Whether current hypotheses for geographic patterns of species richness (SR) have a strong explanatory power for the Tibetan Plateau (TP) with extreme climatic conditions remains unclear. In comparison with the classic “water–energy dynamics hypothesis”, the unique climate factors (e.g., extreme low temperature and low oxygen partial pressure) on the TP likely significantly affect the spatial variation of SR. Here, we investigate geographic patterns and determinants of SR on the TP through a systematic field investigation. We systematically analyzed a total of 2,013 plant communities covering 11 different vegetation types on the TP.  The SR per 400 m² in the forests and shrubs and that per 1 m² in alpine grasslands and deserts was 62.76 (±1.80 SE), 44.53 (±7.57 SE), 16.84 (±0.39 SE), and 3.62 (±0.55 SE), respectively. Unique climate factors, such as extremely low temperature, mean diurnal temperature, and oxygen partial pressure, act synergistically with water–energy dynamics and influence the spatial pattern of SR on the TP. Our findings provide novel insights into the mechanisms underlying the spatial variation in plant diversity, especially on plateaus and in high-latitude regions. Our findings and the SR map with 1 km resolution provide important benchmarks for biodiversity conservation and may help to improve predictions of the effect of climate change on biodiversity.

Field sampling was conducted during the high growth period of mid-July to late August in 2018, 2019, and 2020. We divided the entire Tibetan Plateau according to the latitude and longitude into 1,000 grids with equal area (0.5° × 0.5°). For the field investigation of the alpine grasslands and deserts, we selected the dominant plant communities via visual observation in each grid and randomly set up three 1 m × 1 m plots located at least 20 m apart (most commonly plot-size for these grasslands). If two or more types of plant communities were abundant in a given grid, the vegetation type was then randomly selected (the same rules apply to forests and shrubs). The “shoot presence” method was used to record the presence of species, whereby an individual is recorded as present if any part of its stem or leaf is inside the plot. For the forests and shrubs in the southeast, three 20 m × 20 m plots located at least 100 m apart were randomly established in each grid.  We recorded all vascular plants (including juveniles and seedlings) in these plots including tree, shrub, and herb species.  A total of 2,013 field plots were investigated. The depopulated zone was not investigated due to access limitations.