Deep placement of nitrogen fertilizer is an alternative for increasing the N use efficiency of DSR besides minimizing the adverse effects of fertilizers on the environment. In present study, the results showed that both D8 and D12 produced higher grain yields compared with all other treatments. The lowest grain yield was observed under the CK treatment (Table 1). Reasonable N application could improve rice grain yield (Peng et al., 2006). Previous study showed that fertilizer deep placement could significantly increase rice grain yields in comparison to surface broadcasting (Alimata et al., 2015; Azmul H et al., 2016). However, some studies found no effects of deep N placement on grain yields (Setyanto et al., 2000; Maria et al., 2016). We found that deep placement of N fertilizer could increase grain yield due to the increase of productive panicle and spikelet number. The main reasons were for larger leaf area index, and stronger enzyme activities of leaves at the late growth stage (Table 3), which can product more carbohydrate by photosynthesis.
It has been well documented that the effects of deep placement of N fertilizer on nitrogen absorption and utilization (Kirda et al., 2005; Geng et al., 2015; Azmul et al., 2016); however, few studies have focused on the differences in fertilization deep placement depth on crop growth. Our results indicated that significantly higher TNA of leaves in both rice cultivars were recorded for both D8 and D12. And the D12 treatment also resulted in highest TNA of stem among all the treatments followed by D8 and D4, whilst the lowest was found for CK (Fig 1). Huda et al. (2016) found that deep placement of urea birquettes of NPK could significantly increase NRE than broadcast prilled briquettes because of slow movement of NH4+-N from the deep soil layer to the soil’s surface or floodwater. Liu et al. (2015) also observed that compound fertilizer point deep placed at 10 cm depth significantly decreased cumulative ammonia volatilization (AV) and increased NRE, nitrogen partial factor productivity, nitrogen agronomic efficiency, and grain yield in no-tillage paddy fields compared to deep placed at 5 cm and 20 cm depths treatments. In present study, the highest NAE and NRE were found under D12 treatment, while the lowest was found in D0 treatment in both TY998 and YXYZ (Table 2), which was consistent with the finding of Li et al (2015). Yao et al. (2018) found that nitrogen deep placement could increase the contact of nitrogen fertilizer granules with soil, decrease nitrogen losses through runoff and AV. The D0 treatment (surface broadcasting) resulted in the lowest NRE in our study. This result agrees with Xu et al. (2013), who found that surface broadcasting ammonium bicarbonate resulted in lower nitrogen recovery efficiency because of much higher ammonia volatilization (AV) compared to urea and control released urea, whilst deep placement of N could reduce AV rates and lag the AV process.
Sun et al. (2015) found that both 10 cm and 15 cm deep fertilization could significantly promoted greater taproot length and dry weight. In present study, we observed that deep placement of N fertilizer (both D8 and D12) could promote rice root growth (such as larger root length, root superficial area, and thicker root diameter) (Fig 2a, 2b), increase nitrogen uptake by rice plants and improve nitrogen utilization efficiency, compared with all other treatments (Table 2). However, lest total root length per hill, total root superficial area per hill, average root diameter per hill and total root volume per hill were recorded for D0 treatment for both rice cultivars. The main reason was that broadcasting N fertilizer onto the soil surface resulted in more AV due to the higher NH4+–N concentration in the flood water at the early growth stage. And the root of rice plant is very small and less plant N demand. However, there is no need to maintain a high N concentration when rice plant requires more N during the late growth stage. Our results agreed with the finding of Li et al. (2016), who found that surface broadcasting fertilizer onto the soil surface as basal fertilizer markedly decreased NRE due to higher NH4+–N concentration in the flood water because of smaller rice root and less plant N demand. Correlation analysis also indicated that there were significantly positive correlations between root superficial area (RSA), average root diameter (ARD), total root volume (TRV) and NRE (Table 4). These results clearly showed that deep placement of N fertilizer could promote rice root growth, thus, absorb more N, increase nitrogen recovery efficiency.
NR is the rate-limiting enzyme in NO3- assimilation. Numerous studies have reported that reasonable N fertilization could increase NR and GS activities and nitrogen accumulation (Zhong et al., 2018; Liu et al., 2019). GS and GOGAT are key enzymes in incorporation with NH4+ into amides and amino acids (Lawlor, 2002; Masclaux-Daubresse et al., 2006). Tian et al. (2018) observed that excess of N fertilization did not increase the leaf area index, flag leaf photosynthesis rate, flag leaf NR and GS activities, dry matter accumulation, and N uptake in wheat. Wang et al. (2018) also found that higher N application increased carbon and N content, NR and GS activities in rice leaves, while it decreased GOGAT and glutamate dehydrogenase activities. In this study, we found that D12 produced the highest nitrate reductase (NR) and glutamine synthetase activity (GOGAT) of roots at the heading stage for both two rice cultivars. Compared to the CK treatment, significantly higher GS activity was found than that under D8 and D12 treatments in both two rice cultivars. No markedly differences were found in GOGAT and GS between D8 and D12. Our finding was not consistent with Wang et al. (2018), which was probably influenced by the nitrogen rate.