Modelling water and salinity distribution in soil under advance fertigation systems in horticultural crops

Abstract

New generation irrigation and fertilizer application methods have revolutionised input applications in agriculture, leading to significant savings of scarce resources like water. These systems are highly capital intensive, and apply water at suitable locations, facilitating their maximum uptake by crops. However there is still a need to evaluate these systems for water and salinity dynamics on a long term basis, to further finetune them and come to terms with the high cost involved, and to control the possible deep drainage of water which transports solutes below the root zone endangering the quality of groundwater. But the soil-plant-atmosphere system evaluation in respect of water and salt distribution involves very complex processes which are unable to be managed by experimentation alone. The numerical models offer good opportunity to analyse the complex systems if reliable input parameters are available. In this study, HYDRUS-2D model was used to evaluate the seasonal water balance and salinity distribution in soil under surface drip irrigation of two horticultural crops, almond and mandarin which are widely grown in Australia. Modelling domain represents the plant spacing of the trees in the field studies. Modelling simulations were carried out on daily basis involving daily input of variable flux equivalent to the depth of irrigation. The daily potential transpiration (Tpot) and potential evaporation (Es) under almond and mandarin trees were estimated by FAO 56 dual crop coefficient method utilizing local weather parameters, soil characteristics and plant estimates. The estimated daily Tpot and Es were then used as daily atmospheric inputs along with rainfall received at the experimental sites. The modelling simulations were carried out for one complete growing season for both crops. The field experiments were equipped with temporal and spatial monitoring of water content and salinity (in mandarin only) distribution following standard procedures. The modelling output on water content and salinity (for mandarin) were compared with the corresponding measured values throughout the growing season. The model was also used for evaluating the impact of water stress and impact of pulsing in irrigation application. The graphical and statistical comparison of weekly measured and simulated values of moisture content at various depth and distances from dripper revealed fairly good matching. The RMSE values of weekly comparison varied from 0.005 to 0.06 cm3cm-3 in almond and 0.01-0.06 cm3cm-3 in mandarin. The variation of this magnitude generally exists in field measurement of water content. Similarly the comparison between measured and model predicted values of salinity (ECsw) under mandarin also matched well and RMSE ranged between 0.09 to 0.93 dS m-1 which is well within the acceptable limit in a complex and highly dynamic soil system. The daily ECsw distribution remained below the threshold salinity values of both crops throughout the growing season. Reducing the irrigation application by 35% (65% ETC) in almond increased the seasonal salinity substantially. The average increase was about 2.5 fold particularly during the summer season. Hence, the temporal and spatial soil solution salinity distribution (ECsw) obtained from the modelling simulation was very well synchronised with the corresponding moisture regime in the soil under both trees. The water balance revealed that only 49% of applied water (irrigation and rain) was used by the mandarin tree, allowing a leaching fraction of 34%, while 54-55% water uptake efficiency was recorded in almond under surface drip irrigation. The deep drainage losses in almond accounted for 25% of the total water application. The pulsing of irrigation events produced a similar seasonal water uptake as obtained in continuous irrigation. The modelling simulations revealed that there is a need to further finetune system design and irrigation scheduling so that significant savings of water can be realised.