Answers are grounded in this paper and cite their sources.
Ground temperature varies annually as heat input at the surface propagates by conduction, but rainfall infiltration and other processes can superimpose temperature changes that conduction cannot explain. To determine the depth to which such changes appear, ground temperature was monitored hourly at z = 3, 6, 9 and 12 m for approximately ten months, from October 2025 to July 2026, in a sandy vadose zone with a water table approximately 15 m deep. The annual variation was fitted with a cosine function to remove the conductive annual component, and PELT (Pruned Exact Linear Time), a changepoint detection method, was applied to the residuals. Fourteen changepoints were detected at z = 3 m, whereas only three, two and three changepoints were detected at z = 6, 9 and 12 m, respectively. The temperature changes unexplained by heat conduction therefore concentrate at z = 3 m and diminish markedly at greater depths.
No takes yet. Share an insight, caveat, or question.
Koichi Nakamura (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: