Long term use of sodic waters in North India and the reliability of empirical equations for predicting their sodium hazard

Various predictive equations (SAR, adjusted R, and adjusted SAR) determining sodium hazard of irrigation waters were tested against in situ ESP data obtained from different agro-climates of Northern India under different crop sequences, where sodic and saline-sodic waters (EC 0.6-10dS m-1SAR 10 43, adjusted SAR 15-100 and RSC 015 me 1-1) are in use for the last 5-30 y. It was observed that no one predictive equation gave a good estimate of soil ESP (0-30 cm depth) developed under wheat-fallow, wheat-rice and wheat-millets crop sequence. For wheat-fallow, SAR and adj.RNa under predicted soil ESP by 1.5 and 1.3 times, respectively, whereas adj.SAR over predicted it by 2 times. For wheat-rice, SAR, adj.RNaand adj.SAR under predicted it by 3.4, 1.9 and 1.5 times, respectively, whereas none of these fitted well for wheat-millets crop sequence. For wheat-fallow and wheat-rice sequences ESP build up of soil was best predicted by relations 1.5 SARiw and 1.5 adj.SARiw, respectively. Thus there is a need to develop predictive models keeping in view the water requirement of crops in relation to different soil types and climates.