Snow is an important factor controlling vegetation functions in high latitudes/altitudes. However, due to the lack of reliable in-situ measurements, the effects of snow on vegetation phenology remains poorly understood. Here, we examine the effects of snow cover duration (SCD) on the start of growing season (SOS) for different vegetation types. SOS and SCD were extracted from in-situ carbon flux and albedo data, respectively, at 51 eddy covariance flux sites in the northern mid-high latitudes. The effects of SCD on SOS vary substantially among different vegetation types. For grassland, preseason SCD outperforms other factors controlling grassland SOS. However, for forests and cropland, the preseason air temperature is the dominant factor in controlling SOS. Preseason SCD mainly influences the SOS by regulating preseason air and soil temperature rather than soil moisture. The CMIP6 Earth system models (ESMs) fail to capture the effect of SCD on SOS. Thus, Random Forest (RF) models were established to predict future SOS changing trends considering the effect of SCD. For grassland and evergreen needleleaf forest, the projected SOS advance rate is slower when SCD is considered. These findings can help us better understand impacts of snow on vegetation phenology and carbon-climate feedbacks in the warming world.

中文翻译：

---

积雪持续时间延迟了北半球草地春季返青的时间


雪是控制高纬度地区植被功能的重要因素。然而，由于缺乏可靠的现场测量，雪对植被物候的影响仍然知之甚少。在这里，我们研究了积雪持续时间 (SCD) 对不同植被类型的生长季节开始 (SOS) 的影响。 SOS和SCD分别从北部中高纬度地区51个涡流协方差通量站点的原位碳通量和反照率数据中提取。 SCD 对 SOS 的影响在不同植被类型之间差异很大。对于草原而言，季前 SCD 优于控制草原 SOS 的其他因素。然而，对于森林和农田来说，季前气温是控制 SOS 的主导因素。季前 SCD 主要通过调节季前空气和土壤温度而不是土壤湿度来影响 SOS。 CMIP6 地球系统模型 (ESM) 未能捕捉到 SCD 对 SOS 的影响。因此，考虑SCD的影响，建立随机森林（RF）模型来预测未来SOS的变化趋势。对于草原和常绿针叶林，考虑 SCD 时预计的 S​​OS 推进速度较慢。这些发现可以帮助我们更好地了解降雪对变暖世界中植被物候和碳气候反馈的影响。