HiCPC: A new 10-km CMIP6 downscaled daily climate projections over China

HiCPC: A new 10-km CMIP6 downscaled daily climate projections over China | Scientific Data

Abstract

Accurate climate projections are critical for various applications and impact assessments in environmental science and management. This study presents HiCPC (High-resolution CMIP6 downscaled daily Climate Projections over China), a novel dataset tailored to China’s specific needs. HiCPC leverages outputs from 22 global climate models (GCMs) from the Coupled Model Intercomparison Project Phase 6 (CMIP6). To address inherent biases in daily GCM simulations, an advanced Bias Correction and Spatial Disaggregation (BCSD) method is employed, using the China Meteorological Forcing Dataset (CMFD) as a reference. HiCPC offers detailed daily precipitation and temperature data across China at an enhanced spatial resolution of 0.1° × 0.1°. It covers both the historical period (1979–2014) and future projections (2015–2100) based on four CMIP6 Shared Socioeconomic Pathways (SSPs) scenarios (SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5). Upon validation, HiCPC demonstrates good performance, surpassing CMIP6 GCMs across the historical period. This reinforces its significance for essential research in climate change evaluation and its associated implications within China.

Similar content being viewed by others

Background & Summary

The escalating intensity of global warming is demonstrably linked to an alarming increase in the frequency and severity of extreme weather events, posing significant threats to human societies and ecosystems across the globe 1, 2. Owing to its distinctive geographical and climatic attributes, China is witnessing this phenomenon with intensified magnitude, manifesting in temperature increments that surpass the global mean and a heightened susceptibility to extreme meteorological events 3, 4. Consequently, in-depth investigations into the nation’s future climate trajectory are paramount for gauging potential consequences across critical sectors, including agriculture, water resources, energy, and the ecological environment 5. Such endeavors will pave the way for developing robust adaptation and mitigation strategies, ultimately providing a vital scientific foundation and decision-making support for effectively addressing the challenges posed by climate change.

Climate models are invaluable tools for investigating climate change implications 2, 6, despite advancements like the Sixth Coupled Model Intercomparison Project (CMIP6), global climate models (GCMs) still face significant challenges in simulating climate change for China 7, 8. Main challenges include: (1) substantial biases and uncertainties: GCMs struggle to accurately capture China’s diverse climate nuances, leading to significant errors in simulations 7; (2) coarse spatial resolution: The limited resolution of GCMs (often exceeding 100 km) hampers their ability to represent fine-scale regional climate dynamics crucial for impact assessments 9, 10, 11, 12. These challenges necessitate downscaling techniques to bridge the gap between large-scale model simulations and local-scale impact assessments 13, 14, 15, 16, 17, 18, 19, 20. However, existing global downscaled datasets like NEX-GDDP-CMIP6 21, 22, 23, ISIMIP 24 and GDPCIR 25 often lack sufficient resolution or optimal reference data and may not be optimal for China 26.

This study addresses these gaps by developing the High-resolution CMIP6 downscaled daily Climate Projections over China (HiCPC). We employ the refined Bias Correction and Spatial Disaggregation (BCSD) method and utilize the China Meteorological Forcing Dataset (CMFD) as reference data. HiCPC incorporates 22 CMIP6 models across four Shared Socioeconomic Pathways scenarios, offering high-resolution (0.1°) daily climate data valuable for assessing future climate changes and variability and driving diverse impact assessment models. The HiCPC dataset is freely available to download from the National Tibetan Plateau/Third Pole Environment Data Center ( https://doi.org/10.11888/Atmos.tpdc.301122) 27.

Methods

Study area and data acquisition

This study concentrates on China, consequently, the downscaling area encompasses the terrestrial region of China, which is delineated by the coordinates 15–55°N and 70–140°E.

To guide the statistical downscaling process, we have chosen datasets of daily precipitation (pr) and near-surface air temperature (daily mean tas, daily maximum tasmax, and daily minimum tasmin) spanning the historical period from 1979 to 2014, which collected from the China Meteorological Forcing Dataset (CMFD, https://data.tpdc.ac.cn/en/data/8028b944-daaa-4511-8769-965612652c49) 26. CMFD synergizes high-quality observations obtained from various sources, including the China Meteorological Administration (CMA) stations, Princeton forcing data, Global Land Data Assimilation System (GLDAS) data, GEWEX-SRB radiation data, and Tropical Rainfall Measuring Mission (TRMM) satellite precipitation analysis. Encompassing the entire country of China, the CMFD dataset offers a wealth of daily meteorological variables—including mean, maximum, and minimum temperatures, surface pressure, relative humidity, and wind speed—with a spatial resolution of 0.1°. The temporal resolution varies from sub-daily to monthly intervals, providing a robust timeframe that extends from 1979 to 2018 28, 29, 30, 31.

In this study, daily precipitation and near-surface air temperatures from 22 CMIP6 GCMs (Table S1) are downscaled for the historical period 1979–2014 and future period 2015–2100 under the four SSPs 1–2.6, 2–4.5, 3–7.0 and 5–8.5. The CMIP6 GCMs data are acquired via the Earth System Grid Federation website (ESGF, https://esgf-node.llnl.gov/projects/cmip6/). In the HiCPC production process, we preserved the individual temporal ranges of each global model. The time ranges for all downscaled models are also shown in Table S1.

Downscaling process

The refined methodology of Bias Correction and Spatial Disaggregation (BCSD) 32, 33, 34 is applied for downscaling the CMIP6 GCMs outputs for China. BCSD, as applied in NEX-CDDP-CMIP5 and NEX-CDDP-CMIP6 downscaling projects, has been widely acknowledged for its effectiveness in refining the resolution of GCM outputs while preserving the integrity of large-scale climate patterns, making it a valuable tool for regional climate impact studies 35, 36, 37. In the development of HiCPC, we chose BCSD method that align with those of NEX-CDDP-CMIP6. This strategy ensures ease of comparison for users, enhancing the utility and accessibility of our data set for comprehensive analysis. The daily BCSD method introduces an innovative process that applies bias-correction and spatial disaggregation directly to daily projections, eliminating the need for temporal disaggregation required by the initial BCSD method. This enhancement substantially augments the downscaling efficacy of daily variables compared to the conventional BCSD approach 34, 38.

The BCSD method comprises two primary steps: bias correction (BC) and spatial disaggregation (SD). These are briefly outlined below using precipitation as an illustrative example:

1. 1)
Bias Correction: Observed precipitation data is remapped to align with the GCM grid. For each day of the year, values from both the GCMs and observed datasets are pooled within a ±15-day window to construct a quantile mapping. Then the GCMs data points are transformed, correcting biases to match the distribution of the observed data. Since daily precipitation data typically lacks a discernible trend, detrending is unnecessary when employing this method.
2. 2)
Spatial Disaggregation: The coarse resolution GCM data is bilinearly interpolated to match the grid of the observation. To maintain the spatial nuances captured in high-resolution observations, the mean precipitation of each day of the years from the observation is calculated and employed as scaling factors. These scaling factors are then applied to the daily interpolated GCM projections by multiplication (or addition for three temperature variables), resulting in downscaled GCM projections.

The implementation of the daily BCSD approach in this study is structured into two primary phases:

- Phase 1 — Testing: The climatological fields from the training period spanning 1979-2004, derived from both the Observation (OBS) and GCMs, are utilized to establish transfer functions that link simulations with observation. Subsequently, the BCSD-master method is applied to an independent testing period covering 2005–2014. The outcomes are then compared against OBS to assess the efficacy of the downscaled data in replicating historical climate conditions.

- Phase 2 — Future downscaling: To maximize the utility of the BCSD approach’s training period, a comprehensive 36-year dataset from 1979 to 2014 is employed to develop the transfer functions. These are used to guide spatial disaggregation for the forthcoming period from 2015 to 2100. The BCSD method is applied to all GCMs, and Shared Socioeconomic Pathways (SSPs) scenarios listed in Table S1, thereby generating the targeted HiCPC dataset.

The process of generating of downscaled minimum daily temperature, which is derived from the difference between the downscaled maximum daily temperature and the downscaled diurnal average temperature range fields, as in Thrasher, et al. 23. It should be noted that our BCSD method, unlike NEX-CDDP-CMIP6, does not employ detrending on temperature variables. This approach is deliberate, with the objective of maintaining the original values of all temperature variables. In the subsequent section “Inspection of the projected change”, it is clear that our method continues to effectively retain the temperature trends of the GCM.

Quality control

A quality control process was applied to all downscaled output to check that values fell within a realistic range 23, 34. Table 1 shows the bounding values used to evaluate each variable. In the process of downscaling, a minor number of instances where precipitation values were less than 0 occurred. These values were uniformly adjusted to be set at 0. After undergoing the aforementioned quality control procedures, all downscaled simulations within the HiCPC datasets have successfully passed the four checks outlined in Table

Evaluation methods

To evaluate the quality of the downscaled data, several statistical and graphical techniques are employed. The performance of the downscaled data is benchmarked against the reference dataset using standard metrics such as the Pearson correlation coefficient (CORR), Root mean square error (RMSE), and Bias. Furthermore, the concept of Relative Added Value (RAV) is introduced to measure the enhanced performance of the downscaled data in comparison to the raw GCM outputs.

The generic Added Value (AV) 38, 39 is defined as a dimensional metric with units, which can be formulated as follows:

Here, X denotes one of the four variables that have been downscaled. The AV is then normalized by removing its units and dimensions to yield the RAV, which is calculated as:

Considering, t represents a specific time instance, while n denotes the aggregate number of such time instances. This formulation articulates the degree of enhancement in the downscaled data relative to the raw GCM output. The RAV is inherently less than 1. A positive RAV denotes an improvement where the downscaled data exhibits superior performance compared to the raw GCM data. Conversely, a negative RAV implies a diminished performance of the downscaled data in relation to the raw GCM data.

Data Records

The high-resolution China Daily Downscaled Projections (HiCPC) is available at the National Tibetan Plateau/Third Pole Environment Data Center ( https://doi.org/10.11888/Atmos.tpdc.301122). This dataset encompasses daily values of precipitation (pr) and average, maximum, and minimum temperatures (tas, tasmax, tasmin), spanning the historical period from 1979 to 2014 and future projections for 2015 to 2100 across four Shared Socioeconomic Pathways (SSPs) scenarios: SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5. Currently, the HiCPC dataset comprises downscaled data products from 22 CMIP6 GCMs. As computations are finalized, this dataset will continuously be expanded with additional model data in the future.

The data are formatted in the Network Common Data Form (netCDF). In addition to the standard daily products, users are also provided with monthly climatological data. Files with varying temporal resolutions are segregated into distinct directories. Each individual file, which represents a single variable’s time series, is substantial in size, approximately 60 GB each. The overall data volume for HiCPC is 14.3 TB (compressed to 4.17 TB). All datasets can be downloaded using wget or curl command. For example: curl -u download_4323096:32671394 -# -O ftp://ftp2.tpdc.ac.cn:6201/HiCPC/HiCPC_daily/pr/ACCESS-CM2.zip.

Technical Validation

Comparison of downscaled and GCM data

The high-resolution downscaled datasets are comparatively analyzed against OBS and GCM data over China for the testing period spanning from 2005 to 2014. Beyond generating high-resolution data, the efficacy of the BCSD method in mitigating biases and errors inherent to GCMs is also evaluated. Figure 1 shows a comparison of spatial distribution between OBS, raw GCM and downscaled (ACCESS-CM2) monthly total precipitation (pr), monthly mean average, maximum, minimum temperature (tas, tasmax, tasmin) in testing period. The CORR and RMSE between spatially averaged OBS and GCM/downscaled variables are also displayed in the upper right corner of each sub-figure. The downscaled data exhibits a spatial distribution that aligns more closely with OBS, demonstrating higher CORR and lower RMSE compared to the GCM. Additionally, the Probability Density Function (PDF) of the downscaled random variable, depicted in the bottom left corner of each sub-figure, aligns more closely with the OBS than the raw GCM. The multi-model ensemble mean (MME), as shown in Figure S1, similarly exhibits the aforementioned pattern.

Spatial distribution of the monthly total precipitation (pr) and average, maximum, minimum temperature (tas, tasmax, tasmin) over China during 2005–2014 by OBS (CMFD, left), raw GCM (ACCESS-CM2, mid) and downscaled GCM (ACCESS-CM2, right), respectively. The PDF of each variable is shown in the bottom left corner of each sub-figure. Average RMSE and CORR over the entire China territory are also displayed.

Figure 2 illustrates the biases of the raw GCM (left) and the downscaled GCM (right) in comparison to OBS for the testing period from 2005 to 2014. With respect to precipitation, the GCM exhibits significant positive biases over southern and western China, whereas the downscaled GCM substantially improves the GCM’s precipitation simulation. The mean bias is reduced from 23.86 mm to −2.13 mm. As for the three temperature variables, the GCM displays pronounced biases in regions of complex terrain, which are notably ameliorated by the downscaled data. Among these variables, the raw GCM shows the smallest average bias for tas, while after downscaling, the smallest average bias is observed for tasmax. Both the GCM and downscaled data reveal that tasmin has the largest average bias. Overall, compared to the downscaled data, the GCMs demonstrate a substantial bias across all variables. The multi-model ensemble mean (MME), as shown in Figure S2, similarly exhibits the aforementioned pattern. It is noteworthy that in certain regions, deviations of up to ± 5 °C are observed across three temperature variables, a pattern also evident in Figure

Spatial distribution of bias between raw GCM (ACCESS-CM2, left), downscaled GCM. (ACCESS-CM2, right) and OBS respectively for monthly total precipitation (pr) and average, maximum, minimum temperature (tas, tasmax, tasmin) during 2005–2014. Average bias of each variable is shown in the top left corner of each sub-figure.

We compare seasonal cycles of the GCM, the downscaled individual model (Downscaled), and OBS to examine further the ability of the downscaled data in reproducing the seasonal temperature/precipitation cycles of the testing period (Fig. 3). The downscaled data demonstrates a better agreement with OBS. On the contrary, the GCM displays a significant systematic divergence in the annual cycle. The reproducibility for seasonal cycles is examined by calculating the temporal correlations between the simulation and observation. The average correlation values for the downscaled data are higher than that of GCM, indicating the good performance of the downscaled products. The multi-model ensemble mean (MME), as shown in Figure S3, similarly exhibits the aforementioned pattern.

Comparison of the seasonal precipitation/temperature cycles according to OBS (black), GCM (ACCESS-CM2, blue), downscaled GCM (ACCESS-CM2, red) in the period 2005–2014. Values of average temporal correlation of each variable and OBS over China are also displayed.

Holistic Evaluation of downscaled effect

The comparison between the downscaled data and GCMs can clearly demonstrates the superiority of the downscaling method in eliminating biases and errors from GCMs. We primarily focus on the downscaling model’s capability to accurately reproduce the climatological features of the reference dataset. Figure 4 shows the global average (averaged over all grids) CORR, Bias and Normalized Root Mean Square Error (NRMSE) for all the models and variables. All downscaled models perform better than raw GCMs for all downscaled variables, expressing in higher correlations, lower average bias and NRMSE. Additionally, there is considerable variation in NRMSE, CORR, and Bias across different GCMs. Yet, these variations are notably diminished in the statistically downscaled models, suggesting that statistical downscaling effectively minimizes the uncertainty between models. This convergence is beneficial for employing ensemble approaches in diverse research settings.

The spatially averaged NRMSE, CORR and Bias between the downscaled models (Downscaled), (a) raw GCMs (GCM) and OBS for precipitation (pr), (b) average temperature (tas), (c) maximum temperature (tasmax), (d) and minimum temperature (tasmin). The top right corner of each sub-figure displays the average values for all GCMs (AVE_GCM) and downscaled models (AVE_Downscaled), respectively.

To confirm the superior performance of the downscaled models compared to the raw GCMs, we utilized the Relative Added Value (RAV) metric (Supplementary Figures S4- 7). Positive RAV grids dominate the majority of China across all downscaled model outputs. This is also evident from the distribution of PDF, where the RAV for all models is predominantly within the 0-1 range. For precipitation (Supplementary Figure S4), high RAV areas in most models are primarily located in Central and Western regions of China, particularly around the periphery of the Tibetan Plateau, indicating a good downscaling effect. However, some models exhibit weak negative RAV values in southeastern China. Overall, the average RAV for precipitation across all models reaches 0.42, demonstrating considerable added value over China (Fig. 5a). For the three temperature variables (Fig. 5b–d), the overall RAV in China is greater than 0, with better downscaling effects (closer to 1) in regions with complex terrain. Nevertheless, some areas along the northern edge of the Tibetan Plateau show significant negative RAV values. In summary, the average RAV for both precipitation and temperature exceeds 0.4.

Average RAVs of downscaled multi-models for precipitation (pr), (a) average temperature (tas), (b) maximum temperature (tasmax), (c) and minimum temperature (tasmin), (d) over the testing period. The PDF of average RAVs is shown in the bottom left corner of each sub-figure. Positive (negative) RAVs indicate better (worse) performance of the downscaled model.

Evaluation of extreme climate

To evaluate the efficacy of the downscaling in delineating extreme climate, we identified four extreme indices (refer to Table 2), as proposed by the Expert Team on Climate Change Detection and Indices (ETCCDI) 40, 41. The CDD and CWD indices represent the frequency characteristics of extreme precipitation, while the SU and ID indices denote the frequency characteristics of extreme temperature. Figure 6 shows a comparison of the spatial distribution of extreme indices calculated by OBS, GCM_MME, and downscaled_MME respectively during the testing period. The downscaled ensemble mean exhibits higher CORR and lower RMSE compared to the GCM ensemble mean.

Spatial distribution of the CDD, CWD, ID and SU over China during 2005–2014 by OBS (CMFD, left), GCM_MME (mid) and downscaled_MME (right), respectively. The PDF of each variable is shown in the bottom left corner of each sub-figure. Average RMSE and CORR over the entire China territory are also displayed.

Inspection of the projected change

It is hoped that the downscaled outputs will correspond with the future trend projections of GCM, thereby necessitating appropriate verification. Focusing on the entire 1979-2100 period of Phase 2, we validated the consistency of the trends in temperature/precipitation variables projected by HiCPC and GCMs, as well as the trends in extreme indices derived from them.The ACCESS-CM2 model under the SSP5-8.5 scenario was utilized as an illustrative example. A comparison of downscaled outputs with observations (CMFD) and GCM projections (Fig. 7) illustrates that the downscaled results effectively retain the intrinsic trends of the original GCM from 1979 to 2100, rather than aligning with OBS. The difference in annual precipitation trends is merely 0.19 mm/decade, and the temperature trend difference is only 0.01 °C/decade. Figure 7c–f illustrate the timeseries for the four extreme indices listed in Table 2. These suggest that the downscaled results adeptly retain the original GCM trends, and extend beyond merely rectifying systematic errors of the GCM. Frequency-related extreme indices show a discrepancy of no more than one day per decade over a hundred-year time scale compared to the original GCM calculations. Frequency-related extreme indices exhibit a discrepancy of no more than 1.0 day/decade comparing to the original GCM calculations.

Timeseries of annual total precipitation (a), annual mean temperature (b), CDD (c), CWD (d), ID (e), and SU (f) from 1979 to 2100. The linear trends for both GCM and Downscaled are displayed at the base of each subfigure.

Usage Notes

HiCPC provides high-resolution (10-km) daily climate change projections for China, derived from the CMIP6 GCMs, the number of which will continue to be updated and expanded. The dataset is designed to assist researchers and policymakers in understanding potential future climate change at a finer spatial scale. Here are some key points to consider when using this dataset:

1. 1.
The central assumption of statistical downscaling rests on the existence of a stable statistical relationship between coarse-scale and fine-scale climate variables
6, 42. Furthermore, it presupposes that the statistical relationships observed in historical climate data persist into future projections, thereby suggesting that the historical datasets employed in model construction are proxies for future climatic conditions 6, 42. Breaching these foundational assumptions can compromise the accuracy and dependability of downscaled outputs. Consequently, the appropriateness and suitability of these assumptions must be cautiously considered when using statistical downscaling datasets.
2. 2.
Apart from the enhanced spatial resolution and bias correction, this dataset does not contribute any additional information beyond what is already encapsulated in the raw CMIP6 scenarios. It preserves the frequency of periods characterized by anomalously high and low values, namely extreme events, within each distinct scenario.
3. 3.
BCSD heavily relies on correction factors derived from historical data to adjust future climate scenarios
34. These correction factors can vary across time and space, and their uncertainty may impact the accuracy of the downscaling results. Furthermore, while BCSD is designed to preserve the frequency of extreme events, it may not accurately simulate changes in these events

Code availability

The script files, developed with Python and Climate Data Operators (CDO), can be accessed on GitHub at https://github.com/Dpasang/HiCPC.

1.
Intergovernmental Panel on Climate, C. Climate Change 2021 – The Physical Science Basis: Working Group I Contribution to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. (Cambridge University Press, 2023).
2.
Eyring, V. et al. Taking climate model evaluation to the next level. Nature Climate Change 9, 102–110,
https://doi.org/10.1038/s41558-018-0355-y(2019).
3.
Shi, C., Jiang, Z.-H., Chen, W.-L. & Li, L. Changes in temperature extremes over China under 1.5 °C and 2 °C global warming targets. Advances in Climate Change Research 9, 120–129,
https://doi.org/10.1016/j.accre.2017.11.003(2018).
4.
Huang, R., Chen, J. & Huang, G. Characteristics and variations of the East Asian monsoon system and its impacts on climate disasters in China. Advances in Atmospheric Sciences 24, 993–1023,
https://doi.org/10.1007/s00376-007-0993-x(2007).
5.
Lian, X., Jiao, L., Hu, Y. & Liu, Z. Future climate imposes pressure on vulnerable ecological regions in China. Science of The Total Environment 858, 159995,
https://doi.org/10.1016/j.scitotenv.2022.159995(2023).
6.
Yang, Y., Tang, J., Xiong, Z., Wang, S. & Yuan, J. An intercomparison of multiple statistical downscaling methods for daily precipitation and temperature over China: present climate evaluations. Climate Dynamics 53, 4629–4649,
https://doi.org/10.1007/s00382-019-04809-x(2019).
7.
Jiang, D., Hu, D., Tian, Z. & Lang, X. Differences between CMIP6 and CMIP5 Models in Simulating Climate over China and the East Asian Monsoon. Advances in Atmospheric Sciences 37, 1102–1118,
https://doi.org/10.1007/s00376-020-2034-y(2020).
8.
Eyring, V. et al. Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization. Geosci. Model Dev. 9, 1937–1958,
https://doi.org/10.5194/gmd-9-1937-2016(2016).
9.
Knutti, R. & Sedláček, J. Robustness and uncertainties in the new CMIP5 climate model projections. Nature Climate Change 3, 369–373,
https://doi.org/10.1038/nclimate1716(2013).
10.
Meenu, R., Rehana, S. & Mujumdar, P. P. Assessment of hydrologic impacts of climate change in Tunga–Bhadra river basin, India with HEC-HMS and SDSM. Hydrological Processes 27, 1572–1589,
https://doi.org/10.1002/hyp.9220(2013).
11.
Gutmann, E. D. et al. A Comparison of Statistical and Dynamical Downscaling of Winter Precipitation over Complex Terrain. Journal of Climate 25, 262–281,
https://doi.org/10.1175/2011JCLI4109.1(2012).
12.
Ahmed, K. F. et al. Statistical downscaling and bias correction of climate model outputs for climate change impact assessment in the U.S. northeast. Global and Planetary Change 100, 320–332,
https://doi.org/10.1016/j.gloplacha.2012.11.003(2013).
13.
Wilby, R. L., Dawson, C. W. & Barrow, E. M. sdsm — a decision support tool for the assessment of regional climate change impacts. Environmental Modelling & Software 17, 145–157,
https://doi.org/10.1016/S1364-8152(01)00060-3(2002).
14.
Gebrechorkos, S. H., Hülsmann, S. & Bernhofer, C. Statistically downscaled climate dataset for East Africa. Scientific Data 6, 31,
https://doi.org/10.1038/s41597-019-0038-1(2019).
15.
Tavakol-Davani, H., Nasseri, M. & Zahraie, B. Improved statistical downscaling of daily precipitation using SDSM platform and data-mining methods. International Journal of Climatology 33, 2561–2578,
https://doi.org/10.1002/joc.3611(2013).
16.
Khan, M. S. & Coulibaly, P. Assessing Hydrologic Impact of Climate Change with Uncertainty Estimates: Bayesian Neural Network Approach. Journal of Hydrometeorology 11, 482–495,
https://doi.org/10.1175/2009JHM1160.1(2010).
17.
Gebrechorkos, S. H., Bernhofer, C. & Hülsmann, S. Impacts of projected change in climate on water balance in basins of East Africa. Sci Total Environ 682, 160–170,
https://doi.org/10.1016/j.scitotenv.2019.05.053(2019).
18.
Gebrechorkos, S. H., Bernhofer, C. & Hülsmann, S. Climate change impact assessment on the hydrology of a large river basin in Ethiopia using a local-scale climate modelling approach. Science of The Total Environment 742, 140504,
https://doi.org/10.1016/j.scitotenv.2020.140504(2020).
19.
Gebrechorkos, S. H., Taye, M. T., Birhanu, B., Solomon, D. & Demissie, T. Future Changes in Climate and Hydroclimate Extremes in East Africa. Earth’s Future 11, e2022EF003011,
https://doi.org/10.1029/2022EF003011(2023).
20.
Abdollahipour, A., Ahmadi, H. & Aminnejad, B. A review of downscaling methods of satellite-based precipitation estimates. Earth Science Informatics 15, 1–20,
https://doi.org/10.1007/s12145-021-00669-4(2022).
21.
Murali, G., Iwamura, T., Meiri, S. & Roll, U. Future temperature extremes threaten land vertebrates. Nature 615, 461–467,
https://doi.org/10.1038/s41586-022-05606-z(2023).
22.
Yuan, H.-H. et al. Evaluation of precipitation extremes over the Tibetan plateau using the NASA global daily downscaled datasets NEX-GDDP-CMIP6. Advances in Climate Change Research 14, 884–895,
https://doi.org/10.1016/j.accre.2023.12.001(2023).
23.
Thrasher, B. et al. NASA Global Daily Downscaled Projections, CMIP6. Scientific Data 9, 262,
https://doi.org/10.1038/s41597-022-01393-4(2022).
24.
Lange, S. Trend-preserving bias adjustment and statistical downscaling with ISIMIP3BASD (v1.0). Geosci. Model Dev. 12, 3055–3070,
https://doi.org/10.5194/gmd-12-3055-2019(2019).
25.
Gergel, D. R. et al. Global downscaled projections for climate impacts research (GDPCIR): preserving extremes for modeling future climate impacts. EGUsphere 2023, 1–35,
https://doi.org/10.5194/egusphere-2022-1513(2023).
26.
He, J. et al. The first high-resolution meteorological forcing dataset for land process studies over China. Scientific Data 7, 25,
https://doi.org/10.1038/s41597-020-0369-y(2020).
27.
Luo, Y., Yuan, H. & Ning, L. HiCPC: High-resolution CMIP6 downscaled daily climate projections over China 1979-2100. National Tibetan Plateau / Third Pole Environment Data Center.
https://doi.org/10.11888/Atmos.tpdc.301122(2024).
28.
Luan, L. & Zhai, P. Changes in rainy season precipitation properties over the Qinghai-Tibet Plateau based on multi-source datasets. Advances in Climate Change Research 19, 173 (2023).
29.
Huang, D. Q., Zhu, J., Zhang, Y. C. & Huang, A. N. Uncertainties on the simulated summer precipitation over Eastern China from the CMIP5 models. J Geophys Res-Atmos 118, 9035–9047,
https://doi.org/10.1002/jgrd.50695(2013).
30.
Yang, J., Huang, M. & Zhai, P. Performance of the CRA-40/Land, CMFD, and ERA-Interim datasets in reflecting changes in surface air temperature over the Tibetan Plateau. Journal of Meteorological Research 35, 663–672 (2021).
31.
Lei, N. et al. Performance Evaluation and Improvement of CMFD’s Precipitation Products Over Shanghai City, China. Earth and Space Science 10, e2022EA002690,
https://doi.org/10.1029/2022EA002690(2023).
32.
Wood, A. W., Leung, L. R., Sridhar, V. & Lettenmaier, D. P. Hydrologic Implications of Dynamical and Statistical Approaches to Downscaling Climate Model Outputs. Climatic Change 62, 189–216,
https://doi.org/10.1023/B:CLIM.0000013685.99609.9e(2004).
33.
Wood, A. W., Maurer, E. P., Kumar, A. & Lettenmaier, D. P. Long-range experimental hydrologic forecasting for the eastern United States. Journal of Geophysical Research: Atmospheres 107, ACL 6-1-ACL 6-15,
https://doi.org/10.1029/2001JD000659(2002).
34.
Thrasher, B., Maurer, E. P., McKellar, C. & Duffy, P. B. Technical Note: Bias correcting climate model simulated daily temperature extremes with quantile mapping. Hydrol. Earth Syst. Sci. 16, 3309–3314,
https://doi.org/10.5194/hess-16-3309-2012(2012).
35.
Falchetta, G., De Cian, E., Sue Wing, I. & Carr, D. Global projections of heat exposure of older adults. Nature Communications 15, 3678,
https://doi.org/10.1038/s41467-024-47197-5(2024).
36.
Rao, K. K. et al. Future changes in the precipitation regime over the Arabian Peninsula with special emphasis on UAE: insights from NEX-GDDP CMIP6 model simulations. Scientific Reports 14, 151,
https://doi.org/10.1038/s41598-023-49910-8(2024).
37.
Eum, H.-I., Cannon, A. J. & Murdock, T. Q. Intercomparison of multiple statistical downscaling methods: multi-criteria model selection for South Korea. Stochastic Environmental Research and Risk Assessment 31, 683–703,
https://doi.org/10.1007/s00477-016-1312-9(2017).
38.
Vandal, T., Kodra, E. & Ganguly, A. R. Intercomparison of machine learning methods for statistical downscaling: the case of daily and extreme precipitation. Theor Appl Climatol 137, 557–570,
https://doi.org/10.1007/s00704-018-2613-3(2019).
39.
Tran-Anh, Q., Ngo-Duc, T., Espagne, E. & Trinh-Tuan, L. A 10-km CMIP6 downscaled dataset of temperature and precipitation for historical and future Vietnam climate. Scientific Data 10, 257,
https://doi.org/10.1038/s41597-023-02159-2(2023).
40.
Karl, T. R., Nicholls, N. & Ghazi, A. in Weather and Climate Extremes: Changes, Variations and a Perspective from the Insurance Industry (eds Thomas R. Karl, Neville Nicholls, & Anver Ghazi) 3-7 (Springer Netherlands, 1999).
41.
Office, I. C. P. Report on the activities of the Working Group on Climate Change Detection and Related Rapporteurs 1998-2001. International CLIVAR Project Office (2001).
42.
Wilby, R. L. & Dawson, C. W. The Statistical DownScaling Model: insights from one decade of application. International Journal of Climatology 33, 1707–1719,
https://doi.org/10.1002/joc.3544(2013).

Acknowledgements

This research is supported by the Second Tibetan Plateau Scientific Expedition and Research Program (2019QZKK1001) and the National Natural Science Foundation of China (U2242207). We acknowledge the World Climate Research Programme, which, through its Working Group on Coupled Modelling, coordinated and promoted CMIP6. We thank the climate modeling groups for producing and making available their model output, the Earth System Grid Federation (ESGF) for archiving the data and providing access, and the multiple funding agencies who support CMIP6 and ESGF.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

About this article

Cite this article

Yuan, H., Ning, L., Zhou, J. et al. HiCPC: A new 10-km CMIP6 downscaled daily climate projections over China. Sci Data 11, 1167 (2024). https://doi.org/10.1038/s41597-024-03982-x

-
Received:
-
Accepted:
-
Published:
-
DOI: https://doi.org/10.1038/s41597-024-03982-x