Tag: Climate Change Downscaling Techniques

Abstract

Regional climate models (RCMs) exhibit greater potential than global models (GCMs) in capturing geographical details of climate change arising from orography and land–water distribution, but dynamical downscalings are only available for a limited number of GCMs. The full GCM ensembles are much more representative. Furthermore, the current EURO-CORDEX RCM runs most likely underestimate future warming. Thus, neither GCMs nor RCMs as such constitute an ideal tool for preparing reliable spatially detailed climate projections. This study introduces an easy-to-use GCM-RCM hybrid method that takes advantage of the best properties of both model categories. The large-scale response is adopted from GCM simulations, but the pattern is enriched with RCM-simulated details. For temperature projections, the procedure resembles the conventional pattern-scaling technique. However, the spatial averages of temperature change used for scaling are calculated over an area surrounding each grid point, either by giving an equal weight to the entire area or by taking into account the land–sea distribution. For precipitation, a linearised version of the method has been formulated. The method is demonstrated by integrating spatial details from 12 EURO-CORDEX RCM simulations with the CMIP6 multi-GCM mean projection. The resulting temperature responses include RCM-generated spatial details of up to \(\sim\) 1 \(^\circ\) C while effectively correcting the general tendency of RCMs to underestimate warming in Europe. For precipitation, geographical details originating from the different CORDEX runs tend to diverge, resulting in a low signal-to-noise ratio. This probably reflects the substantial impact of internal variability on small-scale changes in precipitation.

Abstract

The scale-dependent predictability of the devastating 7·20 extreme rainstorm in Zhengzhou, China in 2021 was investigated via ensemble experiments, which were perturbed on different scales using the stochastic kinetic-energy backscatter (SKEB) scheme in the WRF model, with the innermost domain having a 3-km grid spacing. The daily rainfall (RAIN24h) and the cloudburst during 1600–1700 LST (RAIN1h) were considered. Results demonstrated that with larger perturbation scales, the ensemble spread for the rainfall maximum widens and rainfall forecasts become closer to the observations. In ensembles with mesoscale or convective-scale perturbations, RAIN1h loses predictability at scales smaller than 20 km and RAIN24h is predictable for all scales. Whereas in ensembles with synoptic-scale perturbations, the largest scale of predictability loss extends to 60 km for both RAIN1h and RAIN24h. Moreover, the average positional error in forecasting the heaviest rainfall for RAIN24h (RAIN1h) was 400 km (50–60) km. The southerly low-level jet near Zhengzhou was assumed to be directly responsible for the forecast uncertainty of RAIN1h. The rapid intensification in low-level cyclonic vorticity, mid-level divergence, and upward motion concomitant with the jet dynamically facilitated the cloudburst. Further analysis of the divergent, rotational and vertical kinetic spectra and the corresponding error spectra showed that the error kinetic energy at smaller scales grows faster than that at larger scales and saturates more quickly in all experiments. Larger-scale perturbations not only boost larger-scale error growth but are also conducive to error growth at all scales through a downscale cascade, which indicates that improving the accuracy of larger-scale flow forecast may discernibly contributes to the forecast of cloudburst intensity and position.

Abstract

The aim of this research is to project climate change-induced rainfall for three future periods using three dynamically downscaling regional climate models (RCM1–RCM3) based on three respective global climate models: ICHEC-ECEARTH, MPI-M-MPI-ESM-MR and NOAA-GFDL-GFDL-ESM2M. The projections are carried out under two representative concentration pathways: RCP 4.5 and 8.5. The three future periods include near- (2022–2040), mid- (2041–2060) and far-future (2061–2099). The study area is Thailand’s central province of Pathumthani, which is a low-lying area and prone to flash flooding. The projected climate-induced rainfall is measured by five future extreme climate indexes: consecutive dry days (CDD), number of heavy precipitation days (R10), number of very heavy precipitation days (R20), consecutive wet days (CWD), and maximum 5-day precipitation amount (RX5day). The findings show that Pathumthani is increasingly susceptible to future flooding, as indicated by lower CDD and higher CWD. The lower CDD (decreasing from 77 days to 38–45 days) and higher CWD (increasing from 7 days to 21–22 days) suggest that Pathumthani is more likely to have more rainfall in the future. In addition, land use and land cover change (LULCC) contributes to persistent flooding in the province. The province’s rapid urbanization results in higher susceptibility to flooding as agricultural land is converted into urban infrastructure, commercial, industrial and residential areas. The conversion of repetitively flooded agricultural areas into urban areas also aggravates the flood situation. To mitigate the impact of climate change-induced floods, provincial authorities should implement non-structural anti-flood strategies (i.e., flood adaptation strategies), in addition to existing structural anti-flood measures. This research is the first to employ the dynamically downscaling regional climate models and LULCC to project future rainfall and flood risks. The projection techniques are also applicable to different geographical settings that are prone to flooding.

Abstract

Groundwater development is key to accelerating agricultural growth and to achieving food security in a climate crisis. However, the rapid increase in groundwater exploitation over the past four decades has resulted in depletion and degradation, particularly in regions already facing acute water scarcity, with potential irreversible impacts for food security and economic prosperity. Using a climate–water–food systems modelling framework, we develop exploratory scenarios and find that halting groundwater depletion without complementary policy actions would adversely affect food production and trade, increase food prices and grow the number of people at risk of hunger by 26 million by 2050. Supportive policy interventions in food and water systems such as increasing the effective use of precipitation and investments in agricultural research and development could mitigate most negative effects of sustainable groundwater use on food security. In addition, changing preferences of high-income countries towards less-meat-based diets would marginally alleviate pressures on food price. To safeguard the ability of groundwater systems to realize water and food security objectives amidst climate challenges, comprehensive measures encompassing improved water management practices, advancements in seed technologies and appropriate institutions will be needed.

Abstract

Foliar fungi on urban trees are important for tree health, biodiversity and ecosystem functioning. Yet, we lack insights into how urbanization influences foliar fungal communities. We created detailed maps of Stockholm region’s climate and air quality and characterized foliar fungi from mature oaks (Quercus robur) across climatic, air quality and local habitat gradients. Fungal richness was higher in locations with high growing season relative humidity, and fungal community composition was structured by growing season maximum temperature, NO₂ concentration and leaf litter cover. The relative abundance of mycoparasites and endophytes increased with temperature. The relative abundance of pathogens was lowest with high concentrations of NO₂ and particulate matter (PM_2.5), while saprotrophs increased with leaf litter cover. Our findings show that urbanization influences foliar fungi, providing insights for developing management guidelines to promote tree health, prevent disease outbreaks and maintain biodiversity within urban landscapes.

Abstract

This study examined how Land Use Land Cover (LULC) and climate affect groundwater recharge in Southern India’s semi-arid region. Coupled Model Intercomparison Project Phase6-Global Circulation Models (CMIP6-GCMs) climatic data is used to generate climate projections for the future. The GCMs are ranked for precipitation and temperatures using the Taylor Skill Score (TSS). Rating Metric (RM) is preferred to establish the final rank of the GCMs. Ensemble of projections from the top four ranked GCMs: Max Planck Institute Earth System Model version 1.2 - Low Resolution (MPI-ESM1-2-LR), European Consortium Earth System Model version 3 (EC-Earth3), Max Planck Institute Earth System Model version 1.2 - High Resolution (MPI-ESM1-2-HR), and Institute for Numerical Mathematics Coupled Model version 5 (INM-CM5-0) are used as they estimated the most reliable forecasts for all the three considered parameters. MPI-ESM1-2-LR is the top-ranked GCM with an RM of 0.92. The future LULC map is produced using Cellular Automata and Artificial Neural Networks (CA-ANN). Soil and Water Assessment Tool (SWAT) is used to evaluate the individual impact of climate change on groundwater recharge and the combined impacts of LULC and climate change on groundwater recharge in water-stressed regions (semi-arid) using standard modelling techniques. The SWAT model has been calibrated using monthly discharge data from a gauging station, resulting in an overall accuracy of R² = 0.83 and NSE = 0.81. The SWAT groundwater module is employed to estimate recharge across different time frames: baseline (1985–2014), near-future (2015–2030), mid-future (2031–2060) and far-future (2061–2100), considering moderate (SSP2-4.5) and extreme (SSP5-8.5) emission scenarios. Results indicate that under constant LULC conditions, recharge varied between 135 and 215 mm/year for SSP2-4.5 and 149 to 316 mm/year for SSP5-8.5. Notably, compared to the baseline recharge of 116.4 mm, future groundwater recharge increased under both SSP scenarios. The observed variations in recharge carry significant implications for understanding how varied emission scenarios may impact groundwater resources over the specified time periods.

Abstract

As our planet is entering into the “global boiling” era, understanding regional climate change becomes imperative. Effective downscaling methods that provide localized insights are crucial for this target. Traditional approaches, including computationally-demanding regional dynamical models or statistical downscaling frameworks, are often susceptible to the influence of downscaling uncertainty. Here, we address these limitations by introducing a diffusion probabilistic downscaling model (DPDM) into the meteorological field. This model can efficiently transform data from 1° to 0.1° resolution. Compared with deterministic downscaling schemes, it not only has more accurate local details, but also can generate a large number of ensemble members based on probability distribution sampling to evaluate the uncertainty of downscaling. Additionally, we apply the model to generate a 180-year dataset of monthly surface variables in East Asia, offering a more detailed perspective for understanding local scale climate change over the past centuries.

Abstract

Access to high-resolution historical climate data is vital for both theoretical and applied climatology. The aim of this paper is to propose a hybrid algorithm, which consists of two main steps. At the first step, using temperature lapse rate, the temperature data of ERA5, MERRA2 and CFS were downscaled. The temperature lapse rate was calculated using MODIS temperature data. Then, at the second step, the ability of machine learning algorithms was examined in generating daily 2m temperature data with spatial resolution of 1km for Iran during 1980-2021. The inputs of the machine learning algorithms include the downscaled outputs of ERA5, MERRA2 and CFS data (which were calculated at the first step), elevation, aspect and Julian day. Several machine learning algorithms were examined including multi-layered perceptron neural networks (MLPNN), Random Forest (RF), Gradient Boosting (GB), Adaptive Boosting (AB) and Cubist Regression (CB). The results indicated that the MLPNN outperforms other techniques. Also, the final temperature map was depicted for Feb and Jul and also the MLPNN output predictions were evaluated for six different climate regions in the country. Evaluation results in regional section indicated that the MLPNN best performance was in very warm region with low rainfall while its worst performance was in mountainous and cold region.

Abstract

Climate projections at fine spatial resolutions are required to conduct accurate risk assessment for critical infrastructure and design adaptation planning. Generating these projections using advanced Earth system models (ESM) requires significant computational resources. To address this issue, various statistical downscaling techniques have been introduced to generate fine-resolution data from coarse-resolution simulations. In this study, we evaluate and compare five deep learning-based downscaling techniques, namely, super-resolution convolutional neural networks, fast super-resolution convolutional neural network ESM, efficient sub-pixel convolutional neural network, enhanced deep residual network (EDRN), and super-resolution generative adversarial network (SRGAN). These techniques are applied to a dataset generated by the Energy Exascale Earth System Model (E3SM), focusing on key surface variables such as surface temperature, shortwave heat flux, and longwave heat flux. Models are trained and validated using paired fine-resolution (0.25 \(^{\circ }\) ) and coarse-resolution (1 \(^{\circ }\) ) monthly data obtained from a 9-year simulation. Next, blind testing is performed using monthly data obtained from two different years outside of the training and validation set. To evaluate the efficiency of each technique, different statistical metrics are used, including mean squared error (MSE), peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and learned perceptual image patch similarity (LPIPS). The results show that EDRN outperforms other algorithms in terms of PSNR, SSIM, and MSE, but struggles to capture fine-scale features in the data. In contrast, SRGAN, a generative model that uses perceptual loss, excels in capturing fine details at boundaries and internal structures, resulting in lower LPIPS than other methods.

Abstract

The study assesses the predictability of rainfall patterns in India through 3-day precipitation forecasts from a regional climate model ensemble framework operating at convection-permitting (CP) scales. Initially, 149 experiments are conducted across four events representing different rainfall mechanisms. The performance of a larger set of 55 ensemble members within the multi-physics ensemble framework is evaluated using quantitative metrics such as composite scaled scores and cross-correlation analyses. This evaluation led to the development of an optimally designed smaller member ensemble framework, WRF-CP7, which reduces turnaround time while maintaining the spatial and temporal performance of simulated precipitation fields. The study further assesses the reliability of this framework over an extended period, utilizing insights from 5544 simulations (792 days × 7 ensemble members running for a 90-h lead time) conducted between September 2015 and December 2017. Comparisons between WRF-CP7 and a global climate model forecasts available at coarser resolution highlight the need of parameterization and ensemble framework at convection-permitting scale. WRF-CP7 demonstrates skill in capturing spatiotemporal variability of rainfall occurrences, evidenced by a higher spread–error correlation (0.9 vs. 0.6 in the global model) among ensemble members. The correlation remains consistent even at higher lead-times, in contrast to the reducing skill of the global model with increasing lead-time. WRF-CP7 also shows reduction in spatial and temporal errors within simulated diurnal precipitation patterns, notably during Indian Summer Monsoon, Pre-Monsoon Thunderstorm activities and North-East Monsoon. A notable 30% increase in predictability for moderate to heavy rain intensities is observed across all seasons, accompanied by a 10% decrease in false alarms compared to global model ensemble forecasts. The spatial skill of WRF-CP7 for moderate-heavy intensity events remains high (50–80%) even with a longer lead time of 72-h on an intra-seasonal timescale. With a substantial sample size, the results underscore the effectiveness of using the multi-physics ensemble framework at convective scales for operational forecasting and dynamic downscaling of climatology across the Indian subcontinent.