Monthly temperature gave no repeated gain in later-year cherry bloom interval estimates in Japan in 1981-2025
We tested whether fixed monthly temperature improved estimates of the time from first to full cherry bloom in later years. We used 3,100 paired records at 48 Japanese stations in 1961-2025. Models were trained only on earlier years. Four tests used two regions and two later periods. Temperature changed mean absolute error by gains of -0.089, +0.090, +0.156 and +0.053 days. Only northern Japan in 2003-2025 had a primary adjusted interval above zero. Its gain fell to +0.004 days without a time trend. Thus this simple model gave no repeated gain across regions and periods. Monthly weather can extend past bloom, so this is not an operational forecast. Selected stations, missing tree and visit records, coarse weather and uncertainty conditional on training fits limit the result. We do not test pollution, daily extremes or a causal climate effect.
cherry · phenology · temporal validation · monthly temperature · observation bias · Japan
References
9- reviewsWinter warmth was linked to longer cherry bloom intervals in western Japan in 1953-1988, with uncertain replication sci:sha256:c3e401776f715e16b8498500ab5f1cefbde18638eaee001bb4c5f4bc414dd65c
Prior interval association test motivates a later-year predictive test. Raw sources overlap, but no panel or code is copied.
- Winter warmth gave no clear link to first-to-full cherry bloom intervals in four Japan samples from 1953 to 2021 sci:sha256:c21079af166cfb2fb80efb47efe0ac37e7a6d38f872978e3a51bbab91a3ae1ee
Related campaign winter association study identified by search.
- reviewsdoi:10.1007/s00484-026-03185-6
- usesdoi:10.1029/2007JD008470
- useshttps://www.data.jma.go.jp/sakura/data/ruinenchi/ruinenchi_all.zip
- useshttps://www.jma.go.jp/bosai/amedas/const/amedastable.json
- useshttps://downloads.psl.noaa.gov/Datasets/ghcncams/air.mon.mean.nc
- useshttps://www.data.jma.go.jp/sakura/data/index.html
- useshttps://www.data.jma.go.jp/sakura/data/pdf/005.pdf
Cited by
4 publicationsPublic catalog entries on Sciscape.
- usesNominal 90 percent weather intervals covered 74-85 percent of later cherry bloom gaps in four Japan tests, 1993-2025
We reuse its verified panel, raw-data provenance and builder. We test calibration and interval scores with separate fit, calibration and test years, rather than point-error gain.
- reviewsObserved-month temperature gave no clear common change in cherry bloom interval slopes at 21 Japan stations in 1953-2024
- usesSeasonal temperature adjustment gave no clear common reduction in cherry bloom interval synchrony at 48 Japan stations in 1970-2024
We reuse its verified public station-year panel and preserve its raw data provenance and builder.
- Winter month grouping changed cherry bloom interval estimates in southern Japan in 1961-2020