Abstract
1. Introduction
Global warming has become an indisputable reality, accelerating the global water cycle and significantly altering atmospheric circulation patterns [1,2]. Against this backdrop, the arid–humid climate patterns in many regions worldwide have undergone notable adjustments, particularly in the spatial configuration of climate transition zones. These zones, typically located at the margins between drylands and non-drylands, are characterized by sensitive climate conditions and fragile ecosystems, making them critical frontiers of climate change impacts. For instance, the 100th meridian in the central United States has long been recognized as the dividing line between the humid eastern and arid western regions of North America. Recent studies have shown its eastward shift, indicating a strong climatic response to global warming [3,4]. Similarly, the Sahel belt in North Africa has shown pronounced interannual to decadal variability and spatial expansion, influenced by both the West African monsoon and oceanic thermal anomalies [5]. Given that China’s arid–humid transition zone is influenced by the East Asian and South Asian monsoons as well as the mid-latitude westerlies [6,7,8], it remains unclear how this transition zone has responded to recent climate change and whether it shows shifts analogous to the 100th meridian in the United States?
Situated in a typical monsoon climate region, China is highly sensitive to global climate change [9]. Its vast territory and complex topography create a pronounced spatial gradient, from the humid southeast coastal regions to the hyper-arid northwest interior. Within this range, China’s arid–humid transition zone exhibits a relatively sharp aridity gradient change to separate drylands and non-drylands [10], reflecting the dominant influence of monsoon systems on the country’s climate structure. Moreover, the transition zone closely overlaps with the ecologically fragile agro-pastoral ecotone, where climate variability can trigger far-reaching socio-environmental chain reactions [11,12].
Under the influence of global warming, aridity patterns across China have exhibited significant but spatially heterogeneous shifts [13,14]. Specifically, Northwest China has experienced a notable transition from warm-dry to warm-wet conditions [15,16], whereas North and Northeast China have predominantly suffered from aridification since the 1980s, although recent evidence suggests a potential moderation or reversal of this drying trend [17,18,19,20,21,22]. In contrast, Southeast China has generally maintained a wetting trend, a pattern further corroborated by GRACE-derived terrestrial water storage anomalies [23,24,25]. Superimposed on these regional trends, prior studies have also identified widespread decadal-scale oscillations (~25–30 years) in China’s aridity variability [26,27]. Despite these extensive regional studies, a unified understanding of how these spatially heterogeneous shifts collectively impact the stability of the arid–humid transition zone remains limited.
Remote sensing enables the systematic, large-scale observation of terrestrial hydrological processes [28,29,30]. The integration of these data provides a powerful means to quantify how hydrological systems respond to climate fluctuations, revealing the impacts of regional aridification and wetting trends on water resources. This is facilitated by long-term satellite missions such as GRACE (Gravity Recovery and Climate Experiment) and Landsat whose products reveal the spatiotemporal dynamics of surface water extent and groundwater storage. Collectively, these datasets deliver a spatially continuous, multi-decadal perspective essential for analyzing continental-scale changes in the hydrological cycle.
However, most conventional investigations have primarily focused on the numerical fluctuations of aridity indices [11,31] or static climatic boundaries based on long-term averages [7,32], often overlooking the intrinsic stability of the transition zone. To address this, we introduce a Shannon entropy-based metric to quantify the stability of the arid–humid transition, enabling a dynamic identification of the ‘Arid–Humid Divide’. Furthermore, we integrate multi-source remote sensing data to reveal the divergent and lagged hydrological responses of surface water and groundwater to aridity changes—a topic that has been less explored in continental-scale assessments. The main contributions of this study are summarized as follows:
A robust ~32-year oscillatory cycle in China’s aridity patterns is identified, providing a new perspective on long-term climate variability.
A distinct ‘Arid–Humid Divide’ is identified using a novel Shannon entropy-based metric, revealing a stable climatic boundary governed by monsoon-driven moisture transport.
We reveal significant divergent hydrological responses, where surface water expands rapidly during the current wetting phase (post-2010), while groundwater recovery lags significantly.
2. Materials and Methods
2.1. Study Area
China is geographically divided into seven major regions (Figure 1): Northeast China (NEC), North China (NC), Central China (CC), East China (EC), South China (SC), Southwest China (SWC), and Northwest China (NWC). China’s terrain features a three-step staircase pattern descending from the Tibetan Plateau in the west to the eastern plains, shaping distinct arid–humid climatic zones. In winter dry seasons, the development and breakdown of the Siberian High and its associated anticyclone drives the westerlies across East Asia, carrying cold, dry air and aeolian dust from high-latitude Asia, resulting in a cold and dry climate in east Asia [33,34,35]. In summer, the northward shift of the westerly belt transports limited moisture to northwestern China, while the East and South Asian summer monsoons carry warm and humid air from the Pacific and Indian Oceans into eastern and southern China, resulting in a markedly humid climate [36,37]. The boundary of monsoon roughly follows the Greater Khingan–Helan–Qilian Mountains and southeastern Tibetan Plateau, forming a transitional zone where arid-humid conditions are sensitive to monsoon variability [36].
Figure 1. Geography of China. Thin orange arrows indicate the westerlies, while thick blue and orange arrows represent the winter and summer monsoons, respectively. The positions of the westerlies, Siberian High, East Asian monsoon (EAM), and South Asian monsoon (SAM) are adapted from Yao, et al. [38] and Wang, et al. [39].
2.2. Datasets
To capture fine-scale variations in arid–humid climate patterns, this study used high-resolution monthly data on precipitation (PRE), potential evapotranspiration (PET), and mean temperature (TMP) from the National Tibetan Plateau/Third Pole Environment Data Center (TPDC, http://data.tpdc.ac.cn (accessed on 20 January 2024)). These datasets, downscaled from the CRU product [40,41,42,43], have a spatial resolution of approximately 1 km and cover the period from 1950 to 2022. PET was estimated using the Hargreaves method [40,41]. To assess the robustness of the CRU-based aridity zone classification, we additionally employed PRE and PET data from the TerraClimate dataset for validation [44v
Area Changes of Aridity Zones
From 1950 to 2022, China’s aridity zones experienced substantial spatial adjustments, displaying a persistent gradient of increasing dryness from the humid southeastern coast toward the arid northwestern interior (Figure 3). The spatial distribution and temporal evolution of these zones align well with previous findings derived from CRU datasets and meteorological station observations [51], confirming the reliability of our high-resolution gridded analysis. Despite these widespread shifts, the Qinling Mountains remained a relatively stable climatic boundary. Owing to their high elevation and east–west orientation, the mountains effectively block the southward intrusion of cold continental air in winter and limit the northward expansion of the East Asian summer monsoon, thereby maintaining a consistent climatic divide even as surrounding regions undergo notable changes.
