Root mass may affect soil water infiltration more strongly than the incorporated residue

Introduction

Soil degradation is a major constraint of food security (Gomiero, 2016; Lal, 2015), and soil erosion represents one of the crucial intervention points for reversing soil degradation (Karlen & Rice, 2015). The no-tillage method is a major approach to tackle erosion; however, tillage remains a major management approach on arable land. The Universal Soil Loss Equation (USLE) (Wischmeier & Smith, 1978) is the standard for estimating erosion. The equation shows that the risk of erosion is reduced when a crop has covered soil surface. This emphasizes the importance of preventing erosion in the early stage of crop growth. In general, it is rare for rain intensity to exceed 30 mm 10 min⁻¹ (Kouhei et al., 2011). Therefore, surface runoff after tillage almost never occurs if the infiltration rate of a field is larger than 30 mm 10 min⁻¹. Therefore, technologies increasing the infiltration rate higher than the precipitation rate are needed to prevent soil erosion in tillage systems.

Crop residue management is a crucial factor in the early stage of crop growth. Tillage makes soil porous, but the physical properties are rapidly lost (Strudley et al., 2008). However, organic matter application increases the stability of soil pores (Turmel et al., 2015). Potter et al. (1995) reported that water infiltration of soil was higher under no-tillage than tillage conditions when the residue input was low, but the opposite result was shown when the residue input was high. The soil erosion decreased according to the degree of water infiltration (Potter et al., 1995). Just after residue incorporation, rill erosion rates were significantly reduced (Brown et al., 1989). A previous study reported that surface water runoff under normal subtillage reduced up to the applied wheat straw quantity, as the water infiltration increased with the quantity of applied straw (Russel, 1940), although the relation between the quantity of applied residue and infiltration rate has been less studied. Conversely, it is known that plant root mass is related to rill and ephemeral gully erosion (Gyssels et al., 2005).

Our research aims were as follows: 1) to determine whether the relation between residue incorporation and infiltration holds under crop rotation, and 2) to determine whether the remaining underground root mass influences this relation. To address these questions, we designed an experiment that involved ensuring growth of different crops for the same amount of input residue with different nutrition levels and different soil moisture levels.

Methods

Results

The incorporated residues of each house were 2.0 and 2.0 Mg ha⁻¹ for the initial stage (before seeding the rose grass), 3.8 and 4.7 Mg ha⁻¹ after the rose grass, and 0.8 and 0.8 Mg ha⁻¹ after the okra. The corresponding infiltration rates were 45 and 36 mm, 97 and 123 mm, and 32 and 47 mm, respectively. There was a strong correlation between the quantity of incorporated residue dry weight and soil water infiltration rate (r = 0.953) in terms of nitrogen level treatment, although initial corn residue showed outliers (Figure 1a). However, aboveground biomass of the prior crop showed a higher correlation with soil water infiltration rate (r = 0.965), without outliers (Figure 1b). The correlation coefficient of the infiltration rate and the aboveground weight decreased to r = 0.872 for the mulch level treatment (Figure 1c). The soil moisture (0–5 cm) of the N0, N1, and N4 treatment at the end of okra cropping was 8.3% (±1.6%), 7.2% (±1.0%), and 7.8% (±1.7%), respectively (Oda et al., 2018). The soil moisture of the unmulched, the fabric, and the film treatments were 6.5% (±0.7%), 7.6% (±0.8%), and 9.7% (±0.9%), respectively (Oda et al., 2018).

Figure 1. Correlation between input residue or aboveground biomass and the soil water infiltration rate.

(a, b) Means of the nitrogen-level treatment. (c) Means of the mulch-level treatment. Crop rotation was conducted as follows: corn, rose grass, and okra in plastic film houses. An interval of 65 days was provided between the corn harvesting and the rose grass planting. There was no interval between rose grass harvesting and okra planting. Houses A and B are replicates. We measured the soil infiltration rates on the ridge using artificial rainfall equipment on the day of making the ridge. The values are the mean of three plots.

Discussion and conclusions

We found a strong correlation between input residue and the infiltration rates for the nitrogen-level treatment (Figure 1a). However, the average infiltration rate of initial stage was almost the same as that of after okra, although the input quantity of the initial stage (2.0 Mg ha⁻¹) was a 2.5-fold higher than after okra (0.8 Mg ha⁻¹). By contrast, the correlation for the aboveground biomass had no outliers (Figure 1b). This reveals that aboveground biomass affects infiltration rate more than the applied residue. A previous study has shown that the decrease in water erosion rates with increasing root mass is exponential, although infiltration was not mentioned (Gyssels et al., 2005). Our data show a positive relationship between resistance to erosion and root mass when assuming that aboveground biomass is proportional to the underground biomass.

The mulch treatment data shows that correlation between the aboveground weight and infiltration rate was erratic (Figure 1c). The soil moisture difference caused large differences in aboveground growth. In addition, the unit performance of infiltration was not stable. The difference of soil water content at most 3.2% (6.5% of the unmulched and 9.7% of the film) in top 5 cm was negligible against the 25–40 mm of applied water before measurement for making the ridge. Therefore, it is considered that the outliers did not come from the difference of soil moisture but came from the difference of root mass, which was affected by the soil dryness.

We should consider the duration of the “after-effect” of the prior crop, such as the roots of rose grass on the soil water infiltration rate measurement of after okra (Wischmeier & Smith, 1978). Two facts support the observation of a small after-effect. Firstly, the infiltration rates were very small (32 and 47 mm) after okra, although they were very large (97 and 123 mm) after rose grass. Secondly, the correlation between the aboveground biomass and the infiltration rate was stable and less was affected by a prior crop. Therefore, we conclude that the effect of the prior crop root mass almost disappears within the next crop growth period under the experimental conditions.

Finally, the key finding of this study is that the effect of aboveground residue quantity, more precisely root mass, was stronger than the incorporated residue. Although the residue quantity was the main factor determining the infiltration rates at tillage, the fairly low performance of initial corn residue is remarkable. This is likely to be because interval between harvesting corn and rose grass planting was very high (65 days). The soil pores would have been decreased. From a physical viewpoint, the area of residue surface is far smaller than that of the root surface. And the gap is easily clogged by sediment caused by rainfall. Therefore, the improvement of soil water infiltration probably comes from root mass (Gyssels et al., 2005). This suggests the importance of sequential cropping in tropical regions.

Data availability

Raw data of this article are presented in figshare: https://doi.org/10.6084/m9.figshare.6741890.v1 (Oda et al., 2018).

Data are available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication).