Search Results (122)

In the present study, we evaluated the potential use of a humic substance (HS)-based biostimulant in mitigating drought stress in lettuce (Lactuca sativa L.) by comparing both root and foliar modes of application. To achieve this, lettuce plants were grown in a growth chamber on a solid substrate composed of vermiculite and perlite (3:1). Plants were exposed to drought conditions (50% of Field Capacity, FC) and 50% FC + HS applied as radicular (‘R’) and foliar (‘F’) at concentrations: R-HS 0.40 and 0.60 mL/L, respectively, and 7.50 and 10.00 mL/L, respectively, along with a control (100% FC). HSs were applied three times at 10-day intervals. Plant growth, nutrient concentration, lipid peroxidation, reactive oxygen species (ROS), and enzymatic and non-enzymatic antioxidants were estimated. Various photosynthetic and chlorophyll fluorescence parameters were also analyzed. The results showed that HS applications alleviated drought stress, increased plant growth, and reduced lipid peroxidation and ROS accumulation. HSs also improved the net photosynthetic rate, carboxylation efficiency, electron transport flux, and water use efficiency. Although foliar HSs showed a greater tendency to enhance shoot growth and photosynthetic capacity, the differences between the application methods were not significant. Hence, in this preliminary work, the HS-based product evaluated in this study demonstrated potential for alleviating drought stress in lettuce plants at the applied doses, regardless of the mode of application. This study highlights HS-based biostimulants as an effective and sustainable tool to improve crop resilience and support sustainable agriculture under climate change. However, further studies under controlled growth chamber conditions are needed to confirm these results before field trials. Full article

With climate warming, the global precipitation patterns have undergone significant changes, which will profoundly impact flood–drought disaster regimes and socioeconomic development in key regions of human activity worldwide. The convergence zone of the Indian Ocean monsoon and Pacific monsoon in China covers most of the middle and lower reaches of the Yangtze River (MLRYR), which is located in the transitional area of the second and third steps of China’s terrain. Changes in precipitation patterns in this region will significantly impact flood and drought control in the MLRYR, as well as the socioeconomic development of the MLRYR Economic Belt. In this study, Huaihua area in China was selected as the study area to study the characteristics of regional precipitation change, and to analyze the evolution in the trends in annual precipitation, extreme precipitation events, and their spatiotemporal distribution, so as to provide a reference for the study of precipitation change patterns in the intersection zone. This study utilizes precipitation data from meteorological stations and the China Meteorological Forcing Dataset (CMFD) reanalysis data for the period 1979–2023 in Huaihua region. The spatiotemporal variation in precipitation in the study area was analyzed by using linear regression, the Mann–Kendall trend test, the moving average method, the Mann–Kendall–Sneyers test, wavelet analysis, and R/S analysis. The results demonstrate the following: (1) The annual precipitation in the study area is on the rise as a whole, the climate tendency rate is 9 mm/10 a, and the precipitation fluctuates greatly, showing an alternating change of “dry–wet–dry–wet”. (2) Wavelet analysis reveals that there are 28-year, 9-year, and 4-year main cycles in annual precipitation, and the precipitation patterns at different timescales are different. (3) The results of R/S analysis show that the future precipitation trend will continue to increase, with a strong long-term memory. (4) Extreme precipitation events generally show an upward trend, indicating that their intensity and frequency have increased. (5) Spatial distribution analysis shows that the precipitation in the study area is mainly concentrated in the northeast and south of Jingzhou and Tongdao, and the precipitation level in the west is lower. The comprehensive analysis shows that the annual precipitation in the study area is on the rise and has a certain periodic precipitation law. The spatial distribution is greatly affected by other factors and the distribution is uneven. Extreme precipitation events show an increasing trend, which may lead to increased flood risk in the region and downstream areas. In the future, it is necessary to strengthen countermeasures to reduce the impact of changes in precipitation patterns on local and downstream economic and social activities. Full article

The study evaluates the accuracy of two FvCB model sub-models (I and II) in estimating the maximum electron transport rate for CO₂ assimilation (J_A-max) by comparing estimated values with observed maximum electron transport rates (J_f-max) in four C₃ species: Triticum aestivum L., Silphium perfoliatum L., Lolium perenne L., and Trifolium pratense L. Significant discrepancies were found between J_A-max estimates from sub-model I and observed J_f-max values for T. aestivum, S. perfoliatum, and T. pratense (p < 0.05), with sub-model I overestimating J_A-max for T. aestivum. Sub-model II consistently produced higher J_A-max estimates than sub-model I. This study highlights limitations in the FvCB sub-models, particularly their tendency to overestimate J_A-max when accounting for electron consumption by photorespiration (J_O), nitrate reduction (J_Nit), and the Mehler reaction (J_MAP). An alternative empirical model provided more accurate J_f-max estimates, suggesting the need for improved approaches to model photosynthetic electron transport. These findings have important implications for crop yield prediction, ecological modeling, and climate change adaptation strategies, emphasizing the need for more accurate estimation methods in plant physiology research. Full article

Modular green wall, or living wall (LW) system, has evolved worldwide over the past decades as a popular green building feature and a nature-based solution. Differential climatic conditions across the globe make the standardisation of practices inapplicable to local scenarios. LW projects with differing goals and preferences like aesthetic (such as plant healthiness), water-saving, and minimal plant growth require optimal combinations of plant species to achieve single or multiple goals. This exploratory study aimed to deploy empirical field LW data to optimise analytical models to support plant species selection and LW design. Plant growth performance and water demand data of 29 commonly used plant species in outdoor modular LW systems without irrigation were collected in subtropical Hong Kong for 3 weeks. The 29 species tested were grouped into five plant forms: herbaceous perennials (16 spp), succulents (2 spp), ferns (2 spp), shrubs (7 spp), and trees (2 spp). Plant species-specific plant height, LAI, plant health rating, and water absorption amount were recorded every 6 days, together with photo records. Total water demand varied widely among plant species, ranging from 52.5 to 342.5 mL in 3 weeks (equivalent to 2.5 to 16.3 mL per day). The random forest algorithm proved that the water demand of the species was a dominant predictor of plant health tendency, among other parameters. Hierarchical clustering grouped plant species with similar water demand and health rating tendencies into four groups. The health rating threshold approach identified the top-performing species that displayed a healthy appearance as a basic prerequisite, coupled with one or two optional objectives: (1) water-saving and (2) slow-growing. The comparison among the plant selection scenarios based on projected LW performance (water demand, plant health, and growth) provided sound evidence for the optimisation of LW design for sustainability. LW projects with multiple objectives inherited a multitude of multi-scalar properties; thus, the simulation of LW performance in this study demonstrated a novel data-driven approach to optimise plant species selection and planting design with minimal resource input. Full article

The evaporative cooling system (ECS) is an energy-efficient and eco-friendly air-cooling technology that is very effective in dry climates, and the conventional method with cellulosic pads is widely used. However, because of the accumulation of dust and salts, these pads have a tendency to degrade quickly. This study aimed to examine the viability of using volcanic stone (Scoria) as an innovative material for evaporative cooling pads. The experiments were conducted in a wind tunnel (0.4 m × 0.6 m) with different pad thicknesses (d = 10 cm and 15 cm), water addition rates (m_w =1.6, 2.4, and 4 kg.min⁻¹.m⁻¹), and air speeds (v = 0.75, 1.25, and 1.75 m·s⁻¹). The results showed that the 10 cm thick pad consistently performed better than the 15 cm thick pad across all air speeds and water addition rates. The 10 cm thick pad achieved the highest cooling efficiency of 82% at a water addition rate of 2.4 L.min⁻¹.m⁻¹ and an air speed of 1.75 m·s⁻¹. In contrast, the cooling efficiency for the 15cm-t hick pad was 64% under the same conditions. The 10 cm thick pad consumed more water (1.8 to 2.8 kg·h⁻¹ compared to 1.0 and 2.4 kg·h⁻¹ for the 15 cm pad), as the ECS performance was directly associated with the amount of water used. Higher air speed led to a drop in pressure, which impacted fan performance. The pressure drops across the pads were between 10 and 13 Pa for an air speed of 1.75 m·s⁻¹. These results suggest that volcanic stone (Scoria) pads can provide effective cooling performance similar to that of commercial cellulosic pads but have added benefits of durability, less maintenance, and biological degradation resistance. The non-evaporative medium, especially the 10 cm thick Scoria pad, could be a more viable medium for evaporative cooling applications in arid areas. Full article

Variability identification and uncertainty characteristic analysis, under the impacts of climate change and human activities, is beneficial for accurately predicting the future evolution trend of hydrological variables. In this study, based on the evolution trend and characteristic analyses of historical precipitation and temperature sequences from monthly, annual, and interannual scales through the Linear Tendency Rate (LTR) index, as well as its variability point identification using the M–K trend test method, we further utilized three cloud characteristic parameters comprising the average Ex, entropy En, and hyper-entropy He of the Cloud Model (CM) method to quantitatively reveal the uncertainty features corresponding to the diverse cloud distribution of precipitation and temperature sample scatters. And then, through an application analysis of the proposed research framework in Anhui Province, China, the following can be summarized from the application results: (1) The annual precipitation of Anhui Province presented a remarkable decreasing trend from south to north and an annual increasing trend from 1960 to 2020, especially in the southern area, with the LTR index equaling 55.87 mm/10a, and the annual average temperature of the entire provincial area also presented an obvious increasing trend from 1960 to 2020, with LTR equaling about 0.226 °C/10a. (2) The uncertainty characteristic of the precipitation series was evidently intensified after the variability points in 2013 and 2014 in the southern and provincial areas, respectively, according to the derived values of entropy En and hyper-entropy He, which are basically to the contrary for the historical annual average temperature series in southern Anhui Province. (3) The obtained result was basically consistent with the practical statistics of historical hydrological and disaster data, indicating that the proposed research methodologies can be further applied in related variability diagnosis analyses of non-stationary hydrological variables. Full article

Methane is a potent greenhouse gas that significantly contributes to global warming, making the accurate quantification of methane emissions essential for climate change mitigation. The traditional matched filter (MF) algorithm, commonly used to derive methane enhancement from hyperspectral satellite data, is limited by its tendency to underestimate methane plumes, especially at higher concentrations. To address this limitation, we proposed a novel approach—the multi-level matched filter (MLMF)—which incorporates unit absorption spectra matching using a radiance look-up table (LUT) and applies piecewise regressions for concentrations above specific thresholds. This methodology offers a more precise distinction between background and plume pixels, reducing noise interference and mitigating the underestimation of high-concentration emissions. The effectiveness of the MLMF was validated through a series of tests, including simulated data tests and controlled release experiments using satellite observations. These validations demonstrated significant improvements in accuracy: In radiance residual tests, relative errors at high concentrations were reduced from up to −30% to within ±5%, and regression slopes improved from 0.89 to 1.00. In simulated data, the MLMF reduced root mean square error (RMSE) from 1563.63 ppm·m to 337.09 ppm·m, and R² values improved from 0.91 to 0.98 for Gaussian plumes. In controlled release experiments, the MLMF significantly enhanced emission rate estimation, improving $R^2$ from 0.71 to 0.96 and reducing RMSE from 92.32 kg/h to 16.10 kg/h. By improving the accuracy of methane detection and emission quantification, the MLMF presents a significant advancement in methane monitoring technologies. The MLMF’s superior accuracy in detecting high-concentration methane plumes enables better identification and quantification of major emission sources. Its compatibility with other techniques and its potential for integration into real-time operational monitoring systems further extend its applicability in supporting evidence-based climate policy development and mitigation strategies. Full article

As an international metropolis, Beijing still plays a key role in the development of national agricultural production technology despite its small regional scale. Climate change has a great impact on agricultural production. Previous studies often focus on a single short-term meteorological factor and lack a more systematic analysis of climate resources in Beijing. Based on the daily temperature, precipitation, relative humidity, wind speed and sunshine hours of 17 meteorological stations in Beijing in the past 42 years, this study analyzed the spatial and temporal distribution characteristics of agro-climatic resources and predicted the future climate change trend under different climate scenarios. The climate resource tendency rate is calculated on a time scale of every ten years (10a). The results showed that (1) the light resources are decreasing at a rate of 44.9~156.3 MJ m⁻² 10 a⁻¹, and the downward trends in light resources in the northeastern plain areas as well as in June and July are more significant; (2) the thermal resources are increasing at a rate of 34.2~176.4 °C·d 10 a⁻¹, and the upward trends in thermal resources in the southeastern plain areas and in March are more remarkable; (3) the water resources represented by the soil humidity index are changing at a rate of −1.6~6.1% 10 a⁻¹. The situation is complex, and the fluctuations of water resources in the central and western regions as well as in July, August and September are more significant; and (4) compared with the low-emission “dual carbon” scenario, the decrease in water volume and the overall increase in temperature in the high-carbon scenario are larger, and this trend is particularly obvious in the long run. This study provides a basis for Beijing’s agricultural layout and response to climate change, and its methods and results are also valuable for other regions to promote green, high-quality and sustainable agricultural development. Full article

One of the main differences between regional climate model and convection-permitting model simulations is not just how well topographic characteristics are represented, but also how deep convection is treated. The convection process frequently occurs within hours, thus a sub-daily scale becomes appropriate to evaluate these changes. To do this, a series of simulations has been carried out at different spatial resolutions (0.11$°$ and 0.025$°$) using the COSMO-CLM (CCLM) climate model forced by the ECMWF Reanalysis v5 (ERA5) between 2011 and 2020 over a domain covering northwestern Türkiye. Hourly precipitation and heavy precipitation simulated by both models were compared with the observations by Turkish State Meteorological Service (TSMS) stations and Integrated Multi-satellitE Retrievals for GPM (IMERG). Subsequently, we aimed to identify the reasons behind these differences by computing several atmospheric stability parameters and conducting event-scale analysis using atmospheric sounding data. CCLM12 displays notable discrepancies in the timing of the diurnal cycle, exhibiting a premature shift of several hours when compared to the TSMS. CCLM2.5 offers an accurate representation of the peak times, considering all hours and especially those occurring during the wet hours of the warm season. Despite this, there is a tendency for peak intensities to be overestimated. In both seasons, intensity and extreme precipitation are highly underestimated by CCLM12 compared to IMERG. In terms of statistical metrics, the CCLM2.5 model performs better than the CCLM12 model under extreme precipitation conditions. The comparison between CCLM12 and CCLM2.5 at 12:00 UTC reveals differences in atmospheric conditions, with CCLM12 being wetter and colder in the lower troposphere but warmer at higher altitudes, overestimating low-level clouds and producing lower TTI and KI values. These conditions can promote faster air saturation in CCLM12, resulting in lower LCL and CCL, which foster the development of low-level clouds and frequent low-intensity precipitation. In contrast, the simulation of higher TTI and KI values and a steeper lapse rate in CCLM2.5 enables air parcels to enhance instability, reach the LFC more rapidly, increase EL, and finally promote deeper convection, as evidenced by higher CAPE values and intense low-frequency precipitation. Full article

This study aims to explore the impacts of climate change on potato planting in Gansu Province so as to be able to adjust potato planting pattern scientifically and rationally. (1) Air precipitation and temperature time-series datasets were obtained from 87 meteorological stations in the study area over the past 50 years. The backpropagation neural network was employed to interpolate irregular and missing data in the time-series data. The altitude, the precipitation from June to July, the average temperature in July and the accumulated temperature above 10 °C were selected as the agricultural zoning indicators for the regionalization of potato planting. (2) The linear propensity rate method, cumulative anomaly method, ArcGIS technology and the Mann–Kendall mutation test were employed to examine the spatial–temporal variation in and mutation testing of the three zoning indicators. (3) The experimental results demonstrated that the amount of precipitation from June to July was registered at 139.94 mm, indicating a slight humidifying trend characterized by an annual increase rate of approximately 1.81 mm/10 a. Furthermore, a significant abrupt change was observed in 1998. The average temperature in July was registered at 20.53 °C, which showed an increasing trend at a rate of 0.55 °C/10 a, marked by a sudden shift in 1998. Lastly, the accumulated temperature above 10 °C was registered at 2917.05 °C, manifesting a significant warming trend at a rate of 161.96 °C/10 a, without any abrupt changes. For spatial distribution, the precipitation from June to July showed a decreasing spatial distribution pattern from south to north and from east to west, while its tendency rate showed a gradually decreasing trend from north to south and from east to west. The average temperature in July showed a decreasing spatial pattern from northeast to southwest, while its tendency rate showed a decreasing trend from west to east and from north to south. The accumulated temperature above 10 °C showed a spatial pattern of high accumulated temperatures in the northwestern and southeastern regions and low accumulated temperatures in the remaining regions, while its tendency rate showed a decreasing trend from west to east and from north to south. (4) The impacts of climate change on potato planting in Gansu Province were mainly manifested as a decrease of 0.30 × 10⁶ hm² in the cultivated land area in the most suitable region for potato planting post-1998, while the suitable area diminished by 0.96 × 10⁶ hm², the sub-suitable area expanded by 0.47 × 10⁶ hm², and the plantable area increased by 0.79 × 10⁶ hm². However, the unsuitable area experienced a reduction of 0.30 × 10⁴ hm². The findings of this study can provide a scientific foundation for optimizing and adjusting the potato planting structure, considering the backdrop of climate change. Moreover, they contribute to regional decision-making, thereby promoting sustainable agricultural development as well as enhancing both the yield and quality of potato in Gansu Province. Full article

The Arctic, characterised by severe climatic conditions and sparse vegetation, is experiencing rapid warming, with temperatures increasing by up to four times the global rate since 1979. Extensive impacts from these changes have far-reaching consequences for the global climate and energy balance. Satellite remote sensing is a valuable tool for monitoring Arctic vegetation dynamics, particularly in regions with limited ground observations. To investigate the ongoing impact of climate change on Arctic and sub-Arctic vegetation dynamics, a review of 162 studies published between 2000 and November 2024 was conducted. This review analyses the research objectives, spatial distribution of study areas, methods, and the temporal and spatial resolution of utilised satellite data. The key findings reveal circumpolar tendencies, including Arctic greening, lichen decline, shrub increase, and positive primary productivity trends. These changes impact the carbon balance in the tundra and affect specialised fauna and local communities. A large majority of studies conducted their analysis based on multispectral data, primarily using AVHRR, MODIS, and Landsat sensors. Although the warming of the Arctic is linked to greening trends, increased productivity, and shrub expansion, the diverse and localised ecological shifts are influenced by a multitude of complex factors. Furthermore, these changes can be challenging to observe due to difficult cloud cover and illumination conditions when acquiring optical satellite data. Additionally, the difficulty in validating these changes is compounded by the scarcity of in situ data. The fusion of satellite data with different spatial–temporal characteristics and sensor types, combined with methodological advancements, may help mitigate data gaps. This may be particularly crucial when assessing the Arctic’s potential role as a future carbon source or sink. Full article

This study assessed the past changes in extreme precipitation and temperature events across the Himalayan Mountains of Pakistan. This cold and mountainous environmental region has witnessed a significant increase in climate-related disasters over the past few decades. Spatiotemporal changes in extreme temperature and precipitation events were analyzed using 24 indices developed by the Expert Team on Climate Change Detection and Indices (ETCCDI). For this study, in situ data of 16 national meteorological stations were obtained from the Pakistan Meteorological Department (PMD) for the past three decades (1991–2020). The significance of the trends was assessed using the modified Mann–Kendall (MMK) test, and the Theil–Sen (TS) slope estimator was used to estimate the slope of the trend. The results showed that there has been a consistent decline in the total precipitation amount across the Himalayan Mountains of Pakistan. The trend exhibited a decrease in the annual average precipitation at a rate of −6.56 mm/year. Simultaneously, there was an increasing trend in the annual average minimum and maximum temperatures at rates of 0.02 °C/year and 0.07 °C/year, respectively. The frequencies of consecutive wet days (CWDs) and maximum 5-day precipitation (RX5day) have decreased significantly, with decreasing rates of −0.40 days/year and −1.18 mm/year, respectively. The amount of precipitation during very wet days (R95p) and extremely wet days was decreased by −19.20 and −13.60 mm/decade, respectively. The warm spell duration (WSDI) and the frequency of warm days (TX90p) across the Himalayan Range both increased by 1.5 and 1.4 days/decade. The number of cold days (TX10p) and cold nights (TN10p) decreased by 2.9 and 3.4 days/decade. The average temperature of the hottest nights (TXn) and the diurnal temperature range (DTR) were increased by 0.10 and 0.30 °C/decade. The results indicated an increasing tendency of dry and warm weather in the Himalayan region of Pakistan, which could have adverse consequences for water resources, agriculture, and disaster management in the country. Therefore, it is essential to prioritize the implementation of localized adaptation techniques in order to enhance sustainable climate resilience and effectively address the emerging climate challenges faced by these mountainous regions. Full article

The present study refers to an extensive investigation of lameness performed countrywide in Greece, on 325 sheep and 119 goat farms. The specific objectives of this work were to present data on the occurrence of lameness on sheep and goat farms and to identify variables (including variables related to climatic factors) associated with the disorder on the farms. Farms were visited and animals on the farm were assessed for the presence of lameness; further, an interview was carried out with the farmer to obtain information regarding practices applied on the farm. Climatic variables at the location of each farm were derived from NASA research. The within farm prevalence rate varied from 0.0% to 25.0% in sheep flocks and from 0.0% to 30.0% in goat herds. The mean ± standard error (median (interquartile range)) within farm prevalence rate among sheep farms was 1.9% ± 0.2 (0.0% (0.0%)); among goat farms, it was 2.6% ± 0.5% (0.0% (0.0%)). Multivariable analysis for within farm prevalence of lameness revealed three significant predictors in sheep farms: application of vaccination against foot-rot, increased precipitation at the farm location and longer annual grazing period for sheep, and one in goat farms: increased precipitation at the farm location. Treatment of lameness involved mostly administration of antibiotics (on 104 farms); the antibiotics administered most often were lincomycin (on 69 farms) and oxytetracycline (on 33 farms). There was a tendency for higher median within farm prevalence of lameness among farms where no antibiotic administration was practiced. Finally, 6.2% of sheep farmers and 4.2% of goat farmers considered lameness as an important health problem for the animals, specifically the third and fifth most important problem on the respective farms. Full article

Forest fire is one of the dominant disturbances in the forests of Heilongjiang Province, China, and is one of the most rapid response predictors that indicate the impact of climate change on forests. This study calculated the Canadian FWI (Fire Weather Index) and its components from meteorological record over past years, and a linear model was built from the monthly mean FWI and monthly fire numbers. The significance test showed that fire numbers and FWI had a very pronounced correlation, and monthly mean FWI was suitable for predicting the monthly fire numbers in this region. Then FWI and its components were calculated from the SRES (IPCC Special Report on Emission Scenarios) A2 and B2 climatic scenarios, and the linear model was rebuilt to be suitable for the climatic scenarios. The results indicated that fire numbers would increase by 2.98–129.97% and −2.86–103.30% in the A2 and B2 climatic scenarios during 2020–2090, respectively. The monthly variation tendency of the FWI components is similar in the A2 and B2 climatic scenarios. The increasing fire risk is uneven across months in these two climatic scenarios. The monthly analysis showed that the FFMC (Fine Fuel Moisture Code) would increase dramatically in summer, and the decreasing precipitation in summer would contribute greatly to this tendency. The FWI would increase rapidly from the spring fire season to the autumn fire season, and the FWI would have the most rapid increase in speed in the spring fire season. DMC (Duff Moisture Code) and DC (Drought Code) have relatively balanced rates of increasing from spring to autumn. The change in the FWI in this region is uneven in space as well. In early 21st century, the FWI of the north of Heilongjiang Province would increase more rapidly than the south, whereas the FWI of the middle and south of Heilongjiang Province would gradually catch up with the increasing speed of the north from the middle of 21st century. The changes in the FWI across seasons and space would influence the fire management policy in this region, and the increasing fire numbers and variations in the FWI scross season and space suggest that suitable development of the management of fire sources and forest fuel should be conducted. Full article

One of the primary factors in the hydrological cycle is reference evapotranspiration (ET₀). The prediction of ET₀ is crucial to manage irrigation water in agriculture under climate change; however, little research has been conducted on the trends of ET₀ changes in Shandong Province. In this study, to estimate ET₀ in the entire Shandong Province, 245 sites were chosen, and the monthly ET₀ values during 1901–2020 were computed using the Hargreaves–Samani formula. A deep learning model, termed SAO-CNN-BiGRU-Attention, was utilized to forecast the monthly ET₀ during 2021–2100, and the predictions were compared to two CMIP6 climate scenarios, SSP2-4.5 and SSP5-8.5. The hierarchical clustering results revealed that Shandong Province encompassed three homogeneous regions. The ET₀ values of Clusters H1 and H2, which were situated in inland regions and major agricultural areas, were the highest. The SAO-CNN-BiGRU-Attention and SSP5-8.5 forecasting results generally displayed a monotonically growing trend during the forecast period in the three regions; however, the SAO-CNN-BiGRU-Attention model displayed a declining tendency at a few points. According to the SAO-CNN-BiGRU-Attention and SSP5-8.5 results, during 2091–2100, H1, H2, and H3 will reach their peaks; the SSP2-4.5 results show that H1, H2, and H3 will peak in 2031–2040. At the end of the forecast period, for Clusters H1, H2, and H3, the prediction rate of SAO-CNN-BiGRU-Attention increased by 1.31, 1.56%, and 1.80%, respectively, whereas SSP2-4.5’s prediction rate increased by 0.31%, 0.95%, and 1.57%, respectively, and SSP5-8.5’s prediction rate increased by 10.88%, 10.76%, and 10.69%, respectively. The prediction results of SAO-CNN-BiGRU-Attention were similar to those of SSP2-4.5 (R² > 0.96). The SAO-CNN-BiGRU-Attention deep learning model can be used to forecast future ET₀. Full article