Optimizing fertigation schemes based on root distribution

Abstract

When fertigation is applied, fertilizer is often injected into drip irrigation systems according to a single-pulse application method (SPAM) where a single dose is given either in the early, middle, or late stage of an irrigation event. To improve the consistency of distributions between nutrients and roots and thus promote root nutrient uptake, a root distribution-based multistage application method (RMAM) was proposed, where fertilizer was injected in stages during each irrigation event and proportionally according to the distribution of normalized root length density. A case study to verify the potential of RMAM and to test it against other fertigation scheduling strategies was conducted in a two-year field experiment with drip-irrigated cotton under film mulch in Xinjiang, China. Results indicated fertilizer application strategies significantly impacted soil nutrient dynamics. Generally, SPAM caused nutrient accumulation in certain soil layers, while uniform multistage application method (UMAM) resulted in a more even nutrient profile. Optimal nutrient profile for root uptake was obtained by RMAM, where more nutrients were located in the upper soil layers containing more roots. Compared to late-stage SPAM, the best among all the schemes of UMAM and SPAM, RMAM did not influence deep leaching of mineral N and available P, but did reduce leachate of available K by an average of 3.5% during the fertigation periods over the two seasons. Furthermore, at the late boll-opening stage, due to 25.8%, 35.7% and 35.3% increases of average aboveground accumulation (representing root nutrient uptake) for N, P and K, average soil residual respectively decreased 23.8%, 6.2% and 3.1%, and average cotton yield enhanced 5.5%. Besides root length density, crop nutrient uptake is also affected by many other complicated factors such as soil water/nutrient dynamics and root uptake activity, and thus RMAM should be further investigated to consider these factors comprehensively.