Soil organic carbon was more strongly linked with soil phosphate fixing capacity than with clay content across 20,000 agricultural soils in Japan: a potential role of reactive aluminum revealed by soil database approach

• Published in: Soil science and plant nutrition (Tokyo) Vol. 67; no. 3; pp. 233 - 242
• Main Authors: Matsui, Kayo, Takata, Yusuke, Matsuura, Shoji, Wagai, Rota
• Format: Journal Article
• Language: English
• Published: Kyoto Taylor & Francis 04.05.2021; Taylor & Francis Ltd
• Subjects: Agricultural land; Aluminum; andosols; Arable land; Carbon storage potential; Clay; Land use; Land use management; lowland soils; Organic carbon; Organic soils; phosphate adsorption coefficient (pac); soil organic matter (som); Soil properties; Soil temperature; Soil testing; Soil types; Temperature dependence; Variation; Volcanic soils; Japan
• ISSN: 0038-0768; 1747-0765
• DOI: 10.1080/00380768.2021.1902750

Identifying the soil properties that control soil organic carbon (SOC) storage is a critical step to monitor and manage SOC at regional and global scales. While clay content has been commonly used to explain SOC variation and to model SOC dynamics, recent studies point to the significant control of extractable metal phases such as pyrophosphate extractable Al (Al p ) for a range of soil types. Phosphate adsorption coefficient (PAC), soil's capacity to adsorb phosphate, is known to show strong positive correlation with Al p . The soil database maintained by the national farmland fixed-point survey contains basic soil attributes including SOC and PAC and covers approximately 20,000 sites that belong to Andosols and Lowland soil great groups, representing roughly 70% of the arable land in Japan (MAT range from 5.3 to 22.3 °C). Using the database, we found that PAC showed more significant positive correlation with SOC compared to clay and silt-clay contents for both Lowland soils and Andosols. The OC:Al p ratio was estimated from the slope of SOC:PAC regression line for all soil groups and great groups ranged from 3.35 to 10.5 (mol:mol), which was within the range shown in previous studies on volcanic and spodic soils. The ANCOVA test further revealed that the SOC-PAC relationship was significantly affected by soil temperature regime in both soil group and soil great group levels (both p < 0.0001). All four soil groups within Lowland soils showed a similar temperature-dependent trend in the SOC:PAC ratio with little variation among the groups. The soil groups within Andosols, however, had no clear temperature trend, suggesting the influence of other factors that remain to be elucidated in future. We concluded that PAC was the more effective indicator of SOC than clay content for the major arable soils in Japan and that the PAC-SOC relationship was partly controlled by soil temperature and soil group. Further study to assess the role of Al p and the contribution of other factors such as land use and management type is important as major portions of the SOC variation was not explained by PAC.