Open-Top Chambers (OTCs), widely used as field warming facilities, not only generate warming effects but also reduce water availability through rain interception effect. While OTCs are extensively applied to explore the effects of warming on seedling recruitment which act as a fundamental driver of population dynamics, the potential experimental biases introduced by the rain interception effect remains largely unexplored. This knowledge gap raises concerns about the interpretation of warming experiments using OTC facilities. We conducted warming experiments in alpine meadows on the Tibetan Plateau using OTCs. Three treatments were applied: control (no OTC), warming alone (W, +2 °C without OTC precipitation interception), and warming with decreased precipitation (WDP, +2 °C with OTC precipitation interception). Focusing on 50 plant species across various functional groups, we disentangled the independent effects of warming and decreased precipitation caused by OTC interception on seedling emergence, survival, and establishment. We applied structural equation modeling (SEM) to analyze the relationships between emergence timing, rate, survival, and establishment success. Our study demonstrates that warming (W) and warming with decreased precipitation (WDP) treatments accelerated seedling emergence but decreased survival and establishment rates at the community level. The WDP reduced survival by 20% compared to W. The SEM analysis revealed that earlier emergence and higher emergence rates under W indirectly reduced establishment success, while WDP directly suppressed survival by limiting water availability. Furthermore, W increased the biomass of grasses and perennials, whereas WDP significantly reduced their biomass. In contrast, annuals and non-grasses showed no significant response to either treatment.

Synthesis and applications:

The OTC-induced precipitation reduction can overestimate the impacts of warming on seedling establishment success, with species and functional group-specific responses. Therefore, interpreting data from OTC warming experiments requires careful consideration of precipitation interception effects. To mitigate this issue, future studies should optimize experimental designs to minimize precipitation interception, incorporating methods such as collecting and redistributing precipitation to better support seedling water availability. Furthermore, to enhance ecosystem resilience and support biodiversity in the context of climate change, conservation strategies should prioritize species and functional groups that are especially vulnerable to changes in warming and precipitation.

Seedling emergence monitoring

Seedling recruitment was monitored from March 2022 until October 2022 in all experimental plots. Seedling emergence was recorded every two days, with time of seedling exposure to the soil surface defined as the initial emergence time. Each emergent seedling was labeled with a colored label and inserted into the soil beside the seedling. Different colored labels were used each month to allow seedling survival to be monitored. Seedlings began to appear in May and continued to emerge until July, with very few emergences afterward. We recorded the seedling mortality rate each month. Additionally, we calculated the seedling emergence rate, mean emergence time , emergence index, survival rate, and establishment rate.

Seedling growth trait measurements

Seedlings were manually thinned out three months after the first emergence. Five healthy, pest- and disease-free, well-grown seedlings of each species were retained for seedling trait measurements. Leaf and root traits were measured before drying. Two weeks after thinning, five fully expanded young leaves from each species were selected, and their leaf area was measured using a leaf area meter (Li-300, Li-COR, Lincoln, USA). Leaves were then dried for 72 hours at 60 °C and weighed. Specific leaf area (SLA, cm²/g) was calculated as the ratio of fresh leaf area to dry weight. At the end of the growing season, the aboveground parts and roots of all plants were harvested. The entire root system of each plant was obtained using a root scanning system (WinRHIZO Pro 2009b), and root traits were measured, including total root length (TRL) and root average diameter (RAD). Specific root area (SRA, root surface area/root weight), specific root length (SRL, total length of absorbing roots/total dry weight), and root tissue density (RTD, total dry weight of absorbing roots/volume) were calculated using the respective formulas. We excluded coarse roots (> 2 mm in diameter). Finally, aboveground biomass and root biomass were measured after drying to a constant weight and total biomass was calculated.