**1. Introduction**

Evidence suggests that global temperatures have continued to rise over the last two decades and will continue to warm over the next three decades [1], which affects many ecosystems [2–4]. Alpine ecosystems are considered to be particularly sensitive to climate change because of harsh natural environments [5–8]. Therefore, accurate assessment of the impacts of climate change in alpine ecosystems is essential.

Land surface phenology (LSP) is defined as the seasonal change pattern of surface vegetation obtained from remote sensing observations, which is usually used to describe

**Citation:** Liu, Y.; Zhou, W.; Gao, S.; Ma, X.; Yan, K. Phenological Responses to Snow Seasonality in the Qilian Mountains Is a Function of Both Elevation and Vegetation Types. *Remote Sens.* **2022**, *14*, 3629. https://doi.org/10.3390/rs14153629

Academic Editor: Xinghua Li

Received: 27 June 2022 Accepted: 25 July 2022 Published: 29 July 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

the start, end and length of the vegetation growing season [9–11]. Unlike traditional groundbased observations that can record dates of budburst or flushing, LSP is used to describe the full process of regional greening. This may not correspond to a specific vegetation event but can provide a rapid understanding of the key stages of the overall greening of a region [9]. LSP is one of the most sensitive and easily observable nature features when analyzing the response of vegetation to climate change [12], and exploring its changes provides an important avenue for studying severe climate anomalies. Climate change can interfere with vegetation germination. In addition, changes in LSP may have a significant impact on carbon and water cycles [13–15]. An integrated analysis of the impact of climate change on land surface phenology is important for understanding the impact of future climate change.

Driving factors for trends in land surface phenology have frequently been attributed to changes in temperature and precipitation [16–18]. However, as one of the typical features in areas with stable snowpack, changes in snow seasonality can also cause changes in land surface phenology [19,20]. Specifically, Snow accumulates or melts on soil and vegetation, which can directly alter the hydrothermal conditions under which vegetation grows and develops (Figure 1). Snow directly affects near-surface temperatures in several ways. Snow cover in winter insulates the soil from cold air and maintains soil temperature [20]. Soil temperature is higher than air temperature in early spring under snow cover [21]. Due to the insulating nature of snow, temperature no longer has a direct effect on vegetation [22]. The timing of snowmelt is sometimes a more important factor in the growing season than air temperature [23]. These regulatory effects of snow accumulation and snowmelt on surface temperature have important implications for land surface phenology and soil moisture content. Snow is the main source of freshwater for alpine vegetation, as snowmelt provides the necessary moisture for vegetation to sprout in the form of soil water [24–27]. Snow cover protects vegetation and soil from harsh natural hazards such as wind erosion, freezing damage, and intense solar radiation, which often occur in high-elevation mountains and can seriously hinder vegetation growth [28–31]. It has also been demonstrated that winter snow can indirectly affect the carbon sequestration capacity of vegetation by altering community structure and activity of soil microorganisms [32].

**Figure 1.** Schematic diagram of snow cover effects.

Although there are many studies focusing on the snow cover effect on land surface phenology, they have inconsistent results [33–38]. In some warm and dry regions, winter snow cover has been shown to favor early spring vegetation germination and prolong the growing season, while the opposite is true in cold, humid regions [35,36], clearly demonstrating that the effect of snow cover varies under different hydrothermal conditions. However, in Inner Mongolia, China [38], and the French Alps [39], two regions with different natural conditions, snow has a similar negative effect on land surface phenology:

late snowmelt delays vegetation emergence. Grass is the predominant vegetation type in these two regions, and the mechanism for the effect of snow on the same species should be consistent. Even so, the impact of snow on a single vegetation type in the same area changes with terrain [33,34]. In short, the influence of snow seasonality on land surface phenology is determined by the coupling of multiple factors such as water and heat conditions, vegetation type and terrain. There remains a lot of views on specific conclusions, especially the effect of snow as a function of elevation gradients and vegetation types, and additional study is required. Additionally, the spring phenology extracted using NDVI is often affected by preseason snow, which may lead to inaccurate conclusions [40,41]. Unraveling the effects of snow on land surface phenology can help identify the mechanisms of change in land surface phenology.

The Qilian Mountains area (QLMA) is an essential part of the Qinghai–Tibet Plateau and is considered as an important ecological barrier in western China. The QLMA has large elevation differences, a variety of vegetation types and significant climate change. In addition, the relatively small area of QLMA could mitigate the impact of spatial differences on the results. It is an ideal laboratory for studying the vegetation response to climate change. Studies on vegetation phenology for the Qilian Mountains region are still limited. We quantified the response of land surface phenology metrics to different snow seasonality metrics from 2002 to 2021. Annual snow seasonality metrics include the first snow day (FSD), last snow day (LSD), snow season length (SSL) and total number of days with snow cover (SCD). Land surface phenology metrics include the start of the growing season (SOS), end of the growing season (EOS) and length of the growing season (LOS) estimated by the normalized difference phenology index (NDPI), which was proven to be an accurate vegetation index for estimating land surface phenology at high elevation [42]. More comprehensive metrics and more reliable vegetation indices enhance the richness and accuracy of our conclusions. In addition, we chose two representative influencing factors, elevation and vegetation type, which may be helpful to explain the complex effects of snow seasonality.

To better understand the relationship between snow and land surface phenology, the following three research questions are proposed:


#### **2. Materials and Methods**
