Introduction
The Quaternary period, comprising the Holocene and Pleistocene Epochs, spanned the last ~2.6 million years (Myr) and has been characterized by distinct climatic oscillations, especially alternating glacial and interglacial cycles in the Northern Hemisphere (Shackleton & Opdyke, 1973; Pillans & Gibbard, 2012; Elias, 2013). The glacial cycles co-varied with, and probably profoundly affected, other aspects of the climate, including the intensity of the Asian monsoon, even in unglaciated regions (An, Kutzbach, Prell, & Porter, 2001; Jiang , Lang, Tian, & Guo, 2011; Liu et al., 2018). Climate fluctuations during the Quaternary glaciations led to dramatic changes in the geographical distribution, genetic structure, and population demography of plant species (Hewitt, 2000; Wang et al., 2009; Jia et al., 2012; Liu et al., 2018).
Quaternary climate change is known to have strongly affected the distributions of plants in Northern China (Tian et al., 2009; Qiu, Fu, & Comes, 2011; Liu, Sun, Ge, Gao, & Qiu, 2012). Northern China is dominated by deserts, which developed in the Quaternary to due to sustained orogenesis of the Qinghai-Tibetan Plateau and surrounding areas that ultimately enhanced aridity within the region (Wu, 1992, 2002; Meng & Zhang, 2011). The climatic processes that gave rise to the deserts also shaped regional plant diversity and yielded highly complex demographic histories of native species (Ge, Yu, Yuan, Huang, & Yan, 2005; Ge et al., 2011; Yang, Fang, Dong, Peng, & Li, 2006). In particular, many plants within the deserts of Northern China underwent adaptive and demographic change in response to cold, arid conditions, which occurred during the intermittent glacial periods and which may be primary mechanisms explaining modern distributions (e.g., Comes & Kadereit, 1998; Hewitt, 2000; Ge et al., 2011; Zhang & Zhang, 2012; Su & Zhang, 2013; Zhang, Zhang, & Williams, 2014; Xu & Zhang, 2015; Shuyskaya, Toderich, Gismatullina, Rajabov, & Khohlov, 2017; Merklinger et al., 2019; El-Tayeh, Galal, Soliman, & Zaki, 2020). For example, Su & Zhang (2013) proposed that the onset of aridity during Quaternary glacial periods was a primary driver of population processes and structures in Nitraria sphaerocarpa Maxim. (Nitrariaceae, Sapindales), and Xu & Zhang (2015) revealed that periods of cold, arid conditions during the Pleistocene glaciations resulted in the genetic differentiation and demographic structuring in Atraphaxis frutescens (L.) K.Koch (Polygonaceae).
Nevertheless, these prior studies on population histories of desert plants of Northern China have focused primarily on woody species. Studies on herbs of the region are largely lacking, and to our knowledge, the only such study is on Delphinium naviculare W. T. Wang (Ranunculaceae) (Zhang & Zhang, 2012), which is endemic at mid-elevations within the Tianshan Mountains of Xinjiang Province of China (Wang & Warnock, 2001). Herbaceous plants of the deserts of Northern China merit further study because they comprise a vital component of desert plant communities, and studies focusing on dominant grass species are especially warranted (Meng & Zhang, 2011). Herbaceous plants may have been more sensitive to Quaternary climatic oscillations because they differ markedly from woody species in their response to cold, often via death of aboveground biomass as either part of an annulment or perennial life cycle. Moreover, dominant species likely achieved their abundance due to their responses to the glacial cycles.
In modern times, deserts and semi-deserts, such as in Northern China, are extremely fragile ecosystems, the stability of which impacts global environmental conditions and influences climate change (Su, 2013). Although deserts typically have sparse vegetation, plants are critical for maintaining their integrity. In Northern China, the desert grassland ecosystems in particular are becoming rapidly degraded due to long-term overgrazing and desertification and, simultaneously, desert is encroaching on arable land within the region (Li, Liu, & Wang, 2004; Hanafi & Jauffret, 2008; Wang, 2009; Peters, Yao, Sala, & Anderson, 2012; Deng, Zhang, & Shangguan, 2014; State Forestry Administration, 2015; Wang & Zhou, 2015; Wei, Wang, & Niu, 2020). These desert grasslands represent a large area within China and adjacent countries and occur at both low and high elevations. Dominant plant species within the grasslands are often psammophytes, which have special adaptations to resist being buried by sand and to tolerate having periodically exposed roots. Simultaneously, these plants help to anchor sands in place and prevented wind erosion, and are thus critical for promoting environmental stability within their ecosystems (Ma, 1991; Pan, 2006; Zhou, Yuan, & Jing, 2011).
In this study, we focus on one psammophyte, Psammochloa villosa(Trin.) Bor, which comprises a monotypic genus in tribe Stipeae of Poaceae. This species, commonly called sand whip, is a perennial rhizomatous herb that is primarily distributed in the desert grasslands of northwestern China, especially the Inner Mongolian Plateau, the Hexi Corridor of Gansu, central and Northern Ningxia, and Northern Shanxi (Liu, 1985; Ma, 1994; Huang, 2003). It also occurs in Mongolia. The flowering and fruiting period of P. villosa is from September to November, and the seeds are 5 - 7 mm long with an average weight of 5.507 ± 0.053 mg (Huang, 2003). Such light seeds are likely dispersed by high winds that occur throughout much of its natural desert habitat. Nevertheless, seedlings are rarely observed (Zhu, 2005).Psammochloa villosa is known to have high resistance to drought, cold, alkaline soils, disease, wind, and burial by sand, all of which likely represent evolutionary adaptations and facilitate its survival in grassland and dune areas (Lu, 1987; Huang, 1995, 1997; Wu & Phillips, 2006). Previous research on P. villosa has been mainly focused on studying its anatomy, embryology and microbiology (e.g., Dong & Alaten, 1999; Huang, Dong, & Gutterman, 2004; Wang et al., 2011; Zhang et al., 2017; Lv et al., 2018), with few studies so far focused on molecular markers (e.g., Li & Ge, 2001). The species is ecologically widespread at low and high elevations (900-2900 m) (Kuo, 1987; Wu & Phillips, 2006).
In our present study, we investigated the influence of climate aridification and oscillations on the genetic structure and evolutionary processes of P. villosaduring the Quaternary in north-western China using amplified fragment length polymorphism (AFLP) markers and ENM. We used AFLPs because they remain extremely efficient for investigating genetic diversity, genetic structure, and population demography due to their high levels of polymorphism (Wang, Wang, Liu, Yang, & Chen, 2008), their reproducible, reliable results that are unaffected by the developmental stage of plant materials, and their universality among plant species. In addition, they have been used to resolve genetic structures and population demography in many diverse grass species such as Oryza sativa , Leymus racemosus , Orinus thoroldii and O. kokonoricus (Zhang & Jia, 2002; Sim, 2005; Li, 2015; Cai, 2017; Liu, Harris, Gao, Su, & Ren, 2019). Our main objectives were to (1) analyze the genetic structure from 43 populations of P. villosa from Inner Mongolian Plateau using an AFLP dataset, (2) test whether historical genetic divergence occurred among populations in response to Quaternary climate oscillations, and (3) evaluate the abiotic factors that are most influential in driving the distributions of P. villosa . Moreover, because no assessment of the conservation needs of P. villosa had previously been accomplished, we also performed a preliminary assessment based on Extent of Occurrence (EOO) with interpretation according to guidelines of the International Union for the Conservation of Nature (IUCN). We believe that, taken together, our results can provide a scientific basis for improved protection and sustainable utilization (e.g., as forage) of P. villosa within the fragile desert grassland ecosystems of Northern China.