Search Results (154)

Central Asia, as a chronically water-stressed region marked by extreme aridity, faces significant environmental challenges from intensifying desertification and deteriorating ecological stability. The region’s vulnerability to shifting precipitation regimes and extreme hydrometeorological events has been magnified under anthropogenic climate forcing. Although global climate models (GCMs) remain essential tools for climate projections, their utility in Central Asia’s complex terrain is constrained by inherent limitations: coarse spatial resolution (~100–250 km) and imperfect parameterization of orographic precipitation mechanisms. This investigation advances precipitation modeling through deep learning-enhanced statistical downscaling, employing convolutional neural networks (CNNs) to generate high-resolution precipitation data at approximately 10 km resolution. Our results show that the deep learning models successfully simulate the high center of precipitation and extreme precipitation near the Tianshan Mountains, exhibiting high spatial applicability. Under intermediate (SSP-245) and high-emission (SSP-585) future scenarios, the increase in extreme precipitation over the next century is significantly more pronounced compared to mean precipitation. By the end of the 21st century, the interannual variability of mean precipitation and extreme precipitation will become even larger under SSP-585, indicating an increased risk of extreme droughts/floods in Central Asia under high greenhouse gas emissions. Our findings provide technical support for climate change impact assessments in the region and highlight the potential of CNN-based downscaling for future climate change studies. Full article

Wetlands in the Yellow River Watershed of Inner Mongolia face significant reductions under future climate and land use scenarios, threatening vital ecosystem services and water security. This study employs high-resolution projections from NASA’s Global Daily Downscaled Projections (GDDP) and the Intergovernmental Panel on Climate Change Sixth Assessment Report (IPCC AR6), combined with a machine learning and Cellular Automata–Markov (CA–Markov) framework to forecast the land cover transitions to 2040. Statistically downscaled temperature and precipitation data for two Shared Socioeconomic Pathways (SSP2-4.5 and SSP5-8.5) are integrated with satellite-based land cover (Landsat, Sentinel-1) from 2007 and 2023, achieving a high classification accuracy (over 85% overall, Kappa > 0.8). A Maximum Entropy (MaxEnt) analysis indicates that rising temperatures, increased precipitation variability, and urban–agricultural expansion will exacerbate hydrological stress, driving substantial wetland contraction. Although certain areas may retain or slightly expand their wetlands, the dominant trend underscores the urgency of spatially targeted conservation. By synthesizing downscaled climate data, multi-temporal land cover transitions, and ecological modeling, this study provides high-resolution insights for adaptive water resource planning and wetland management in ecologically sensitive regions. Full article

Bangladesh, one of the most vulnerable countries to climate change, has been experiencing significant climate change-induced risks. Particularly, the northwest region of the country has been severely affected by climate extremes, including droughts and heat waves. Therefore, proper understanding and assessment of future climate change scenarios is crucial for the adaptive management of water resources. The current study used the statistical downscaling model (SDSM) to downscale and analyze climate change-induced future changes in temperature and precipitation based on multiple global climate models (GCMs), including HadCM3, CanESM2, and CanESM5. A quantitative approach was adopted for both calibration and validation, showing that the SDSM is well-suited for downscaling mean temperature and precipitation. Furthermore, bias correction was applied to enhance the accuracy of the downscaled climate variables. The downscaled projections revealed an upward trend in mean annual temperatures, while precipitation exhibited a declining trend up to the end of the century for all scenarios. The observed data periods for the CanESM5, CanESM2, and HadCM3 GCMs used in SDSM were 1985–2014, 1975–2005, and 1975–2001, respectively. Based on the aforementioned periods, the projections for the next century indicate that under the CanESM5 (SSP5-8.5 scenario), temperature is projected to increase by 0.98 °C, with a 12.4% decrease in precipitation. For CanESM2 (RCP8.5 scenario), temperature is expected to rise by 0.94 °C, and precipitation is projected to decrease by 10.3%. Similarly, under HadCM3 (A2 scenario), temperature is projected to increase by 0.67 °C, with a 7.0% decrease in precipitation. These downscaled pathways provide a strong basis for assessing the potential impacts of future climate change across the northwestern region of Bangladesh. Full article

The current study is essential for obtaining an accurate representation of weather conditions in the Ziz and Guir watersheds, characterized by an arid climate. This study combined climate data from the ERA5 model with data from observation stations in order to evaluate the ERA5 model in Morocco’s arid environment and increase the temporal and geographical coverage of climate data. From the data collected, precipitation, minimum and maximum temperatures were predicted under the RCP4.5 and RCP8.5 scenarios by applying the SDSM model in the two watersheds for the 2025 and 2100 periods. These forecasts contribute to the development of adaptation strategies in the face of climate change by giving precise indications of future trends and providing local communities with tools for enhancing their resilience capacity. At all climatic stations, the temperature changes predicted under these scenarios showed a marked positive trend for both minimum and maximum temperatures. By the end of the century, minimum temperatures may increase by 1.84 °C and 2.39 °C under the RCP4.5 and RCP8.5 scenarios, respectively. Similarly, maximum temperatures may increase by 1.78 °C and 2.9 °C under the RCP4.5 and RCP8.5 scenarios, respectively. In addition, the precipitation forecast under the RCP 4.5 scenario showed a significant negative trend at the Ait Haddou station, while under the RCP 8.5 scenario, significant negative trends were predicted for the Sidi Hamza, Ait Haddou, Tit N’Aissa, and Bouanane stations. Full article

The Yellow River Basin is an important grain-production base in China, playing a crucial role in the country’s agricultural production and overall national economy and social development. However, due to the impact of climate change, China’s food security is facing challenges. Therefore, this article takes the Yellow River Basin as an example to reveal the temporal and spatial evolution patterns of the main crop yields in the basin. Based on a coupled statistical downscaling model (SDSM) and ten General Circulation Models (GCMs) from CMIP5, it estimates the future temporal and spatial evolution characteristics of rainfall and evaporation in the basin. Furthermore, a distributed crop-growth model (AquaCrop) is constructed to reveal the temporal and spatial evolution patterns of agricultural irrigation water requirements from a future perspective, clarifying the impact of multi-source uncertainty on the prediction uncertainty of agricultural irrigation water needs. The results indicate that the ten climate models constructed in this study can be effectively applied to the Yellow River Basin, and their ability to capture light-rain events is superior to that of moderate- and heavy-rain events. The simulation accuracy of the AquaCrop model significantly improves with an increase in precipitation frequency. The agricultural irrigation water demand in the middle and upper reaches of the basin is greater than that in the lower reaches, and the uncertainties from GCMs and RCPs have a significant impact on the uncertainty of agricultural irrigation water demand. The research results provide important references for formulating agricultural development plans for irrigation areas under climate-change conditions and for developing response measures for irrigation areas to cope with climate change. Full article

In agricultural production, droughts occurring during the crucial growth periods of crops hinder crop development, while the daily-scale standardized precipitation evapotranspiration index (SPEI) can be applied to accurately identify the drought characteristics. In this study, we used the statistical downscaling method to obtain the daily precipitation (P_r), maximum air temperature (T_max) and minimum air temperature (T_min) during the rice growing season in Heilongjiang Province from 2015 to 2100 under the SSP1-2.6, SSP2-4.5 and SSP5-8.5 in CMIP6, to study the spatial and temporal characteristics of drought during the rice growing season in cold region and the effect of climate change on drought characteristics. The potential evapotranspiration (PET₀) was calculated using the regression correction method of the Hargreaves formula recommended by the FAO, and the daily SPEI was calculated to quantitatively identify the drought classification. The Pearson correlation coefficient was used to analyze the correlation between the meteorological factors (P_r, T_max, T_min), PET₀ and SPEI. The results showed that: (1) Under 3 SSP scenarios, P_r showed an increasing trend from the northwest to the southeast, T_max showed an increasing trend from the northeast to the southwest, and higher T_min was mainly distributed in the east and west regions. (2) PET₀ indicated an overall interannual rise in the three future SSP scenarios, with higher values mainly distributed in the central and western regions. The mean daily PET₀ values ranged from 4.8 to 6.0 mm/d. (3) Under SSP1-2.6, rice mainly experienced mild drought and moderate drought (−0.5 ≥ SPEI > −1.5). The predominant drought classifications experienced were mild, moderate, and severe drought under SSP2-4.5 and SSP8.5 (−0.5 ≥ SPEI > −2.0). (4) The tillering stage experienced the highest drought frequency and drought intensity, with the longest drought lasting 24 days. However, the heading flower stage had the lowest drought frequency and drought intensity. The drought barycenter was mainly in Tieli and Suihua. (5) The PET₀ was most affected by the T_max, while the SPEI was most affected by the P_r. This study offers a scientific and rational foundation for understanding the drought sensitivity of rice in Northeast China, as well as a rationale for the optimal scheduling of water resources in agriculture in the future. Full article

This study evaluated the effectiveness of the NASA Earth Exchange Global Daily Downscaled models from CMIP6 experiments (hereafter; NEX-GDDP-CMIP6) in reproducing observed precipitation and temperature across East Africa (EA) from 1981 to 2014. Additionally, climate changes were estimated under various emission scenarios, namely low (SSP1-2.6), medium (SSP2-4.5), and high (SSP5-8.5) scenarios. Multiple robust statistics metrics, the Taylor diagram, and interannual variability skill (IVS) were employed to identify the best-performing models. Significant trends in future precipitation and temperature are evaluated using the Mann-Kendall and Sen’s slope estimator tests. The results highlighted IPSL-CM6A-LR, EC-Earth3, CanESM5, and INM-CM4-8 as the best-performing models for annual and March to May (MAM) precipitation and temperature respectively. By the end of this century, MAM precipitation and temperature are projected to increase by 40% and 4.5 °C, respectively, under SSP5-8.5. Conversely, a decrease in MAM precipitation and temperature of 5% and 0.8 °C was projected under SSP2-4.5 and SSP1-2.6, respectively. Long-term mean precipitation increased in all climate scenarios (SSP1-2.6, SSP2-4.5, and SSP5-8.5), with near-term MAM precipitation showing a 5% decrease in Rwanda, Burundi, Uganda, and some parts of Tanzania. Under the SSP5-8.5 scenario, temperature rise exceeded 2–6 °C in most regions across the area, with the fastest warming trend of over 6 °C observed in diverse areas. Thus, high greenhouse gas (GHG) emission scenarios can be very harmful to EA and further GHG control is needed. Full article

In this research, we introduce a novel method leveraging the Transformer architecture to generate high-fidelity precipitation model outputs. This technique emulates the statistical characteristics of high-resolution datasets while substantially lowering computational expenses. The core concept involves utilizing a blend of coarse and fine-grained simulated precipitation data, encompassing diverse spatial resolutions and geospatial distributions, to instruct Transformer in the transformation process. We have crafted an innovative ST-Transformer encoder component that dynamically concentrates on various regions, allocating heightened focus to critical spatial zones or sectors. The module is capable of studying dependencies between different locations in the input sequence and modeling at different scales, which allows it to fully capture spatiotemporal correlations in meteorological element data, which is also not available in other downscaling methods. This tailored module is instrumental in enhancing the model’s ability to generate outcomes that are not only more realistic but also more consistent with physical laws. It adeptly mirrors the temporal and spatial distribution in precipitation data and adeptly represents extreme weather events, such as heavy and enduring storms. The efficacy and superiority of our proposed approach are substantiated through a comparative analysis with several cutting-edge forecasting techniques. This evaluation is conducted on two distinct datasets, each derived from simulations run by regional climate models over a period of 4 months. The datasets vary in their spatial resolutions, with one featuring a 50 km resolution and the other a 12 km resolution, both sourced from the Weather Research and Forecasting (WRF) Model. Full article

Land suitability analyses are crucial for identifying sustainable areas for agricultural crops and developing appropriate land use strategies. Thus, the present study aims to analyze the current and future land suitability for wheat (Triticum aestivum L.) cultivation in Ethiopia. Twelve variables including soil properties, climate variables, and topographic characteristics were used in the evaluation of land suitability. Statistical methods such as Rotated Empirical Orthogonal Functions (REOF), Coefficient of Variation (CV), correlation, and parametric and non-parametric trend analyses were used to analyze the spatiotemporal variability in current and future climate data and identified significant patterns of variability. For future projections of land suitability and climate, this study employed climate models from the Coupled Model Intercomparison Project Phase 6 (CMIP6) framework, downscaled using regional climate model version 4.7 (RegCM4.7) under two different Shared Socioeconomic Pathway (SSP) climate scenarios: SSP1 (a lower emission scenario) and SSP5 (a higher emission scenario). Under the current condition, during March, April, and May (MAM), 53.4% of the country was suitable for wheat cultivation while 44.4% was not suitable. In 2050, non-suitable areas for wheat cultivation are expected to increase by 1% and 6.9% during MAM under SSP1 and SSP5 climate scenarios, respectively. Our findings highlight that areas currently suitable for wheat may face challenges in the future due to altered temperature and precipitation patterns, potentially leading to shifts in suitable areas or reduced productivity. This study also found that the suitability of land for wheat cultivation was determined by rainfall amount, temperature, soil type, soil pH, soil organic carbon content, soil nitrogen content, and elevation. This research underscores the critical importance of integrating spatiotemporal climate variability with future projections to comprehensively assess wheat suitability. By elucidating the implications of climate change on wheat cultivation, this study lays the groundwork for developing effective adaptation strategies and actionable recommendations to enhance management practices. The findings support the county’s commitment to refining agricultural land use strategies, increasing wheat production through suitability predictions, and advancing self-sufficiency in wheat production. Additionally, these insights can empower Ethiopia’s agricultural extension services to guide farmers in cultivating wheat in areas identified as highly and moderately suitable, thereby bolstering production in a changing climate. Full article

The hydrological cycle is altered by climate change and human activities, amplifying extreme precipitation and heightening the flood risk regionally and globally. It is imperative to explore the future possible alterations in flood risk at the regional scale. Focusing on the Hanjiang river basin (HRB), this study develops a framework for establishing a scientific assessment of spatio-temporal dynamics of future flood risks under multiple future scenarios. In this framework, a GCMs statistical downscaling method based on machine learning is used to project future precipitation, the PLUS model is used to project future land use, the digitwining watershed model (DWM) is used to project future runoff, and the entropy weight method is used to calculate risk. Six extreme precipitation indices are calculated to project the spatio-temporal patterns of future precipitation extremes in the HRB. The results of this study show that the intensity (Rx1day, Rx5day, PRCPTOT, SDII), frequency (R20m), and duration (CWD) of future precipitation extremes will be consistently increasing over the HRB during the 21st century. The high values of extreme precipitation indices in the HRB are primarily located in the southeast and southwest. The future annual average runoff in the upper HRB during the near-term (2023–2042) and mid-term (2043–2062) is projected to decrease in comparison to the baseline period (1995–2014), with the exception of that during the mid-term under the SSP5-8.5 scenario. The high flood risk center in the future will be distributed in the southwestern region of the upper HRB. The proportions of areas with high and medium–high flood risk in the upper HRB will increase significantly. Under the SSP5-8.5 scenario, the area percentage with high flood risk during the future mid-term will reach 24.02%. The findings of this study will facilitate local governments in formulating effective strategic plans for future flood control management. Full article

The skill, assumptions, and uncertainty of machine learning techniques (MLTs) for downscaling global climate model’s precipitation to the local level in Bolivia were assessed. For that, an ensemble of 20 global climate models (GCMs) from CMIP6, with random forest (RF) and support vector machine (SVM) techniques, was used on four zones (highlands, Andean slopes, Amazon lowlands, and Chaco lowlands). The downscaled series’ skill was evaluated in terms of relative errors. The uncertainty was analyzed through variance decomposition. In most cases, MLTs’ skill was adequate, with relative errors less than 50%. Moreover, RF tended to outperform SVM. Robust (weak) stationary (perfect prognosis) assumptions were found in the highlands and Andean slopes. The weakness was attributed to topographical complexity. The downscaling methods were shown to be the dominant source of uncertainties. This analysis allowed the derivation of robust future projections, showing higher annual rainfall, shorter dry spell duration, and more frequent but less intense high rainfall events in the highlands. Apart from the dry spell’s duration, a similar pattern was found for the Andean slopes. A decrease in annual rainfall was projected in the Amazon lowlands and an increase in the Chaco lowlands. Full article

High-resolution precipitation forecasts play a pivotal role in formulating comprehensive disaster prevention and mitigation plans. As spatial resolution enhances, striking a balance between computation, storage, and simulation accuracy becomes imperative to ensure optimal cost-effectiveness. Convolutional neural networks (CNNs), a cornerstone of deep learning, are examined in this study for their downscaling capabilities in precipitation simulation. During a precipitation event on 23 June 2022, in Jiangsu Province, China, distinct rain belts emerged in both southern and northern Jiangsu, precisely captured by a numerical model (the Weather Research and Forecasting, WRF) with a 3 km spatial resolution. Specifically, precipitation was prevalent in northern Jiangsu from 00:00 to 11:00 Beijing Time (BJT), transitioning to southern Jiangsu from 12:00 to 23:00 BJT on the same day. Upon dynamic downscaling, the model reproduced precipitation in these periods with an average error of 12.35 mm at 3 km and 12.48 mm at 1 km spatial resolutions. Employing CNN technology for statistical downscaling to a 1 km spatial resolution, samples from the initial period were utilized for training, while those from the subsequent period served for validation. Following dynamic downscaling, CNNs with four, five, six, and seven layers exhibited average errors of 8.86 mm, 8.93 mm, 9.71 mm, and 9.70 mm, respectively, accompanied by correlation coefficients of 0.550, 0.570, 0.574, and 0.578, respectively. This analysis indicates that for this precipitation event, a shallower CNN depth yields a lower average error and correlation coefficient, whereas a deeper architecture enhances the correlation coefficient. By employing deep network architectures, CNNs are capable of capturing nonlinear patterns and subtle local features from complex meteorological data, thereby providing more accurate predictions during the downscaling process. Leveraging faster computation and reduced storage requirements, machine learning has demonstrated immense potential in high-resolution forecasting research. There is significant scope for advancing technologies that integrate numerical models with machine learning to achieve higher-resolution numerical forecasts. Full article

Predicting local precipitation patterns over the European Alps remains an open challenge due to many limitations. The complex orography of mountainous areas modulates climate signals, and in order to analyse extremes accurately, it is essential to account for convection, requiring high-resolution climate models’ outputs. In this work, we analyse local seasonal precipitation in Trento (Laste) and Passo Tonale using high-resolution climate data and neural network downscaling. Then, we adopt an ensemble and generalized leave-one-out cross-validation procedure, which is particularly useful for the analysis of small datasets. The application of the procedure allows us to correct the model’s bias, particularly evident in Passo Tonale. This way, we will be more confident in achieving more reliable results for future projections. The analysis proceeds, considering the mean and the extreme seasonal anomalies between the projections and the reconstructions. Therefore, while a decrease in the mean summer precipitation is found in both stations, a neutral to positive variation is expected for the extremes. Such results differ from model’s, which found a clear decrease in both stations in the summer’s mean precipitation and extremes. Moreover, we find two statistically significant results for the extremes: a decrease in winter in Trento and an increase in spring in Passo Tonale. Full article

This study presented an assessment of climate extremes in the Southeast Asia (SEA) region, utilizing downscaled climate projections from the Coupled Model Intercomparison Project Phase 6 (CMIP6) Global Climate Models (GCMs). The study outputs uncovered statistically significant trends indicating a rise in extreme precipitation and temperature events throughout SEA for both the near-term (2021–2060) and long-term (2061–2100) future under both SSP245 and SSP585 scenarios, in comparison to the historical period (1950–2014). Moreover, we investigated the seasonal fluctuations in rainfall and temperature distributions, accentuating the occurrence of drier dry seasons and wetter rainy seasons in particular geographic areas. The focused examination of seven prominent cities in SEA underscored the escalating frequency of extreme rainfall events and rising temperatures, heightening the urban vulnerability to urban flooding and heatwaves. This study’s findings enhance our comprehension of potential climate extremes in SEA, providing valuable insights to inform climate adaptation, mitigation strategies, and natural disaster preparedness efforts within the region. Full article

In this study, we explored the efficacy of random forest algorithms in downscaling CHIRPS (Climate Hazards Group InfraRed Precipitation with Station data) precipitation data to predict Argane stand traits. Nonparametric regression integrated original CHIRPS data with environmental variables, demonstrating enhanced accuracy aligned with ground rain gauge observations after residual correction. Furthermore, we explored the performance of range machine learning algorithms, encompassing XGBoost, GBDT, RF, DT, SVR, LR and ANN, in predicting the Leaf Area Index (LAI) and crop yield of Argane trees using condition index-based drought indices such as PCI, VCI, TCI and ETCI derived from multi-sensor satellites. The results demonstrated the superiority of XGBoost in estimating these parameters, with drought indices used as input. XGBoost-based crop yield achieved a higher R² value of 0.94 and a lower RMSE of 6.25 kg/ha. Similarly, the XGBoost-based LAI model showed the highest level of accuracy, with an R² of 0.62 and an RMSE of 0.67. The XGBoost model demonstrated superior performance in predicting the crop yield and LAI estimation of Argania sinosa, followed by GBDT, RF and ANN. Additionally, the study employed the Combined Drought Index (CDI) to monitor agricultural and meteorological drought over two decades, by combining four key parameters, PCI, VCI, TCI and ETCI, validating its accuracy through comparison with other drought indices. CDI exhibited positive correlations with VHI, SPI and crop yield, with a particularly strong and statistically significant correlation observed with VHI (r = 0.83). Therefore, CDI was recommended as an effective method and index for assessing and monitoring drought across Argane forest stands area. The findings demonstrated the potential of advanced machine learning models for improving precipitation data resolution and enhancing agricultural drought monitoring, contributing to better land and hydrological management. Full article