Accurate and early predictions in agriculture are essential for sustainable farming and optimizing field management. Crop yield prediction significantly impacts the farmer's decisions on crop insurance, storage demand and other important factors during the growing season. Due to non-linearity in crop yield, the use of non-linear models for forecasting purposes has become popular these days. In this paper, ANN and LSTM models were trained using weather parameters to forecast the cotton yield for Punjab, AIndia. Predicting the yield with minimum error is aAmain challenge. ANN (10 10 10 1) Amodel with ReLU activation function in hidden layers performed better thanAother forecasting models with a minimum MSE (0.0182). AThe analysis using the NN model concluded that the weather parameters played an important role in affecting the plant growth. AThese variables may enhance or reduce the yield significantly. Sensitivity analysis showed that relative humidity was the most important weather parameter followed rainfall.
An unprecedented heat wave lasting approximately two weeks occurred in Odisha during the third and fourth weeks of May 2015, Aresulting in meteorological hazards. In this study, an extreme temperature event (ETE) occurred on 25-27th May 2015, Awith about 15 meteorological observation stations in the state of Odisha recording maximum temperatures exceeding 45 degrees C, Aresulting in an intense heat wave. TheAmesoscale modeling framework (WRF4.0) is configured and optimized to simulate ETE at the regional scale in this study using different land-use scenarios. The maximum temperature from a time-ensemble simulation using the current land-use scenario based on Indian Space Research Organization (ISRO) data is found to be more accurate than simulations based on US Geological Survey (USGS) data at India Meteorological Department (IMD) meteorological stations. The mean percentage errors of simulated maximum temperatures over Odisha with respect to IMD station-scale observations are 1.6% (ISRO) and 3% (USGS) on 25th May 2015, and 4.2% (ISRO) and 4.7% (USGS) on 26th May 2015, respectively. Compared with the simulation based on ISRO data (more urbanized), the simulated horizontal surface wind at the different locations in Odisha is generally higher in the simulation based on USGS data. Changes in land use increase the roughness length, reducing surface wind speed. The dynamical aspects are also explored by analyzing humidity, outgoing longwave radiation (OLR), Convective Available Potential Energy (CAPE), and Convective Inhibition (CIN), etc., from the model and validated with the reanalysis products, which support the model performance in capturing the conducive environment resulting in a regional heat wave. The land use analysis reveals that the state-wide increase in urbanization between 1992 and 2015 was about 0.4%, with the highest percentage increase occurring in cities like Bhubaneswar, Sambalpur, Jharsuguda, and Rourkela in the Sundergarh district, which regularly experience the heat wave in the month of May, and the same are accurately simulated by the optimized and calibrated model configuration. Land-use change, including urban expansion and shifting cropping patterns, along with increased anthropogenic activities, is directly linked to the rise in maximum temperatures.
The National Centre for Medium Range and Weather Forecasting (NCMRWF) operated the MIHIR High-Performance Computing (HPC) Facility, delivering up to 2.8 petaflops of processing capacity to run Numerical Weather Prediction (NWP) models, thereby enabling accurate and timely weather forecasting. This study presents a comprehensive performance profiling and optimization analysis of the NCMRWF Unified Model (NCUM) version 13.0 on the MIHIR Cray XC40 HPC Facility. The model requires computations in the order of peta floating-point operations per second (PFLOPS). AThe NCUM is profiled at two horizontal resolutions n96e (similar to 130 km) taken as baseline and n1280e (similar to 10 km) using the Cray Performance Analysis Tool (CrayPAT) to identify runtime bottlenecks and communication overhead. The optimization experiments focused on halo size reduction, domain decomposition strategies, OpenMP threading, and MPI rank reordering. Results demonstrate that reducing the extended halo size from 10 to 5 grid points improves execution time by similar to 10 seconds and thus reduces communication costs and lowers imbalance in key routines from 15.2% to 3.8%. A change in domain decomposition from 4 & times;8 to 3 & times;12 further reduced MPI collective imbalance, while rank reordering improved execution time by up to 12.8%. These findings will be valuable in guiding domain and communication optimizations to efficiently scale high-resolution NWP models on the Cray XC40 system and future HPC architectures to be deployed at NCMRWF.
Climate variability, particularly fluctuations in temperature and rainfall, showed significant impacts on the agricultural sector, especially on cropping systems critical to ensuring food security. A comprehensive study assessed the influence of both positive and negative climate variability on agriculture and horticulture in the Sirmour district of Himachal Pradesh. This investigation analyzed the complex relationship between climate variables and crop productivity over 27 yearsA(1995-2021), Aproviding insights crucial for developing climate-resilient farming strategies. The statistical tools and models were employed to evaluate long-term trends in key climatic variables, with a focus on determining the significance and rate of observed changes. Over the study period, a significant decline in both maximum and minimum temperatures during the Kharif (summer) season was observed, with rates of-0.080 degrees C/year andA-0.090 degrees C/year, respectively. Trend analysis of Kharif crops indicated a significant increase in the productivity of rice (0.046 t/ha/year) and pulses (0.020 t/ha/year), Awhile maize showed aAnon-significant increase in productivity. In the horticulture sector, significant declines in both maximum and minimum temperatures were observed during the fruit-setting stage in the temperate zone, suggesting climate-induced alterations across multiple phenological stages. Furthermore, in the sub-tropical region, a significant decrease in maximum temperature was recorded during the fruit-setting stage. These findings indicate that over the past 27 years, climatic changes in the Sirmour district have had a generally positive impact on the productivity of most fruit crops, except guava, which exhibited a decline in yield. This analysis highlighted the critical need for adaptive strategies in the face of evolving climate patterns to sustain and enhance agricultural productivity in the region.
This study investigates the causes of riverbank erosion along the Hau Bassac River in Hoa Lac commune, An Giang province, Vietnamese Mekong Delta (VMD), based on integrating hydrodynamic, sediment, and morphology characteristics. Field data collected in November 2024, representing receding flood season conditions, indicate that high flow velocities, especially on the outer banks of meanders, frequently approached or exceeded calculated permissible non-eroding velocities. Steep bank slopes (> 50 degrees), erodible alluvial materials, and tidal fluctuations further contribute to instability. While regional factors like upstream damming and sand mining exacerbate systemic vulnerability, our findings pinpoint the critical role of local concentrated flow energy and geotechnical weaknesses. The analysis reveals a multi-stage failure mechanism, where extreme velocities during peak floods likely act as the primary trigger for basal scour, while sustained, near-threshold flows during the receding season prevent natural recovery and perpetuate bank instability. This study provides a critical snapshot of erosion mechanisms that are likely intensified during peak flood events, offering essential data for local mitigation and emphasizing the need for continued field-based monitoring in this dynamic river system.
Winter precipitation in the Western Himalayan Region (WHR) and the plains of Northwest India (NWI) is a notable weather phenomenon that occurs between December and early March. ATheAwinter months bring critical precipitation to the WHR and NWI, impacting agriculture and water resources. This paper investigates a significant wet spell during the first week of March, characterized by heavy rainfall, intense thunderstorms, and hailstorms across both regions. The wet spell during 1st-3rdMarch, 2024 was primarily influenced by a Western Disturbance (WD), Awhich led to the formation of induced cyclonic circulation and induced low-pressure systems. These atmospheric conditions created a conducive environment for the occurrence of heavy precipitation. AAdditionally, a steady supply of moisture from the Arabian Sea significantly contributed to the event. This influx of moisture enhanced rainfall intensity and facilitated the development of severe thunderstorms and hail. This investigation highlights the intricate relationship between Western Disturbances and moisture supply in generating winter precipitation events in the WHR and NWI. Understanding these interactions of weather systems mainly WD is vital for improving weather forecasting and managing the effects of extreme weather in the region. During this episode some of the stations like Karta, Jammu, Nahan and Manali received record breaking rainfall and placed in top 5 categories in climatological records. The heavy rainfall data revealed that on 3rdMarch, the subdivisions of Jammu & Kashmir and Uttarakhand recorded the highest rainfall, with some stations receiving 12 cm, followed by Haryana at 10cm. Station-wise data indicates that on 2ndMarch, the highest number of stations in the Himachal Pradesh subdivision reported heavy rainfall, followed by Jammu & Kashmir. On 3rd March, however, stations experienced heavy to very heavy rainfall in Jammu & Kashmir, followed by Himachal Pradesh, Haryana, and Uttarakhand. Future research should explore the long-term implications of these weather patterns in the context of climate change.
. Visibility at Chennai International Airport has shown a marked decline between 2016 and 2024, with a strong negative trend (R2 = 0.89) Alinked to increasing weather disturbances and rising urban pollution. This reduction inAvisibility has contributed to more frequent flight delays, particularly during early morning hours (around 06:30-07:00 IST), Awhen the lowest values are typically recorded in METAR. Seasonal analysis highlights November as the month with the poorest overall visibility, while the most severe reductions often below 1200 m and sometimes under 400 m occur in December through February. January records the highest frequency of extremely low visibility, including occurrences below 200 m. Fog typically begins near 03:30 IST and peaks around 07:00 IST, directly affecting early morning aviation operations. On average, fog occurs 5.2 days in January, 4.8 days in February, 3.9 days in December, and 3.1 days in March, with October-November adding another 3.2 days annually. Backward trajectory analysis suggests aerosol transport contributes to these conditions, reinforcing the role of both local pollution and regional atmospheric processes. The findings highlight an urgent need for stricter air-quality measures, improved airport infrastructure, and adaptive operational strategies to mitigate the growing challenges of low visibility.
This study aims to investigate the hydrometeorological dynamics of monsoon floods on the Kaveri River, Southern India. It encompasses the analysis of interannual rainfall variability and its correlation with floods, characterization of flood-inducing low-pressure systems (LPS), depth-area-duration (DAD) evaluation, assessment of the relationship between annual rainfall totals and flood frequencies, application of the normalized accumulated departure from mean (NADM) method to identify flood-linked anomalies, and investigation of the interplay between El Ni & ntilde;o southern oscillation (ENSO) phenomena and monsoon rainfall patterns governing flood dynamics. The study employs a 121-year rainfall dataset, Acoupled with a 45-year annual maximum series for two hydrological monitoring stations, enabling long-term statistical analysis of rainfall variability and flood dynamics. The results indicate that the interannual variability in the basin is characterized by a notable increase in the frequency and magnitude of floods, particularly post-1960s. Moreover, substantial flood events coincided with periods of above-average rainfall, underscoring the significant role of rainfall anomalies in modulating flood dynamics. Intense LPS predominantly drive major floods in the basin, especially in the deltaic zone. DAD curve of 23-25 July 1924 event has contributed the highest averageAdepth of rainfall over the basin for 1-day, 2-day and 3-day duration. Analysis of synoptic conditions associated with major floods indicate that most of the floods were associated with positive departure from mean rainfall in the basin. The NADM graphs show epochal behaviour of high and low rainfall and therefore, floods. The frequency of floods is generally high in years with normal or weak ENSO conditions.
Weather forecasting is an important and challenging attribute to predict because most atmospheric and agricultural fields depend on day-to-day weather fluctuations. Rainfall is one of the most important parameters dependent on various climatic conditions. We used the Random Forest (RF) and Long Short-Term Memory (LSTM) Neural Network models to predict rainfall for the major cities of Uttar Pradesh in the months of June, July, AAugust, and September (JJAS) Afrom 1901 to 2020. We used various statistical indices like the correlation coefficient (CC), ARoot Mean Square Error (RMSE), Aand Mean Absolute Error (MAE) to assess the quality of the forecast. Several climate indices were used as predictors to forecast rainfall as mentioned above. These indices include the North Atlantic Sea Surface Temperature, Nino 3.4, AEquatorial south-east Indian Sea, and rainfall at a lag value of 12. The prediction of rainfall utilizes a predictive neural network model, and the output is compared to real-time observed rainfall data for the forecasted period. The investigation revealed that the LSTM generally performed better compared to the RF. AThe generated output is promising and can be widely extended in this type of application.
The microphysical processes like nucleation, condensation, evaporation and coalescence are pivotal in precipitation formation. However, the parameterization of these processes areAa major source of uncertainty in numerical weather prediction models (NWP) due to their complex and highly variable nature. Warm clouds, characterized by temperatures above freezing, playAa significant role in precipitation generation through processes like collision-coalescence. In this study, we explore the coalescence dynamics of warm rain using a stochastic coalescence modelA(SCM) Aimplemented within the Python framework for the super droplet model (PySDM). We utilized RD-80 Impact Disdrometer data to validate the model results, which provide insights into raindrop size distributions and rainfall characteristics. By Disdrometer, we mean an instrument used to measure the size and velocity of precipitation particles, providing valuable data for understanding precipitation dynamics. Our experimental setup involved simulating the coalescence process of super droplets within a designated coalescence cell and comparing the resulting mass density data with ground observations. By conducting 200 experiments spanning a duration of 20,000 seconds, we captured the evolution of droplet populations. Our findings demonstrate the utility of PySDM in bridging the observational gap between ground-level measurements and atmospheric droplet dynamics, thereby enhancing our understanding of warm rain processes and improving precipitation forecasting capabilities.
This study investigates the spatio-temporal variability and long-term trends in rainfall across the major Makhana-growing wetland districts of Bihar, AIndia, using data spanning from 1901 to 2022. ToAdetect trends in seasonal and annual rainfall, statistical methods such as the Mann-Kendall (MK) test, trend-free pre-whitened MK (TFPW-MK) test, Sen's slope estimator, and simple linear regression (SLR) were employed. Innovative Trend Analysis (ITA) was applied as a complementary graphical method to enhance interpretation of trend behavior. Change-point detection was conducted using Pettitt's and the cumulative sum (CUSUM) tests at 1%, 5%, and 10% significance levels. The analysis revealed predominantly negative trends in annual and monsoon rainfall across most districts, except for Supaul, which exhibited a statistically significant increasing trend. Sen's slope estimates indicated spatial heterogeneity in trend magnitudes, with the most pronounced decline observed in Araria. Change-point analysis identified significant temporal shifts, particularly in monsoon rainfall, which strongly influenced annual totals. ITA effectively corroborated the results of the statistical tests, offering a nuanced visual understanding of trend patterns. The integrated use of statistical and graphical approaches provides a comprehensive assessment of rainfall dynamics, with implications for climate-resilient agricultural planning and water resource management in Bihar's wetland agro-ecosystems.
Understanding precipitation projections under climate change in semi-arid regions such as South Bihar is crucial for water-resources planning. This study examines the impacts of a high-emissions scenario (RCP 8.5) on drought patterns across the southern districts of Bihar using outputs from 14 downscaled, regionally adjusted global climate models (GCMs). AMeteorological drought is characterized with the Standardized Precipitation Index (SPI). AModel performance was evaluated by correlating GCM precipitation with India Meteorological Department (IMD) observations for 2006-2020; the best-performing models were HadGEM2-AO, MRI-CGCM3, MIROC5, CESM1-CAM5, and CCSM4. Projections indicate an similar to 30% increase in the severity of mild-to-moderate droughts, with Gaya, Jamui, and Lakhisarai most affected. The likelihood of rainless spells during 2021-2050 is projected to rise, signalling a growing risk of dry spell driven meteorological drought. While individual drought events cannot be predicted deterministically, the evolving rainfall regime suggests greater alternation between wet and dry conditions. Strong agreement between simulated and observed precipitation (R-2 approximate to 0.63-0.73) Asupports the suitability of these models for assessing future drought risk in this region.
The geothermal prospects of the Gede-Pangrango volcano complex in West Java, situated along the Pacific Ring of Fire, exhibit significant potential indicated by surface manifestations such as solfataras, fumaroles, and hot springs. Previous studies have reported the presence of a high-temperature reservoir (290-300 degrees C) within Quaternary volcanic rocks at depths of 2000-2800 meters; however, uncertainties remain regarding the subsurface structures and faults that control the geothermal system. This study employs satellite-derived gravity data (GGMPlus) and elevation data (ERTM) to investigate these features. Edge-detection filters, including the First Horizontal Derivative (FHD) and Second Vertical Derivative (SVD), Awere applied to delineate subsurface structures associated with geothermal activity. The residual anomaly map reveals three primary zones: high, moderate, and low anomalies. High anomalies are concentrated around Mount Gede and Mount Pangrango, reflecting dense volcanic rocks or shallow intrusions, whereas low anomalies correspond to pyroclastic deposits and hydrothermally altered rocks. FHD and SVD analyses highlight northeast-southwest (NE-SW) trending faults that serve as primary pathways for geothermal fluid migration, confirmed by the alignment of fractures with hot springs and the Kawah Ratu crater. The 2D cross-sectional model indicates a low-density reservoir at depths of 1-2 km controlled by a major fault, underlain by intrusive rocks acting as the heat source. Furthermore, the 3D density modeling demonstrates a connected geothermal system between Mount Gede and Mount Pangrango, characterized by high-permeability zones beneath both volcanoes.
We applied inverse probability weighting (IPW) causal modeling with three air pollutants (PM2.5, O3, Aand NO2), Ameteorological parameters, and potential lag effect of exposures to estimate the marginal effect of shortterm exposure to air pollution (AP) on cause-specific mortality risks in the Indo-Gangetic Plain (IGP). A causal linkage between short-term exposure to PM2.5 and O3 was found and all the four assessed causes of deaths (viz. neurological, respiratory, cardiovascular, and nephrological deaths), whereas the NO2 was found to have a linkage only with the cardiovascular mortality risks-for every 10-ppb increase in NO2 exposure, the mean risk increased by 0.84% (95% Confidence Interval (CI) = 0.58, 1.10). For every 10 mu g m-3 increase in short-term PM2.5 exposure, the increase in the risk of in-hospital all-cause deaths, neurological deaths, respiratory deaths, cardiovascular deaths, and nephrological deaths was 1.76% (1.57, 1.95), 0.28% (0.07, 0.50), 0.73% (0.47, 0.98), 0.72% (0.48, 0.96), and 0.11% (0.09, 0.14), respectively. Likewise, after controlling for PM2.5 and NO2, Athe causational linkage between acute exposure to O3 and respiratory mortality risk was found to be highest among all studied scenarios-for every 10-ppb increase in O3 there was a 2.24% increase in the respiratory death risk (2.02, 2.45). AThis set of results from IPW causal modeling could serve as the causational evidence documenting relative risk (RR) estimates of premature non-accidental cause-specific mortalities attributable to ambient AP in a subset of Indian population.
This study examines the interannual variability of pre-monsoon convective activity over Telangana, India, from 1951 to 2024, focusing on its influence on rainfall distribution, rainy days, and associated atmospheric processes. Normalized analysis of rainy days during the March-May (MAM) season reveals above-average pre-monsoon rainfall activity in Strong Convective Years (SCY) (e.g., 1951, 1990, 2006, 2015, 2023) and below-average rainfall activity in Weak Convective Years (WCY) (e.g., 1964, 1965, 1966, 1973, 1984). A marginally increasing trend in pre-monsoon rainfall due to convective activity is observed in recent years. To understand the role of atmospheric processes, composite anomaly analysis is performed for key meteorological variables, including rainfall, rainy days, Convective Available Potential Energy (CAPE), vertically integrated moisture divergence (VIMD), Atotal cloud cover (TCC), Atotal column cloud liquid water (TCLW), Asolar radiation, and soil water content. ASpatial analyses highlight regional variability, showing enhanced pre-monsoon rainfall activity during SCY and suppressed rainfall activity during WCY. Wind patterns at 850 hPa reveal a significant discontinuity over peninsular India, with intensified circulations during SCY and an eastward shift in winds during WCY, emphasizing the role of atmospheric dynamics in modulating convective activity. SCY years exhibit increased atmospheric instability, moisture convergence, cloudiness, and soil water content, along with reduced solar radiation, whereas WCY years show opposite trends. These findings underscore the complex interplay of atmospheric processes driving convective activity and offer critical insights for regional water resource management and climate adaptation. By linking convective activity to broader climatic factors, this study enhances the understanding of pre-monsoon dynamics in Telangana and provides valuable input for forecasting and mitigating extreme weather impacts.
. Monitoring environmental factors such as soil moisture and precipitation on a frequent basis plays a vital role in identifying early signs of flooding, particularly in regions prone to extreme weather. Passive microwave remote sensing stands out in this context due to its ability to collect data regardless of weather conditions, along with its regular daily coverage. In this study, changes in the Polarisation Index (PI) were examined before and after the severe flooding that struck Derna, Libya, on 11 September 2023. AThe data used were obtained from the AMSR2 sensor operating at X-band (10 GHz) aboard Japan's GCOM-W1 satellite. Notably, a sharp rise in PI was recorded one day ahead of the flood, aligning with increased soil moisture linked to intense rainfall in the area. Following the event, PI values remained elevated, indicating continued ground saturation. These findings point to the potential of PI as an early warning indicator for heavy rainfall and flood risk. With appropriate selection of observation points, this method could support the development of flood forecasting systems in other high-risk regions around the world.
Accurate crop yield forecasting is essential for sustainable agricultural management and food security. This study leverages meteorological parameters and machine learning techniques to develop a robust yield prediction model for sugarcane in the major sugarcane growing district of Gujarat. AThis study evaluates the performance of Stepwise Multiple Linear Regression (SMLR) and three machine learning (ML) models: "Artificial Neural Networks (ANN)," A"Random Forest Regression (RFR)," and "Support Vector Regression (SVR)" Afor predicting sugarcane yield in four districts of Gujarat, India (Navsari, Bharuch, ASurat, and Tapi). AHistorical yield data (2001-2019) Aand weather variables were used to train and test the models, with validation performed on a holdout dataset (2020-2022). AResults indicate that ANN outperformed other models inAmost districts, achieving the lowest errorsAand highest predictive accuracy. Specifically, in Bharuch, ANN achieved an RMSE of 2491.28 t/ha and MAPE of 3.57%; in Surat, the RMSE was 8139.02 t/ha and MAPE 9.92%; in Tapi, the RMSE was 3630.44 t/ha and MAPE 4.43%. In Navsari, the model also performed well with an RMSE of 5388.97 t/ha and MAPE of 8.55%. SMLR demonstrated strong performance in Navsari but required further optimization in other regions. RFR and SVR showed mixed results, with significant errors in Surat and Tapi, highlighting challenges in capturing regional variability. Feature importance analysis revealed that weather variables, such as relative humidity and rainfall, were critical predictors across all districts. The study underscores the importance of integrating remote sensing data with meteorological variables to enhance model accuracy, particularly for SMLR. ANN is recommended for yield forecasting in Bharuch, Surat, and Tapi, while SMLR is suitable for Navsari. These findings provide valuable insights for improving sugarcane yield prediction models, supporting sustainable agricultural practices, and aiding policymakers in resource allocation and risk management.
Approximately 60% of India's population is exposed to extreme temperatures that exceed critical health risk thresholds for 10 to 20 days each year. However, the distribution of heat and the associated vulnerabilities vary significantly across the country’s physiographical regions, making it challenging to identify high-risk areas. Recognizing these hotspots and establishing a comprehensive framework to integrate risk management systems into decision-making processes is crucial. In particular, there is a need to focus on threshold detection methods to highlight the importance of further research in regions susceptible to heatwaves. This study utilized MODIS Land Surface Temperature (LST) data alongside daily mean gridded surface air temperature data from the India Meteorological Department (IMD) to analyze 33 heatwave events that occurred between 2009 and 2020 in western Madhya Pradesh, central India. The findings demonstrate that satellite-derived LST data can effectively identify regional heatwave patterns, with a strong correlation (correlation coefficient approximately 0.7) observed between heatwaves detected via air temperature anomalies and those estimated using LST departures. Furthermore, the study established optimal LST thresholds for heatwave detection: the 70th percentile (52.5 °C) for standard heatwaves and the 90th percentile (53.5 °C) for severe heatwaves. Additionally, the results indicate a noticeable trend of summer warming in northern and central India, leading to increased heatwaves' frequency, intensity, and duration over the past decade.
This study presents a data-driven analysis of temperature and rainfall patterns across 12 districts in Southern Telangana using 43 years of gridded meteorological data A(1981-2023). A Employing descriptive statistics, correlation matrices, and six predictive models, Random Forest (RF), A Artificial Neural Network (ANN), A Support Vector Regression (SVR), A Long Short-Term Memory (LSTM), ARIMA, and TBATS, we evaluated forecasting accuracy for maximum temperature, minimum temperature, and rainfall. The RF model demonstrated superior performance with the lowest Test RMSE (0.1178) and Test MAE (0.0601) A across all parameters, outperforming traditional time series models. Correlation analysis revealed strong inter-location temperature synchrony (r approximate to 0.98-1.00), Awhile rainfall exhibited high spatial variability (r = 0.15-0.77), A indicating localized climatic influences. Feature importance analysis identified L332 and L333 as dominant predictors, with scores of 0.1209 and 0.0557, A respectively. Novel contributions include: A (1) a comparative evaluation of six models on long-term regional climate data, (2) integration of feature importance to enhance interpretability, and (3) prediction interval analysis confirming model stability with consistent upper bounds (similar to 0.187) and zero lower bounds. These findings offer actionable insights for climate adaptation, agricultural planning, and resource management in semi-arid regions.
Uttarakhand state receives around 79% of the total annual rainfall during the southwest monsoon season, i.e. from June to September. In this study, precipitation data of the monsoon season from 1983 to 2023 of four departmental observatories of the India Meteorological Department (IMD) in Uttarakhand have been used to analyse the trend of frequency of Heavy, Very Heavy and Extremely Heavy rainfall. Tehri & Mukteshwar observatories represent the hilly region of Uttarakhand, Awhile Dehradun & Pantnagar observatories represent the plains of Uttarakhand. AHeavy rainfall climatology reveals that the plains of Uttarakhand receive a higher frequency of heavy rainfall days than the hilly stations. TheAstate experiences higher frequency and lower variability of heavy rainfall days during August & July months, followed by September & June. Mann-Kendall non-parametric trend test has been used to evaluate the existence of monotonic trends. The results show a weak, statistically insignificant increasing tendency in the number of heavy and more rainfall days in Dehradun station, while no trend is observed in other stations. Quantitatively, aAnon-significant rise in the percentage of extreme rainfall to the total monsoon rainfall is observed over Dehradun, Pantnagar & Mukteshwar stations, while Tehri exhibits a non-significant fall in theAextreme rainfall quantity. AAsAper the IMD criteria, the 24-hour accumulated rainfall is categorized into Very Light, Light, Moderate, Heavy, Very Heavy & Extremely Heavy. The trend analysis of individual categories of rainfall shows a significant increasing trend in very light rainfall days over three stations (Dehradun, Mukteshwar & Pantnagar) and aAnon-significant decreasing trend of dry days over three stations (Dehradun, Mukteshwar & Tehri). AThe other rainfall categories showAno trend.
