Rainfall memory and spring atmospheric demand predict herbaceous biomass production in Mediterranean ecosystems
Marcelo Sternberg, Shay Adar, Darya Perry, Danila Beliavskii, Jaime KigelAbstract
Aboveground herbaceous biomass production in Mediterranean ecosystems is often related to precipitation amount, yet similar rainfall totals can produce contrasting biomass. Rainfall totals alone therefore do not capture how water supply is experienced by plants through time. Long‐term experimental evidence integrating rainfall persistence and continuity, climatic legacies and atmospheric demand remains scarce, limiting prediction under increasingly warm and irregular Mediterranean climates.
We analysed a long‐term record of herbaceous biomass production spanning 2002–2026 from a rainfall‐manipulation experiment in a Mediterranean shrubland. Treatment‐specific rainfall records were combined with meteorological data and Random Forest modelling. A Season‐Specific Climate Model evaluated hydrological, thermal and atmospheric controls across autumn, winter and spring using year‐grouped nested cross‐validation.
Rainfall memory, the short‐term influence of recent rainfall on growing conditions, emerged as a central predictor of biomass production. Greater winter and spring rainfall memory was associated with higher biomass, whereas longer winter dry spells were associated with lower biomass. Favourable rainfall conditions during the previous growing season were also strongly associated with greater current biomass, revealing a climatic legacy across growing seasons. In contrast, high spring mean daily vapour pressure deficit (VPD) and high mean temperature were associated with reduced biomass. Diurnal temperature range also contributed to prediction, but its relationship with biomass was context dependent rather than consistently indicative of drought stress.
These findings show that rainfall effectiveness depends on more than seasonal totals. Biomass integrates the persistence and continuity of current rainfall, conditions inherited from the previous growing season, and the atmospheric environment during late‐season growth. This supports a compound‐drought framework in which water supply limitation and atmospheric demand become important at different stages of the growing season.
Synthesis . Our study provides rare long‐term experimental evidence that Mediterranean herbaceous biomass reflects climatic processes operating across multiple timescales. By identifying when rainfall memory, climatic legacy and spring atmospheric demand matter most, we help close a key gap in precipitation–productivity research and move drought interpretation beyond rainfall amount alone. Projections based only on seasonal rainfall totals may therefore underestimate ecosystem vulnerability as rainfall becomes more irregular and spring atmospheric demand rises.