain filling stages (full irrigation)
T3 = Irrigation at vegetative, silking and grain filling stages (stress at an early stage)
T4 = Irrigation at seedling, vegetative and grain filling stages (stress at mid-stage)
T5 = Irrigation at seedling, vegetative and grain filling stages (stress at a late stage)
AquaCrop requires the input data files for climate, crop, soil, irrigation, and initial soil water conditions (Raes et al., 2009) which were assembled using the field data.
Crop management: Irrigation water was applied to bring the soil moisture to field capacity. Soil water content was measured gravimetrically in each 0.15 m layer down to 0.6 m depth before and after each irrigation and at every 10 days intervals. It was also monitored at planting and at harvest to find out the soil moisture contribution to crop growth. Furrow method of irrigation was used and a measured amount of water was applied to each plot by polyethylene hose pipe. The amount of irrigation water needed in each irrigation event was calculated using the formula suggested by Michael (1978).
The nutrient requirements were determined based on soil analysis and were adequately met by fertilizer applications. Nutrients were applied before planting and nitrogen was also applied as topdressing at 30 and 55 DAP. Intercultural operations such as weeding and pesticide application were done as and when necessary. So, no pests or disease infestations were observed during the crop growing season. The crop was harvested on 29 April 2016.
Data recording: The date of sowing and date of emergence was recorded. The date of emergence was considered when 90% of seedlings had emerged. Flowering initiation and duration of flowering, maximum canopy cover (CC), senescence and maturity were also observed. Senescence was assumed to be reached when the canopy starts to decline. Growth data including canopy cover, above-ground biomass and rooting depth were measured at every 10 days interval. Canopy cover was monitored using a grid system and canopy meter. Aboveground biomass and rooting depth were determined by removing 10-plants at the seedling stage and three plants per plot at other stages from two replications. Samples of above-ground biomass were then oven-dried at 720 C for 3 days to document biomass production (Ehdaie and Waines, 2001). The harvesting of experiments was accomplished manually at physiological maturity and above-ground dry biomass and grain yield were recorded.
Weather and soil data: Daily maximum and minimum air temperatures (0C), sunshine hours, rainfall (mm), wind speed (km/hr) and evaporation (mm/day) data were obtained from weather stations situated about 1km distance from the experimental field. With these meteorological data, the daily ETo was calculated using the Penman-Monteith equation as described in Allen et al. (1998), with the FAO ETo calculator (Version 3.1). The soil was silty clay loam having a bulk density of 1.46 gm/cc and soil water contents at permanent wilting point (PWP) and at field capacity (FC) equaled 13.5% and 27.5%, respectively. Saturated hydraulic conductivity was not measured in the field and a default value suggested by the model was adopted.
Model description: AquaCrop (Steduto et al., 2009), based on Doorenbos and Kassam (1979) principles, simulates the attainable crop biomass and harvestable yield as a function of water consumption under rainfed, supplemental, deficit, and full irrigation conditions. The growth engine of AquaCrop is water-driven, in that transpiration is calculated first and translated into biomass using a conservative, crop-specific parameter (Geerts et al., 2009), the biomass water productivity, normalized for atmospheric evaporative demand and air CO2 concentration. The normalization is to make AquaCrop applicable to diverse locations and seasons. Simulations are performed on thermal time but can be on calendar time, in daily time-steps. The model uses canopy ground cover instead of leaf area index (LAI) as the basis to calculate transpiration and to separate soil evaporation from transpiration. Crop yield is calculated as the product of above-ground dry biomass and harvest index (HI), while water stress effects are segregated into; canopy growth, canopy senescence, transpiration, and harvest index (Steduto et al., 2009). The model strikes a balance between accuracy, simplicity, robustness, and ease of use, and is aimed at practical end-users such as extension specialists, water managers, personnel of irrigation organizations, economists, and policy specialists who use simple models for planning and scenario analysis (Hsiao et al., 2009). A schematic representation of the evolution of AquaCrop.