Spatial variability of soil pH and land use as the main influential factor in redbeds of the Nanxiong Basin, China

Soil pH is the main factor affecting soil nutrient availability and chemical substances in soil. It is of great significance to study the spatial variability of soil pH for soil nutrient management and soil pollution prediction. In order to explore the causes of spatial variability of soil pH in redbed areas, the Nanxiong Basin in south China was selected as an example, and soil pH was measured in the topsoil by nested sampling (0–20 cm depth). The spatial variability characteristics of the soil pH were analysed by geostatistics and classical statistical methods, and the main factors influencing the spatial variability of soil pH are discussed. The results showed that the coefficient of variation in the redbed areas of Nanxiong Basin was 17.18%, indicating moderate variability. The geostatistics analysis showed that the spherical model is the optimal theoretical model for explaining the soil pH’s variability, which is influenced by both structural and random factors. The spatial distribution and pattern analysis showed that soil pH content in the northeast and southwest is relatively high, and is lower in the northwest. These results indicate that topographic factors and land use patterns are the main factors. ABSTRACT 15 Soil pH is the main factor affecting soil nutrient availability and chemical substances in soil. It is of great 16 significance to study the spatial variability of soil pH for soil nutrient management and soil pollution 17 prediction. In order to explore the causes of spatial variability of soil pH in redbed areas, the Nanxiong 18 Basin in south China was selected as an example, and soil pH was measured in the topsoil by nested 19 sampling (0–20 cm depth). The spatial variability characteristics of the soil pH were analysed by 20 geostatistics and classical statistical methods, and the main factors influencing the spatial variability of 21 soil pH are discussed. The results showed that the coefficient of variation in the redbed areas of Nanxiong 22 Basin was 17.18%, indicating moderate variability. The geostatistics analysis showed that the spherical 23 model is the optimal theoretical model for explaining the soil pH’s variability, which is influenced by both 24 structural and random factors. The spatial distribution and pattern analysis showed that soil pH content in 25 the northeast and southwest is relatively high, and is lower in the northwest. These results indicate that 26 topographic factors and land use patterns are the main factors. 27


INTRODUCTION
Soil pH is an indicator of the acidity or alkalinity of soil, which is a reflection of important physical and 31 chemical properties determining soil quality (Nagy & Kónya, 2007). Soil pH also has a profound impact 32 on a number of other properties of soil. Extremes in acidity or alkalinity will change the nutrients 33 available and result in element imbalances in plants (Zhao et al., 2011). 34 Spatial heterogeneity refers to the inhomogeneity and complexity of the distribution in space of 35 properties of a system. The spatial heterogeneity of soil parameters such as pH and content of organic 36 matter and of nitrogen, phosphorus and potassium, has an important influence on the distribution and 37 spatial pattern of plants (Stoyan et al., 2000;Augustine & Frank, 2001;Li et al., 2008;Silvia et al., 2016). 38 The study of spatial heterogeneity and of the driving factors behind soil properties is significant for 39 revealing ecosystem function and biodiversity (Augustine & Frank, 2001). 40 With the continuous development of geographic information technology, studying the spatial variability 41 of soil properties by a combination of geostatistics and GIS technology has become one of the hot topics 42 in the different fields in which soil is investigated (Romano, 1993;Foroughifar et al., 2013).
In the formula, N(h) is the logarithm of the distance when the distance equals h, and Z(xi) is the value at 110 location xi; Z(xi + h) is the value at distance h from xi (Yang et al., 2016;Rosemary et al., 2017). 111 Appropriate model functions were fitted to the semivariograms. The semivariograms were used to 112 determine the degree of spatial variability on the basis of the classes of spatial dependence distinguished 113 by Cambardella (1994): strong spatial dependence (C0/(C0 + C) ＞ 75%), moderate spatial dependence 114 (25% < C0/(C0 + C) ＜ 75%) and weak spatial dependence (C0/(C0 + C) ＜25%). In ArcGIS 9.2, we used 115 kriging interpolation in the geostatistics module to draw the soil pH spatial distribution map and trend 116 analysis chart in order to analyse the spatial variability characteristics. According to the soil type map, 117 slope, aspect, elevation, and land use type distribution map, the degree of influence and main control 118 factors of the soil's spatial variation of pH were analysed.

120
Descriptive statistics of soil pH 121 Descriptive statistics of the soil pH are presented in  consisted of the climate, parent material and terrain; these can enhance the spatial dependency of soil pH. 146 In contrast, the random factors, which are the result of human activity such as farming and fertilization, 147 can make the spatial dependency of soil pH weaker (Isaaks & Srivastava, 1989). This moderate spatial 148 dependence of soil pH in the redbeds implies that the spatial variation of soil pH in the study area is 149 mainly caused by both structural and random factors. 150 According to Figure 4, when the separation distance is more than 161 m, the semivariance fluctuates 151 only slightly, and then stabilizes. This trend might be caused by differences in directional variation. The 152 variance at 250 m implies that the range of the spatial dependence is much wider than the sampling 153 interval. Therefore, the current sampling design was appropriate for this study. 154 In order to understand the characteristics of spatial variation in soil pH, the semivariogram was drawn 155 in four directions, E-W (0°), NE-SW (45°), S-N (90°) and SE-NW (135°), using the GS+7.0 software. 156 As shown in Figure 5, the spatial variation exhibits large differences in different directions, showing the 157 heterogeneity. Table 3 shows that the best-fitting models in the four directions are all spherical. The 158 nugget (C0) and sill (C0 + C) values are different and their ratio ranges from 25% to 75%, indicating 159 moderate variation. 160 As shown in Figure 5, The range of the soil pH values from the northeast to the southwest (45°) and 161 from the southeast to the northwest (135°) is significantly smaller than from east to west (0°) and from north to south (90°), indicating that the variation in the 0°and 90°directions is more complex than those 163 at 45°and 135°. 164 From east to west (0°), when the separation distance is greater than 161 m, the difference in the 165 semivariance of the soil pH begins to fluctuate, first increasing and afterward decreasing to around 0.0388. 166 The semivariance from north to south (90°) shows the same trend, alternating between high and low, but 167 the degree of fluctuation in the east-west (0°) direction is smaller. When the separation distance is larger 168 than 169 m, the variation of the soil pH in the NE-SW (45°) and SE-NW (135°) directions is more stable 169 near 0.0388, and the degree of variation is not very different. The main reason is that the area is near the 170 badlands hills in the NE-SW and the SE-NW directions; the topography and parent materials are of great 171 influence, and in the SE-NW direction there are more hills and larger undulations. However, in the N-S 172 and E-W directions (0°and 90°, respectively), the soil pH shows high spatial homogeneity because the 173 relief is low and the only land use is farmland in these directions. Taken together, the soil pH in this study 174 area has an obvious spatial heterogeneity, which is suitable for further interpolation analysis. 176 The effect of trends is a prerequisite for and the basis of prediction by kriging interpolation. The lower the 177 order of the trend effect is, the smaller the number of parameters will be that are required for kriging 178 interpolation. Thus, a lower order of the trend effect can reduce error, and many scholars take the  (Fig. 7); the soil pH is higher in the 194 northeast and the southwest, increases towards the southwest, and decreases towards the northwest. The 195 result of inverse distance weighting interpolation of the 3D map shows that the overall trend for the pH in 196 the study area is consistent with the results from kriging interpolation (Fig. 8).

Analysis of the spatial distribution of soil pH
197 198 Although the spatial variation of soil pH in the study area is determined by structural factors such as 199 topographic factors, and the random factors of human fertilization, it is still not known what extent each 200 factor affects the spatial variation of soil pH. Therefore, two factors (topographic factors and land use) will 201 be further discussed here to demonstrate their influence.

202
Topographic factors 203 (1) Influence of slope and position along the slope on the spatial distribution of soil pH 204 Severe soil erosion can cause a decrease in the pH value (Schindelbeck et al., 2008). Due to the humid 205 monsoon climate and the high erodibility of purple soil caused by its high content of sandy particles, the 206 pH value is generally lower than in the weathering sediments of redbeds, which have a pH value higher 207 than 8. Table 4 shows that the pH value of the 0-20 cm soil layer tends to decrease from downslope to 208 middle slope to upper slope; this decrease is especially significant at slopes of 20°and 25°(P < 0.05). This ArcGIS software, the spatial distribution map of the soil pH was analysed synthetically (Figures 9 and 10). 221 The result shows that the average pH value varies with aspect of the slope in the study area. The soil pH 222 values on north-and southwest-facing slopes are relatively higher than on slopes of other aspects.   Wang et al., 2011), indicating that the soil pH had good spatial structure in the study area. 256 The effects of topographic factors on soil pH were discussed in this study. The pH of soil is highest on 257 the downslope, followed by the middle slope, and is lowest on the upper slope. Similar results were 258 reported by Tsui (2004), who confirmed that slope, which is involved in the transport and accumulation of 259 solutes, resulted in higher pH. It can be seen that to some extent factors affecting soil erosion have an 260 influence on the change in soil pH. 261 In addition, as we know, the topography is a structure factor influencing the spatial variability of soil pH. 262 In the E-W and N-S directions (0°and 90°, respectively), the soil pH shows high spatial homogeneity 263 because the relief is low and the only land use is farmland. In this study, one rarely acknowledged but 264 important result is that the topography influences the soil pH mainly through the slope and indirectly via 265 the effect of topography on land use patterns. The investigated parameters follow a normal distribution. For pH, the best-fitting variogram model was a 276 spherical one. A practical application of our research results may be that the inclusion of the models we 277 established for application in directional semivariograms in interpolation analysis can improve the 278 reliability of local assessments of the analysed soil pH, thus reducing the cost of the production cycle. In 279 order to reduce production costs, a sampling interval of 80-100 m is recommended for soil pH. The spatial 280 distribution maps based on the kriging interpolation method were successfully applied in soil pH studies. 281 This study shows that soil pH in the study area has moderate spatial autocorrelation, which means that 282 the soil pH is affected by both structural and random factors. This study focused on the spatial variability 283 of soil pH as a result of the interaction of topographic factors, soil and land use patterns. In general, 284 studying the spatial variability of soil pH can provide a theoretical basis for the restoration and                       Table 1 Statistical characteristic values of soil pH.         Table 4 The influence of slope and slope position on soil pH.
The difference between the letters in the same column is significant (P < 0.05), and the letters in brackets indicate significant difference (P < 0.05). Table 4 The influence of slope and slope position on soil pH. The difference between the letters in the same column is significant (P < 0.05), and the letters in brackets indicate significant difference (P < 0.05).