Next Article in Journal
Effect of Vegetation on the Intensity of Low-Level Coastal Jets: The Case of Tenerife
Previous Article in Journal
Assessing Climate Impact on Heritage Buildings in Trentino—South Tyrol with High-Resolution Projections
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Atmosphere
Volume 16
Issue 7
10.3390/atmos16070800
atmosphere-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Mitigating Model Biases in Arid Region Precipitation over Northwest China Through Dust–Cloud Microphysical Interactions

by
Anqi Wang
1,
Xiaoning Xie
2,
Zhibao Dong
1,*,
Xiaoyun Li
1,
Ke Shang
3,
Xiaokang Liu
1 and
Zhijing Xue
1,*
1
School of Geography and Tourism, Shaanxi Normal University, Xi’an 710119, China
2
State Key Laboratory of Loess Science, Institute of Earth Environment, Chinese Academy of Sciences, Xi’an 710061, China
3
School of Civil Aviation, Xihang University, Xi’an 710077, China
*
Authors to whom correspondence should be addressed.
Atmosphere 2025, 16(7), 800; https://doi.org/10.3390/atmos16070800
Submission received: 19 May 2025 / Revised: 19 June 2025 / Accepted: 24 June 2025 / Published: 1 July 2025
(This article belongs to the Section Atmospheric Techniques, Instruments, and Modeling)
Browse Figures
Versions Notes

Abstract

Accurate projection of future climate trends in arid regions critically depends on reliable precipitation simulations. However, most Coupled Model Intercomparison Project Phase 6 (CMIP6) models exhibit systematic overestimations of precipitation in Northwest China, a bias that undermines the credibility of climate projections for this vulnerable region. This persistent bias likely stems from the omission of key physical processes in traditional models. In this study, we incorporate a dust–ice-cloud interaction scheme into the Community Atmosphere Model version 5 (CAM5) model to investigate its role in regulating precipitation over dust-rich arid regions. This physical mechanism, which is rarely included in conventional models, is particularly relevant for Northwest China where dust aerosols are abundant. Our results show that accounting for dust-induced ice nucleation leads to a significant reduction in total precipitation, especially in the convective component, thereby alleviating the longstanding wet bias in the region. These findings underscore the critical importance of dust–ice-cloud interactions in simulating precipitation in arid environments. To improve the accuracy of future climate projections in Northwest China, climate models must incorporate realistic representations of dust-related microphysical processes.

1. Introduction

Simulating precipitation over arid and semi-arid regions remains one of the most persistent challenges in current climate modeling, due to the complex interplay between surface dryness, atmospheric stability, and aerosol–cloud interactions [1,2]. In dryland regions such as Northwest China, where convective precipitation is infrequent yet critically important, even minor shifts in thermodynamic conditions can lead to large differences in rainfall occurrence and intensity [3,4]. Precipitation here is modulated by a range of tightly coupled processes, including land–atmosphere exchange, moisture availability, and cloud microphysics. Among these, interactions involving atmospheric aerosols—particularly their roles in cloud formation and convective suppression—are increasingly recognized but remain poorly constrained in global climate models [5,6]. This makes precipitation projections over arid zones especially uncertain.
The latest CMIP6 multi-model ensembles exhibit a pronounced wet bias across Northwest China, apparent in annual means, seasonal cycles, and extreme precipitation metrics [7,8,9]. Biases are especially large in southern Xinjiang and the Hexi Corridor, where precipitation is often overestimated by more than 100% [10]. Similar deficiencies are seen in simulating snowfall phase changes, drought severity, and extreme events under future SSP scenarios [11,12]. The systematic overestimation of precipitation in these dryland zones is attributed not only to model resolution but also to insufficient representations of convection and land–atmosphere coupling [13]. While increasing horizontal resolution helps reduce local biases [14], model improvements must also focus on regional cloud microphysics, aerosol interactions, and thermodynamic structure [1,15].
As Northwest China is a major source region for dust aerosols, the absence of detailed dust–cloud microphysical interactions may contribute to model bias by overlooking their atmospheric stabilizing effects. Dust particles, abundant over the deserts of Northwest China, are highly efficient ice-nucleating particles (INPs) due to their mineral composition and large surface area [16,17]. When lofted into the middle and upper troposphere, they can initiate heterogeneous freezing, altering ice-cloud microphysics and convective dynamics [18,19,20]. Such microphysical modifications affect precipitation efficiency, particularly by suppressing rainfall through the cold-rain pathway, in which small ice particles delay coalescence and sedimentation [21,22]. Additionally, dust can stabilize the atmosphere by absorbing solar radiation in elevated layers, enhancing thermal inversions, increasing the level of free convection (LFC), and raising convective inhibition (CIN), all of which suppress deep convective development [23,24]. Despite these well-documented effects, most CMIP6 global climate models still treat INPs in a highly simplified or fixed manner, often neglecting key variables such as aerosol vertical distribution, mineralogy, and interactions with latent heat release [25,26]. This lack of process-level representation likely contributes to persistent wet biases in dryland precipitation simulations across Northwest China.
To address this gap, we conduct a set of targeted experiments using Community Atmosphere Model version 5 (CAM5) to isolate the microphysical and radiative impacts of dust aerosols on precipitation over Northwest China, aiming to improve understanding and simulation of dust–precipitation feedbacks. We implement a series of numerical experiments using the CAM5, including a baseline simulation with full dust effects (TOTAL), and two sensitivity experiments removing either ice-nucleating ability (NOICE) or dust entirely (NODUST). These experiments allow us to separate the individual contributions of dust’s microphysical and direct radiative effects. Our central hypothesis is that incorporating dust–ice-cloud interactions (DICIs) into GCMs can reduce the prevalent wet bias in Northwest China by suppressing convective precipitation through thermodynamic and microphysical modifications. Through detailed analysis of cloud properties, convective indices, and rainfall responses, this study aims to build a physically consistent pathway from dust aerosol perturbations to precipitation suppression, offering new insights into model improvement for dryland hydrology.

2. Methods

2.1. Model Description and Configuration

This study employs the Community Atmosphere Model version 5 (CAM5), the atmospheric component of CESM1, which features improved representations of aerosol–cloud–radiation interactions [27,28]. The model is run with the finite-volume dynamical core at a horizontal resolution of 0.9° × 1.25°, and it has 30 vertical hybrid sigma-pressure levels. CAM5 includes a two-moment stratiform cloud microphysics scheme, prognostic dust emission and transport modules, and updated parameterizations for heterogeneous ice nucleation.
To simulate the impact of dust aerosols on cloud processes and precipitation, we adopt a suite of physically consistent schemes for dust emission, transport, radiation interactions, cloud microphysics, and convection. These core physical process schemes are summarized in Table 1.

2.2. Sensitivity Experiment Design

To quantitatively assess the impact of dust–ice-cloud interactions on precipitation over Northwest China, we conduct three CAM5 sensitivity experiments under prescribed sea surface temperature (SST) and sea ice boundary conditions (HadISST):
  • TOTAL (Control Run): Includes both the radiative and microphysical effects of dust aerosols, representing full aerosol–cloud–radiation coupling.
  • NOICE: Identical to TOTAL, but disables heterogeneous ice nucleation from dust aerosols. This isolates the microphysical effect of dust via INP interactions.
  • NODUST: Completely removes dust emissions and transport, excluding both microphysical and radiative effects. This isolates the total dust impact by providing a dust-free baseline.
Each experiment is integrated for 25 years, with the first 5 years discarded as spin-up. The ensemble mean of the remaining 20 years is used for analysis. By comparing the results across the three experiments, we evaluate the total effect of dust aerosols (TOTAL–NODUST), the ice-nucleating effect (TOTAL–NOICE), and the residual radiative effect (NOICE–NODUST) on cloud structure, atmospheric stability, and regional precipitation characteristics.

2.3. Observational Data and Aridity Index

To evaluate model performance and define arid regions, we use monthly precipitation data from the Global Precipitation Climatology Project (GPCP v2.3), which provides merged satellite and gauge-based observations at 2.5° × 2.5° resolution from 1979 onward [37]. GPCP serves as the observational reference for diagnosing precipitation biases over Northwest China.
To assess the long-term aridity conditions, we employed annual mean Aridity Index (AI) data derived from the Global Land Data Assimilation System (GLDAS). Specifically, we used GLDAS-Noah Version 2.1 model outputs for the period 1985–2014, with a spatial resolution of 0.25° × 0.25°. The AI was calculated as the ratio of annual precipitation (P) to potential evapotranspiration (PET), i.e., AI = P/PET, where precipitation and PET were extracted from the GLDAS land surface water balance components. The GLDAS data were obtained from NASA’s GES DISC data archive (https://disc.gsfc.nasa.gov, accessed on 18 May 2025), and the AI values were computed and averaged annually to characterize spatial patterns of aridity over the 30-year period. This dataset provides a physically consistent and observation-constrained estimate of surface hydrological conditions, suitable for evaluating climate model performance in arid and semi-arid regions.

3. Results

3.1. Systematic Overestimation of Precipitation in CMIP6 Models

Accurately simulating precipitation over arid and semi-arid regions remains a persistent challenge in global climate models. Figure 1a presents the spatial distribution of the Aridity Index (AI) across China, identifying Northwest China (outlined by the black dashed box) as one of the driest climatological zones. However, most CMIP6 models systematically overestimate precipitation in this region. As shown in Figure 1b, the ratio of simulated annual mean precipitation from 23 CMIP6 models to the GPCP observations reveals widespread wet biases over Northwest China, co-located with the areas of low AI. This systematic overestimation is further confirmed in Figure 1c, where annual mean precipitation over the Northwest China region (NWC) is shown for each individual CMIP6 model. All 23 models simulate higher precipitation values than observed, with the multi-model ensemble mean also exceeding the GPCP reference (black line). Figure 1d illustrates the monthly evolution of precipitation, showing that the CMIP6 ensemble mean (red line) consistently overestimates rainfall throughout the year compared to GPCP observations (black line). The standard deviation (grey shading) also indicates a large inter-model spread, underscoring persistent structural uncertainties in simulating regional hydrological processes. These results suggest that existing climate models may lack key processes, particularly those related to cloud–aerosol interactions needed for accurate precipitation simulations in dry regions.

3.2. Improvements in Precipitation Simulation by Incorporating Dust–Ice-Cloud Interactions

To explore whether more detailed dust–cloud interactions can reduce the wet bias, we conducted a series of sensitivity experiments using CAM5, including a control run with all dust effects (TOTAL), a simulation with dust removed (NODUST), and a simulation excluding only ice-nucleation effects of dust (NOICE). Figure 2 shows the spatial distribution of dust effects on annual mean precipitation, focusing on three precipitation components: PRECT, PRECC, and PRECL. The dust total effect (TOTAL-NODUST; Figure 2a–c) leads to a notable reduction in precipitation over Northwest China, particularly through suppression of convective rainfall. Further decomposition reveals that the dust–ice-cloud effect (TOTAL-NOICE; Figure 2d–f) is the primary driver of this reduction, especially in PRECC, while the direct radiative effect (NOICE-NODUST; Figure 2g–i) contributes only marginally to precipitation changes in this region. A two-tailed Student’s t-test at the 99% confidence level was applied to assess the statistical significance of differences between the sensitivity experiments. These rigorous statistical tests help to ensure the robustness and reliability of our model results. These results highlight the dominant role of dust-induced ice nucleation in modulating cloud microphysics and reducing precipitation efficiency.

3.3. Monthly Evolution of Dust-Induced Precipitation Changes

To further clarify the seasonal characteristics of dust–cloud interaction impacts on precipitation, Figure 3 presents the monthly evolution of dust-induced changes over the Northwest China region. As shown in Figure 3a, the total dust effect (green bar) leads to a clear reduction in precipitation during the spring and early summer (March to July), with the ice-cloud effect (purple bar) accounting for the majority of this reduction. In contrast, the direct radiative effect (orange bar) shows only minor fluctuations and negligible contribution to the total reduction. Figure 3b further disaggregates the monthly evolution of the dust–ice-cloud effect into individual precipitation components. It reveals that the spring–summer reduction in PRECT is primarily driven by a sharp decline in PRECC, while changes in PRECL are minimal. These results suggest that dust-induced ice nucleation suppresses convective development, thereby reducing rainfall efficiency during the convective season.

4. Physical Mechanisms of Dust–Ice-Cloud Induced Precipitation Reduction over Arid Northwest China

4.1. Ice-Cloud Microphysical Responses to Dust Aerosols

The inclusion of dust–ice-cloud interactions in the CAM5 model substantially alters the microphysical properties of both ice and liquid clouds over the arid regions of Northwest China. As shown in Figure 4a–c, the presence of mineral dust acting as efficient ice-nucleating particles (INPs) significantly increases the ice-cloud number concentration (INNC), particularly during spring and early summer when dust loading is strongest. Correspondingly, the effective radius of ice particles (Rei) decreases, indicating that more numerous but smaller ice crystals are formed due to enhanced heterogeneous ice nucleation. This microphysical shift leads to an overall increase in the ice water path (IWP), implying more extensive accumulation of ice condensate in the atmosphere. These responses suggest a strengthened cold cloud process, whereby mineral dust facilitates more rapid glaciation of mixed-phase clouds, enhancing both cloud optical depth and precipitation efficiency. The increased IWP may also be associated with longer cloud lifetimes or greater horizontal cloud coverage, both of which can influence radiative forcing and cloud dynamics in arid regions.
Interestingly, the impact of dust aerosols also extends to liquid-phase clouds, as illustrated in Figure 4d–f. Unlike typical aerosol–cloud interactions that often lead to increased cloud droplet number concentration (CDNUMC), we observe a decline in CDNUMC over the arid region. At the same time, the effective radius of cloud droplets (Re) increases, while the liquid water path (LWP) decreases. This counterintuitive pattern suggests that the dust-induced intensification of ice processes may deplete supercooled liquid water, leading to fewer but larger cloud droplets. The reduction in CDNUMC could also indicate a suppression of cloud droplet activation, potentially due to increased competition for available water vapor or enhanced droplet scavenging by growing ice particles. Taken together, these results highlight a physically consistent response of the cloud system to dust perturbations: enhanced ice nucleation consumes available liquid water and modifies the cloud droplet spectrum, resulting in a net decrease in LWP. These changes are especially important in arid regions, where small shifts in cloud microphysics can have amplified effects on precipitation formation and surface water availability. Our findings underscore the need to represent dust–ice-cloud interactions explicitly in climate models to improve the fidelity of precipitation simulations over dust-prone drylands.
To further elucidate the influence of dust–ice-cloud interactions on cloud vertical structure and cloud type distribution, Figure 5 presents the changes in cloud properties derived from the modified CAM5 simulations. Figure 5a–d shows the vertical cross-sections (averaged over the study region) of cloud fraction, cloud extinction coefficient, ice water content (IWC), and ice-cloud number concentration, respectively. The results consistently demonstrate that dust-induced ice nucleation substantially enhances upper-tropospheric cloud characteristics, particularly between 400 hPa and 300 hPa. These increases are most pronounced in the regions corresponding to the elevated ice-cloud layers identified in Figure 4, confirming the vertically confined enhancement of ice clouds due to dust–ice interactions. The increase in cloud fraction and extinction at these levels suggests denser and optically thicker clouds, which are likely associated with increased ice particle concentration and IWC. The elevated ice-cloud number further supports the role of dust as efficient ice-nucleating particles, accelerating glaciation and promoting the formation of more extensive and optically active upper-level clouds.
Figure 5e–g display the spatial distribution of changes in low-level, mid-level, and high-level cloud cover, respectively. While dust–ice interactions do not produce significant changes in low-level clouds over Northwest China, both mid-level and especially high-level cloud cover exhibit marked increases across the arid region. This is particularly evident in the Tarim Basin and Hexi Corridor, where high cloud amounts are significantly enhanced. These spatial patterns are consistent with the vertical cross-sections and the microphysical responses observed in Figure 4, further corroborating that dust primarily influences high-level ice clouds in this region.
The increased high cloud coverage implies potential changes in atmospheric radiative balance and vertical temperature structure. By enhancing longwave trapping and modifying the lapse rate, high clouds may exert feedbacks on local thermodynamic stability, a mechanism that will be further explored in the next section.

4.2. Enhanced Upper-Level Cloudiness Stabilizes Atmospheric Stratification

Following the increase in high-level cloud fraction associated with dust–ice-cloud interactions (as shown in Figure 4 and Figure 5), the thermodynamic structure of the atmosphere over Northwest China undergoes a significant transformation. Figure 6a,b illustrate that during the March–July period, temperature decreases at 700 hPa while increasing at 400 hPa, resulting in a notable reduction in the vertical temperature gradient between these two layers (Figure 6c). Given the elevated terrain of the Northwest China arid region, 700 hPa is representative of the lower troposphere, while 400 hPa corresponds well with the levels of increased high cloudiness. This dipole temperature response implies enhanced atmospheric stability due to reduced buoyancy.
This conclusion is further supported by changes in thermodynamic stability indices. The K index, a conventional convective potential indicator, shows a significant decrease (Figure 6d), while the Richardson number, which quantifies dynamic stability, increases substantially (Figure 6e). Both indicators confirm a more stable atmospheric stratification. Additionally, Figure 6f shows that the vertical gap between the lifting condensation level (LCL) and the level of free convection (LFC) widens, suggesting a higher energy threshold for parcel ascent and more suppressed convection.
Taken together, these results indicate that enhanced upper-level cloudiness induced by dust–ice interactions warms the upper troposphere and cools the lower troposphere, thereby strengthening the static stability of the atmosphere. This mechanism provides a physical basis for the observed suppression of convective precipitation in the region, as reported in earlier sections.

4.3. Suppressed Updrafts and Moisture Convergence

As a direct response to the dust-induced thermodynamic stabilization, both large-scale and convective vertical motions experience substantial weakening over Northwest China. Figure 7a displays positive anomalies in 500 hPa vertical velocity (Ω) during March–July, signifying a notable reduction in upward motion under dusty conditions. This suppression of vertical ascent hampers the vertical transport of heat and moisture, thereby inhibiting the development and sustenance of convective systems. Correspondingly, the vertical profile of convective mass flux (CMF) shown in Figure 7b exhibits a widespread decrease throughout the troposphere. The reduction in CMF reflects diminished buoyancy and convective vigor, indicating a weakened capacity for latent heat release and vertical momentum transfer. Together, these results illustrate a reinforcing feedback loop initiated by DICI: microphysical changes enhance thermodynamic stability, which in turn suppresses vertical motion and moisture convergence, ultimately leading to decreased convective activity and rainfall over the arid region.

5. Discussion

This study reveals the critical role of DICI in mitigating longstanding precipitation biases over arid regions in climate models. By incorporating the microphysical effects of dust as ice nuclei, the CAM5 simulations demonstrate notable improvements in capturing the spatial distribution and intensity of precipitation in Northwest China. These improvements primarily arise from enhanced ice-nucleation rates, cloud optical thickness, and convective stability, factors that are typically underestimated in simulations neglecting aerosol–cloud interactions. The results underscore the necessity of integrating realistic aerosol–cloud microphysics into climate models, particularly in regions where cloud formation is highly sensitive to ice-phase processes.
However, several sources of uncertainty should be acknowledged. First, the representation of dust emissions and transport in CAM5, though physically based, may not fully capture the subgrid-scale variability of dust source strength and aerosol vertical profiles, which are essential for determining the activation of ice nuclei at cloud-forming altitudes. Second, while our results point to the predominant role of heterogeneous ice nucleation, other pathways—such as secondary ice production (SIP) and aerosol-induced thermodynamic feedbacks—may also influence cloud properties and convective development, but are less constrained in current parameterizations. However, SIP is known to play a significant role mainly in deep convective systems with strong updrafts and abundant supercooled water. Given the relatively weak convective activity and limited supercooled liquid in our study region and period, we consider the impact of SIP to be secondary. Additionally, observational datasets used for evaluation carry their own retrieval uncertainties, particularly over topographically complex and cloud-sparse regions like Northwest China.
Looking forward, our findings suggest that further improvements in precipitation simulation over arid regions hinge on two key developments: (1) better observational constraints on dust–cloud–precipitation processes through in situ campaigns and satellite synergy, and (2) refined model parameterizations that capture the spectrum of ice-nucleation pathways and their environmental dependence. The incorporation of size-resolved and composition-resolved dust particle information, as well as dynamic coupling between aerosols and cloud systems, is expected to enhance predictive skill. More broadly, this study highlights the importance of region-specific aerosol–cloud interactions in addressing persistent model biases, thereby improving the fidelity of climate projections in vulnerable arid and semi-arid regions.

6. Conclusions

Accurate simulation of precipitation over arid regions remains a critical challenge in global climate modeling. This study identifies a key mechanism—dust–ice-cloud interactions—that substantially improves the simulation of precipitation in Northwest China within the Community Atmosphere Model version 5 (CAM5) model framework. By explicitly accounting for the heterogeneous ice-nucleation effect of mineral dust, our simulations show reduced biases in total precipitation, better agreement with observations, and a more realistic seasonal evolution of hydrometeorological processes.
Figure 8 presents the conceptual mechanism underlying the improved precipitation simulation achieved by incorporating dust–ice-cloud interactions into the CAM5 model. This mechanism demonstrates how dust aerosols, acting as efficient ice-nucleating particles, modify both the microphysical structure of clouds and the thermodynamic stability of the atmosphere, ultimately reducing the longstanding wet bias in precipitation over Northwest China and enhancing model accuracy. From a microphysical perspective, the presence of dust leads to a significant increase in the number concentration of ice crystals, while simultaneously decreasing the effective radius of these ice particles. This shift enhances ice-cloud formation and increases the optical thickness of high-level clouds. Consequently, the ice water path becomes larger, and high cloud cover increases, accompanied by stronger extinction coefficients associated with these clouds. These physically coherent changes indicate a more realistic representation of the cloud system in dust-laden conditions. From a dynamical and thermodynamic standpoint, the presence of dust-induced ice clouds contributes to changes in atmospheric vertical structure. The temperature lapse rate becomes steeper, the K index—a measure of convective potential—decreases, and the Richardson number increases, all of which suggest a more stable atmosphere. Additionally, vertical motion at 500 hPa shows a stronger downward tendency, convective mass flux is reduced, and the lifting of air parcels to form deep clouds becomes more difficult. These conditions collectively inhibit the development of deep convection, making precipitation less likely. Importantly, these mechanisms are most prominent during the March to July period, when both dust loading and convective activity reach seasonal peaks. The improvements in precipitation simulation manifest not only in total rainfall amounts but also in the separate convective and large-scale components, confirming that dust–ice interactions influence both cloud microphysics and larger-scale atmospheric dynamics. The consistency of these findings across multiple physical variables reinforces the credibility of the proposed mechanism illustrated in Figure 8.
Overall, this work demonstrates the necessity of incorporating aerosol–cloud interactions, especially those involving dust and the ice phase, into climate models to reduce longstanding biases over arid regions. The improved representation of precipitation in such regions is vital for water resource assessments, ecosystem modeling, and regional climate adaptation strategies. Future research should focus on enhancing dust emission schemes, integrating observational constraints on ice nucleation, and extending the analysis to other arid regions globally to assess the broader applicability of this mechanism.

Author Contributions

Material preparation was performed by Z.D., X.L. (Xiaoyun Li) and X.L. (Xiaokang Liu); data collection were performed by Z.X. The first draft of the manuscript was written by A.W., and all authors commented on previous versions of the manuscript. X.X. and K.S. have read and approved the final manuscript. All authors contributed to the study conceptualization and design. All authors have read and agreed to the published version of the manuscript.

Funding

This study was funded by the National Natural Science Foundation of China (Grant no. 42241110), the Shaanxi Province Postdoctoral Research Funding Project (Grant no. 2023BSHEDZZ192), Open Fund of the State Key Laboratory of Loess Science (Grant no. SKLLQG2312), the Fundamental Research Funds for the Central Universities (GK202301003, GK202309006), the Natural Science Basic Research Program of Shaanxi Province (2021JCW-17), the National Science and Technology Fundamental Resources Investigation Program of China (Grant no. 2022FY202304), the National Natural Science Foundation of China (Grant nos. 42175059, 42205037, 42177202, 41807060).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data supporting the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of Interest

The authors declare no conflicts of interest relevant to this study.

Abbreviations

Coupled Model Intercomparison Project Phase 6CMIP6
Community Atmosphere Model version 5CAM5
Global Precipitation Climatology ProjectGPCP
Ice-Nucleating ParticlesINPs
Level of Free ConvectionLFC
Lifting Condensation LevelLCL
Convective InhibitionCIN
Sea Surface TemperatureSST
Aridity IndexAI
Northwest China regionNWC
Total PrecipitationPRECT
Convective PrecipitationPRECC
Large-Scale PrecipitationPRECL
Cloud Droplet Number ConcentrationCDNUMC
Liquid Water PathLWP
Ice Water ContentIWC
Convective Mass FluxCMF

References

  1. Luo, R.; Liu, Y.; Zhu, Q.; Tang, Y.; Shao, T. Effects of aerosols on cloud and precipitation in East-Asian drylands. Int. J. Climatol. 2021, 41, 4603–4618. [Google Scholar] [CrossRef]
  2. Xie, X.; Liu, X.; Shi, Z.; Li, X.; Xie, X.; Sun, H.; He, J.; Che, H.; Zhang, X.-Y.; An, Z.; et al. Sharp decline of dust events induces regional wetting over arid and semi-arid Northwest China in the NCAR Community atmosphere model. Environ. Res. Lett. 2023, 19, 014061. [Google Scholar] [CrossRef]
  3. Han, H.; Hou, J.; Jiang, R.; Jiahui, G.; Ganggang, B.; Kang, Y.; Qi, W.; Li, Y.; Bingyao, L. Spatial and temporal variation of precipitation characteristics in the semiarid region of Xi’an, northwest China. J. Water Clim. Change 2021, 12, 2697–2715. [Google Scholar] [CrossRef]
  4. Xu, J.; Ping, F.; Li, J.; Du, H. A study of cloud microphysical processes and a mesoscale environment in a heavy rainfall case over Yan’an. Front. Earth Sci. 2023, 11, 1149856. [Google Scholar] [CrossRef]
  5. Tian, J.; Zhang, Z.; Zhao, T.; Tao, H.; Zhu, B. Warmer and wetter climate induced by the continual increase in atmospheric temperature and precipitable water vapor over Northwest China. J. Hydrol. Reg. Stud. 2022, 42, 101151. [Google Scholar] [CrossRef]
  6. Qian, P.; Zhu, J.; Yang, J.; Jiang, P. Terrestrial moisture sources dominate summer precipitation fluctuations in Northwest China. Environ. Res. Lett. 2024, 19, 124052. [Google Scholar] [CrossRef]
  7. Tian, J.; Zhang, Z.; Ahmed, Z.; Zhang, L.; Su, B.; Tao, H.; Jiang, T. Projections of precipitation over China based on CMIP6 models. Stoch. Environ. Res. Risk Assess. 2021, 35, 831–848. [Google Scholar] [CrossRef]
  8. Zhang, X.; Wang, X.; Zhou, B. CMIP6-projected changes in drought over Xinjiang, Northwest China. Int. J. Climatol. 2023, 43, 6560–6577. [Google Scholar] [CrossRef]
  9. Song, X.; Xu, M.; Kang, S.; Wang, R.; Wu, H. Evaluation and projection of changes in temperature and precipitation over Northwest China based on CMIP6 models. Int. J. Climatol. 2024, 44, 5039–5056. [Google Scholar] [CrossRef]
  10. Guo, H.; Bao, A.; Chen, T.; Zheng, G.; Wang, Y.; Jiang, L.; Ma, P. Assessment of CMIP6 in simulating precipitation over arid Central Asia. Atmos. Res. 2021, 252, 105451. [Google Scholar] [CrossRef]
  11. Lei, X.; Xu, C.; Liu, F.; Song, L.; Cao, L.; Suo, N. Evaluation of CMIP6 models and multi-model ensemble for extreme precipitation over arid Central Asia. Remote Sens. 2023, 15, 2376. [Google Scholar] [CrossRef]
  12. Xu, M.; Hou, Z.; Kang, S.; Wu, X.; Han, H.; Wang, P. Projection of snowfall and precipitation phase changes over Northwest China based on CMIP6 multimodels. J. Hydrol. 2024, 641, 131743. [Google Scholar] [CrossRef]
  13. Zhao, P.; Yang, K.; Wang, C. Impacts of cumulus parameterizations on extreme precipitation simulation in semi-arid region: A case study in Northwest China. Atmosphere 2022, 13, 1464. [Google Scholar] [CrossRef]
  14. Jin, Z.; Ge, F.; Chen, Q.; Lin, Z. To what extent horizontal resolution improves the simulation of precipitation in CMIP6 HighResMIP models over Southwest China? Front. Earth Sci. 2023, 10, 1003748. [Google Scholar] [CrossRef]
  15. Wang, R.; Tian, Y.; Li, J. Analysis of convective and stratiform precipitation characteristics in the summers of 2014–2019 over Northwest China based on GPM observations. Atmosphere 2021, 12, 879. [Google Scholar] [CrossRef]
  16. Hoose, C.; Möhler, O. Heterogeneous ice nucleation on atmospheric aerosols: A review of results from laboratory experiments. Atmos. Chem. Phys. 2012, 12, 9817–9854. [Google Scholar] [CrossRef]
  17. Kanji, Z.A.; Ladino, L.A.; Wex, H.; Boose, Y.; Burkert-Kohn, M.; Cziczo, D.J.; Krämer, M. Overview of ice nucleating particles. Meteorol. Monogr. 2017, 58, 1.1–1.33. [Google Scholar] [CrossRef]
  18. Min, Q.; Li, R.; Lin, B.; Joseph, E.; Wang, S.; Hu, Y.; Morris, V.; Chang, F. Evidence of mineral dust altering cloud microphysics and precipitation. Atmos. Chem. Phys. 2009, 9, 3223–3231. [Google Scholar] [CrossRef]
  19. Zhao, C.; Liu, X.; Leung, L.R.; Hagos, S. Radiative impact of mineral dust on monsoon precipitation variability over West Africa. Atmos. Chem. Phys. 2011, 11, 1879–1893. [Google Scholar] [CrossRef]
  20. Fan, J.; Leung, L.R.; DeMott, P.J.; Comstock, J.M.; Singh, B.; Rosenfeld, D.; Tomlinson, J.M.; White, A.B.; Prather, K.A.; Minnis, P.; et al. Aerosol impacts on California winter clouds and precipitation during CalWater 2011: Local pollution versus long-range transported dust. Atmos. Chem. Phys. 2014, 14, 81–101. [Google Scholar] [CrossRef]
  21. Kubar, T.L.; Hartmann, D.L.; Wood, R. Radiative and convective driving of tropical high clouds. J. Clim. 2007, 20, 5510–5526. [Google Scholar] [CrossRef]
  22. Seigel, R.B.; Van den Heever, S.C.; Saleeby, S.M. Mineral dust indirect effects and cloud radiative feedbacks of a simulated idealized nocturnal squall line. Atmos. Chem. Phys. 2013, 13, 4467–4485. [Google Scholar] [CrossRef]
  23. Yin, Y.; Chen, L. The effects of heating by transported dust layers on cloud and precipitation: A numerical study. Atmos. Chem. Phys. 2007, 7, 3497–3505. [Google Scholar] [CrossRef]
  24. Zhu, H.; Li, R.; Yang, S.; Zhao, C.; Jiang, Z.; Huang, C.Y. The impacts of dust aerosol and convective available potential energy on precipitation vertical structure in eastern China as seen from multiple source observations. Atmos. Chem. Phys. 2023, 23, 2421–2440. [Google Scholar] [CrossRef]
  25. Li, X.; Liu, Y.; Wang, M.; Jiang, Y.; Dong, X. Assessment of the CMIP6 model performance in simulating aerosol optical depth over eastern China. Atmos. Res. 2021, 261, 105747. [Google Scholar] [CrossRef]
  26. Tian, J.; Zhang, Z.; Fu, Y.; Tao, H.; Zhu, B.; Yang, L. Does CMIP6 inspire more confidence in projecting precipitation over China? Int. J. Climatol. 2023, 43, 4965–4986. [Google Scholar] [CrossRef]
  27. Neale, R.B.; Gettelman, A.; Park, S.; Conley, A.J.; Kinnison, D.; Dan, M.; Smith, A.K.; Vitt, F.; Morrison, H.; Cameronsmith, P. Description of the NCAR Community Atmosphere Model (CAM5.0) Technical Report NCAR/TN-486+STR; National Center for Atmospheric Research: Boulder, CO, USA, 2010. [Google Scholar]
  28. Ghan, S.; Liu, X.; Easter, R.C.; Zaveri, R.; Rasch, P.J.; Yoon, J.H.; Eaton, B. Toward a minimal representation of aerosols in climate models: Comparative decomposition of aerosol direct, semidirect, and indirect radiative forcing. J. Clim. 2012, 25, 6461–6476. [Google Scholar] [CrossRef]
  29. Mahowald, N.M.; Muhs, D.R.; Levis, S.; Rasch, P.J.; Yoshioka, M.; Zender, C.S.; Luo, C. Change in atmospheric mineral aerosols in response to climate: Last glacial period, preindustrial, modern, and doubled carbon dioxide climates. J. Geophys. Res. Atmos. 2006, 111, D10202. [Google Scholar] [CrossRef]
  30. Kok, J.F. A scaling theory for the size distribution of emitted dust aerosols suggests climate models underestimate the size of the global dust cycle. Proc. Natl. Acad. Sci. USA 2011, 108, 1016–1021. [Google Scholar] [CrossRef]
  31. Albani, S.; Mahowald, N.M.; Perry, A.T.; Scanza, R.A.; Zender, C.S.; Heavens, N.G.; Maggi, V.; Kok, J.F.; Otto-Bliesner, B.L. Improved dust representation in the Community Atmosphere Model. J. Adv. Model. Earth Syst. 2014, 6, 541–570. [Google Scholar] [CrossRef]
  32. Miller, R.L.; Knippertz, P.; García-Pando, C.; Perlwitz, J.P.; Tegen, I. Impact of dust radiative forcing upon climate. In Mineral Dust: A Key Player in the Earth System; Knippertz, P., Stuut, J.-B.W., Eds.; Springer: Dordrecht, The Netherlands, 2014; pp. 327–357. [Google Scholar]
  33. Morrison, H.; Gettelman, A. A new two-moment bulk stratiform cloud microphysics scheme in the Community Atmosphere Model, version 3 (CAM3). Part I: Description and numerical tests. J. Clim. 2008, 21, 3642–3659. [Google Scholar] [CrossRef]
  34. Zhang, G.J.; McFarlane, N.A. Sensitivity of climate simulations to the parameterization of cumulus convection in the Canadian Climate Center general circulation model Atmos. Ocean 1995, 33, 407–446. [Google Scholar]
  35. DeMott, P.J.; Prenni, A.J.; McMeeking, G.R.; Sullivan, R.C.; Petters, M.D.; Tobo, Y.; Niemand, M.; Möhler, O.; Snider, J.R.; Wang, Z.; et al. Integrating laboratory and field data to quantify the immersion freezing ice nucleation activity of mineral dust particles. Atmos. Chem. Phys. 2015, 15, 393–409. [Google Scholar] [CrossRef]
  36. Tan, I.; Storelvmo, T. Sensitivity study on the influence of cloud microphysical parameters on mixed-phase cloud thermodynamic phase partitioning in CAM5. J. Atmos. Sci 2016, 73, 709–728. [Google Scholar] [CrossRef]
  37. Adler, R.F.; Huffman, G.J.; Chang, A.; Ferraro, R.; Xie, P.; Janowiak, J.; Rudolf, B.; Schneider, U.; Curtis, S.; Bolvin, D.; et al. The Version-2 Global Precipitation Climatology Project (GPCP) Monthly Precipitation Analysis (1979–Present). J. Hydrometeorol. 2003, 4, 1147–1167. [Google Scholar] [CrossRef]
Atmosphere 16 00800 g001aAtmosphere 16 00800 g001b
Figure 1. (a) Spatial distribution of the Aridity Index (AI) over China, with the black dashed box indicating the Northwest China (NWC) domain defined in this study. (b) Relative changes in annual mean precipitation simulated by the CMIP6 multi-model ensemble to Global Precipitation Climatology Project (GPCP) observational data. (c) Annual mean precipitation simulated by each of the 23 CMIP6 models and ensemble mean for the NWC domain compared with GPCP observations (black line). (d) Monthly climatology of precipitation in the NWC domain from the 23 CMIP6 models (colored lines corresponding to (c)), CMIP6 ensemble mean (red line with ±1 standard deviation shaded), and GPCP observations (black line).
Figure 1. (a) Spatial distribution of the Aridity Index (AI) over China, with the black dashed box indicating the Northwest China (NWC) domain defined in this study. (b) Relative changes in annual mean precipitation simulated by the CMIP6 multi-model ensemble to Global Precipitation Climatology Project (GPCP) observational data. (c) Annual mean precipitation simulated by each of the 23 CMIP6 models and ensemble mean for the NWC domain compared with GPCP observations (black line). (d) Monthly climatology of precipitation in the NWC domain from the 23 CMIP6 models (colored lines corresponding to (c)), CMIP6 ensemble mean (red line with ±1 standard deviation shaded), and GPCP observations (black line).
Atmosphere 16 00800 g001aAtmosphere 16 00800 g001b
Atmosphere 16 00800 g002
Figure 2. Spatial distribution of relative changes in annual mean in total precipitation (PRECT), convective precipitation (PRECC), and large-scale precipitation (PRECL) over China due to the following: (a–c) the total dust effect, (d–f) the ice-cloud interactions, and (g–i) the direct radiative effect. All shaded regions with black hatching indicate 99% significance.
Figure 2. Spatial distribution of relative changes in annual mean in total precipitation (PRECT), convective precipitation (PRECC), and large-scale precipitation (PRECL) over China due to the following: (a–c) the total dust effect, (d–f) the ice-cloud interactions, and (g–i) the direct radiative effect. All shaded regions with black hatching indicate 99% significance.
Atmosphere 16 00800 g002
Atmosphere 16 00800 g003
Figure 3. Monthly variations in (a) total precipitation (PRECT) changes over Northwest China (NWC) due to the total dust effect, ice-cloud effect, and direct radiative effect and (b) total precipitation (PRECT), convective precipitation (PRECC), and large-scale precipitation (PRECL) induced by the ice-cloud effect (Unit: mm/day).
Figure 3. Monthly variations in (a) total precipitation (PRECT) changes over Northwest China (NWC) due to the total dust effect, ice-cloud effect, and direct radiative effect and (b) total precipitation (PRECT), convective precipitation (PRECC), and large-scale precipitation (PRECL) induced by the ice-cloud effect (Unit: mm/day).
Atmosphere 16 00800 g003
Atmosphere 16 00800 g004
Figure 4. Cloud-top microphysical responses of cloud properties to dust-ice-cloud interactions during March–July over Northwest China. (a–c) Changes in ice-cloud number concentration (INNC, cm−3), ice effective radius (Rei, µm), and ice water path (IWP, g m−2). (d–f) Changes in liquid cloud droplet number concentration (CDNUMC, cm−3), liquid effective radius (Re, µm), and liquid water path (LWP, g m−2). All shaded regions with black hatching indicate 99% significance.
Figure 4. Cloud-top microphysical responses of cloud properties to dust-ice-cloud interactions during March–July over Northwest China. (a–c) Changes in ice-cloud number concentration (INNC, cm−3), ice effective radius (Rei, µm), and ice water path (IWP, g m−2). (d–f) Changes in liquid cloud droplet number concentration (CDNUMC, cm−3), liquid effective radius (Re, µm), and liquid water path (LWP, g m−2). All shaded regions with black hatching indicate 99% significance.
Atmosphere 16 00800 g004
Atmosphere 16 00800 g005
Figure 5. Vertical and horizontal responses of cloud properties to dust–ice-cloud interactions during March–July over Northwest China. (a–d) Vertical cross-section changes in cloud fraction (%), cloud extinction (km−1), ice water content (IWC, mg m−3), and ice-cloud number concentration (ppm). (e–g) Spatial distribution changes in low-level cloud fraction, mid-level cloud fraction, and high-level cloud fraction. All shaded regions with black hatching indicate 99% significance.
Figure 5. Vertical and horizontal responses of cloud properties to dust–ice-cloud interactions during March–July over Northwest China. (a–d) Vertical cross-section changes in cloud fraction (%), cloud extinction (km−1), ice water content (IWC, mg m−3), and ice-cloud number concentration (ppm). (e–g) Spatial distribution changes in low-level cloud fraction, mid-level cloud fraction, and high-level cloud fraction. All shaded regions with black hatching indicate 99% significance.
Atmosphere 16 00800 g005
Atmosphere 16 00800 g006
Figure 6. Cloud-induced thermodynamic responses to dust–ice-cloud interactions during March–July over Northwest China. (a–c) Changes in temperature at 700 hPa, 400 hPa, and the vertical temperature difference between 400 hPa and 700 hPa (ΔT700–400, K); (d,e) changes in atmospheric stability indices: K index (K) and Richardson number (Ri); (f) changes in the vertical distance between the lifting condensation level (LCL) and the level of free convection (LFC, hPa). All shaded regions with black hatching indicate 99% significance.
Figure 6. Cloud-induced thermodynamic responses to dust–ice-cloud interactions during March–July over Northwest China. (a–c) Changes in temperature at 700 hPa, 400 hPa, and the vertical temperature difference between 400 hPa and 700 hPa (ΔT700–400, K); (d,e) changes in atmospheric stability indices: K index (K) and Richardson number (Ri); (f) changes in the vertical distance between the lifting condensation level (LCL) and the level of free convection (LFC, hPa). All shaded regions with black hatching indicate 99% significance.
Atmosphere 16 00800 g006
Atmosphere 16 00800 g007
Figure 7. Dynamic responses to dust–ice-cloud interactions over Northwest China during March–July. (a) Changes in 500 hPa vertical velocity (Ω): positive anomalies indicate weakened large-scale upward motion. (b) Vertical profile of changes in convective mass flux (CMF). Shaded areas with black slashes denote regions passing the 0.99 significance level based on Student’s t-test.
Figure 7. Dynamic responses to dust–ice-cloud interactions over Northwest China during March–July. (a) Changes in 500 hPa vertical velocity (Ω): positive anomalies indicate weakened large-scale upward motion. (b) Vertical profile of changes in convective mass flux (CMF). Shaded areas with black slashes denote regions passing the 0.99 significance level based on Student’s t-test.
Atmosphere 16 00800 g007
Atmosphere 16 00800 g008
Figure 8. Mechanism of precipitation reduction by dust–ice–cloud interactions in Northwest China. Red upward arrows indicate an increase, while blue downward arrows indicate a decrease.
Figure 8. Mechanism of precipitation reduction by dust–ice–cloud interactions in Northwest China. Red upward arrows indicate an increase, while blue downward arrows indicate a decrease.
Atmosphere 16 00800 g008
Table 1. Key physical process schemes implemented in CAM5.
Table 1. Key physical process schemes implemented in CAM5.
ProcessDescriptionReference(s)
Dust EmissionSurface wind, soil properties and vegetation cover; Size distribution follows brittle fragmentation theory[29,30,31]
Dust Transport and DepositionIncludes advection, gravitational settling, dry and wet removal[31]
Radiative EffectsDust–radiation interaction alters shortwave and longwave fluxes[28,32]
Cloud MicrophysicsTwo-moment scheme predicting mass and number concentrations of cloud droplets and ice crystals[33]
Convection SchemesDeep convection: Zhang–McFarlane scheme; Shallow convection: University of Washington scheme[34]
Ice-Nucleation ParameterizationUpdated dust-INP scheme using temperature and large dust particle concentration; Replaces standard Meyers scheme[35]
Mixed-Phase Cloud AdjustmentReduces Wegener–Bergeron–Findeisen (WBF) process efficiency to optimize cloud-phase partitioning[36]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Wang, A.; Xie, X.; Dong, Z.; Li, X.; Shang, K.; Liu, X.; Xue, Z. Mitigating Model Biases in Arid Region Precipitation over Northwest China Through Dust–Cloud Microphysical Interactions. Atmosphere 2025, 16, 800. https://doi.org/10.3390/atmos16070800

AMA Style

Wang A, Xie X, Dong Z, Li X, Shang K, Liu X, Xue Z. Mitigating Model Biases in Arid Region Precipitation over Northwest China Through Dust–Cloud Microphysical Interactions. Atmosphere. 2025; 16(7):800. https://doi.org/10.3390/atmos16070800

Chicago/Turabian Style

Wang, Anqi, Xiaoning Xie, Zhibao Dong, Xiaoyun Li, Ke Shang, Xiaokang Liu, and Zhijing Xue. 2025. "Mitigating Model Biases in Arid Region Precipitation over Northwest China Through Dust–Cloud Microphysical Interactions" Atmosphere 16, no. 7: 800. https://doi.org/10.3390/atmos16070800

APA Style

Wang, A., Xie, X., Dong, Z., Li, X., Shang, K., Liu, X., & Xue, Z. (2025). Mitigating Model Biases in Arid Region Precipitation over Northwest China Through Dust–Cloud Microphysical Interactions. Atmosphere, 16(7), 800. https://doi.org/10.3390/atmos16070800

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop