Objectives  The stability of soil aggregates and the related potassium (K) release were studied, in an attempt to provide scientific basis for guiding the soil fertilization in the tea garden and promoting the sustainable development of the tea garden ecosystem.

Methods  Soil samples were collected in 0–15 and 15–30 cm deep of undisturbed soil layers in the tea garden of 5, 10, 15 and 30 years old in Ya'an City, Sichuan Province. Each soil sample was separated into four particle sizes of > 2 mm, 0.25–2 mm, 0.053–0.25 mm and < 0.053 mm by wet screen method. The stability of the four sizes of aggregates and the contents of soil readily available potassium (RAK) and slow available potassium (SAK) inside were analyzed. Meanwhile, the sodium tetracenyl borate (NaTPB) method was utilized to explore the release characteristics of available potassium.

Results  The soils in all the investigated tea plantations were mainly composed of large aggregates (size > 0.25 mm), with the mass ratio of 75.87%–95.75%, however, the mass ratio tended to decrease with the increase of plantation ages significantly (P < 0.05), and the mean weight diameter (MWD) of soil aggregates in the two soil layers decreased obviously. The contents of RAK and SAK in different sizes of aggregates at 15 and 30 years after tea planting were higher than those at 5 and 10 years. The content of soil RAK in different sizes of aggregate was similar at tea plantation age of 5 and 10 years, while that was affected by the content of large aggregate at tea plantation age of 15 and 30 years. On the contrary, the content of SAK was higher in microaggregates (size < 0.25 mm) than that in larger aggregates. The cumulative release amount of available potassium from different aggregates varied in the range of 102.3–236.5 mg/kg. Furthermore, the release of available potassium from soil showed a trend of rapid release in the younger tea gardens and stable release in the elder ones, and the release process could be well described using the parabolic diffusion model. Specifically, the cumulative release of available potassium from soil aggregate at 15 and 30 years old tea plantations was significantly higher than those at 5 and 10 years old tea plantations.

Conclusions  Large aggregates (size > 0.25 mm) are dominant in all the tested tea plantations, but the stability of large soil aggregates tends to decrease with the increase of tea plantation age, leading to the release of the RAK and SAK contained in the large aggregates, especially in the 15–30 cm soil layers. So, the high soil K levels in tea plantation at 15 a and 30 a reflect the consumption of soil K pool, which is not good for the sustainable use of tea plantation.