Species and climate databases: Future climate change induced range shifts in Chinese ash along three geographical dimensions

Species and climate databases Occurrence data The specimen data for Chinese ash was obtained from the Chinese Virtual Herbarium (2382 specimens, http://www.cvh.ac.cn/) and Global Biodiversity Information Facility (1426 specimens, http://www.gbif.org/) databases. The duplicate specimens and specimens without locational information or coordinates were removed. Then, the remaining specimens were rasterized onto a raster layer with 10-arcmin resolution. A grid cell was considered a suitable habitat when one or more specimens were located in it. Finally, the binary occurrence map with 10-arcmin resolution was converted into points, and we obtained 293 records with latitudinal and longitudinal values. Climatic variables There are many climatic factors have been demonstrated in previous studies to characterize the climatic niche of plant species or vegetation, such as temperature, precipitation, growing degree days, thermal and moisture variables. For example, 19 bioclimatic factors were used to characterize the global climatic niches of species in the BIOCLIM software and the WorldClim database (Hijmans et al., 2005; Booth, 2018); three factors were used to characterize global climatic niches of vegetation in the Holdridge life zone model (annual biotemperature, potential evapotranspiration and annual precipitation) (Holdridge, 1947); and three factors were used to characterize climatic niches of east Asia plant or vegetation in Kira’s index system (warmth index, coldness index and humidity index) (Kira, 1945). The three sets of climatic factors are widely used in research on the relationship between species/vegetation and climate, at a regional or global scale (e.g. Li et al., 2018; Huang et al., 2018). In this study, we integrated the three sets of climatic factors based on BIOCLIM, Holdridge life zone model and Kira’s index system. An excess of climatic factors can cause overfitting for simulating process, so we only selected 8 of the 19 BIOCLIM variables based on our previous research. A total of 13 climatic factors were used to define the climatic niches of in China, which is sufficient for research on any species at a regional scale, including Chinese ash. The 13 climatic variables are introduced as follows: Temperature factors include annual mean temperature (AMT), max temperature of the warmest month (MTWM), min temperature of the coldest month (MTCM), and annual range of temperature (ART=MTWM - MTCM). Precipitation factors include annual precipitation (AP), precipitation of wettest month (PWM), precipitation of driest month (PDM), and precipitation of seasonality (PSD=Monthly coefficient of variation of precipitation). Growing degree days is correlated with measures of accumulated growing-season warmth (Prentice et al., 1992). Here, we use annual biotemperature (ABT) represents growing degree days, which is calculated by ∑T/12 ( all temperatures below freezing and above 30 °C adjusted to 0 °C, as plants are dormant at these temperatures. ). Thermal factors include warmth index [WI= ∑(T-5), T is >5 °C mean monthly temperature], coldness index [CI=-∑(5-T), T is <5 °C mean monthly temperature]. Moisture factors include potential evapotranspiration rate (PER=58.93×ABT/AP, ABT is annual biotemperature, AP is annual precipitation), humidity index (HI=AP/WI, AP is annual precipitation, WI is warmth index). Current and future climate layers The basic climatic layers of current and future climate scenarios were obtained from the WorldClim database (http://www.worldclim.org/). In the database, the current climatic layers were generated from thin plate smoothing splines using latitude, longitude, altitude, monthly temperature, and precipitation data from the averages of 51-year (1950–2000) climate station records (Hijmans et al., 2005). The future climatic layers were generated from many general circulation models (GCMs) with four representative concentration pathways (RCP2.6, RCP4.5, RCP6.0, and RCP8.5). Here, the climatic layers of future scenarios are averaged by combining seven GCMs to deal with the uncertainty of GCMs under four representative concentration pathways (Huang et al., 2018). The seven GCMs were from seven modeling centers of six countries: BCC-CSM1-1, CCSM4, GISS-E2-R, HadGEM2-AO, IPSL-CM5A-LR, MIROC-ESM-CHEM, and NorESM1-M. The time-period from 2061–2080 was selected as the target future, in which the annual temperature in China will increase from 6.4°C to 8.2–10.6°C and the annual precipitation will increase from 576 mm to 603–623 mm based on ensemble average results of the seven GCMs in contrast to that of 1950-2000.