The development of urban areas in Manokwari Regency has the potential to reduce the availability of land containing vegetation or Green Open Space (GOS) which may be a contributing factor to increasing land surface temperature that triggers the Urban Heat Island (UHI) phenomenon. This study aims to estimate land surface temperature, analyze its relationship with land cover, and examine the pattern of temperature distribution in the four districts of the study area. The research method integrates remote sensing and Geographic Information Systems, using Landsat 8 OLI/TIRS imagery data, topographic maps, and land cover maps, which were analyzed in GIS through an overlay method and supported by field surveys. The results show that land surface temperatures range from 18.29–30.79°C with an average of 23.23°C and are dominated by the 25–30°C temperature class. Areas with extensive vegetation cover have lower temperatures, whereas residential and built-up areas have higher temperatures. Field surveys at 49 points indicate an average air temperature of 36.25°C with a low correlation to Landsat-derived temperatures (r = 0.34). The areal distribution of high temperatures is, in sequence: West Manokwari District, 2,201.560 ha (28.517%); South Manokwari District, 1,514.092 ha (5.891%); East Manokwari District, 413.156 ha (15.866%); and North Manokwari District, 184.431 ha (0.982%). West Manokwari District is the top priority for mitigating increasing land surface temperature, followed by East Manokwari District and South Manokwari District, through the development of GOS in built-up areas in Manokwari Regency.
Sago starch extraction is aimed to separate starch as much as possible from waste or fibre. The efficiency and effectiveness of extraction process depend on several factors i.e. the characteristics of extraction machine and fineness degree of disintegrated sago pith is being processed. In previously study, it had been resulted stirrer rotary blade type of sago starch extraction machine powered by gasoline engine. The purpose of this study was to develop and test the performances of stirrer rotary blade type of sago starch extractor. The developed machine was tested it’s performances under 4 levels of extraction duration time i.e. 10 minutes, 15 minutes, 20 minutes and 25 minutes. The Parameter have been measured to evaluate the machine’s performances were (1) extraction capacity, (2) starch rendement, (3) starch yield and (4) starch losses in sago pith waste. Based on performance test results have shown that the developed machine has higher performance than previously prototype. The results showed that extraction duration time significantly affect extraction capacity, starch rendement, and starch losses in sago pith waste, however the starch yield was not affected significantly. The highest performance was resulted at extraction duration time 15 minutes. The performances of the machine at the condition were (1) extraction capacity 247 kg repos/hour, (2) starch rendemen 37,35%, (3) starch yield 78,81 kg/hour, and (4) starch losses in sago pith waste 2,00%.
Subjective financial satisfaction does not always reflect objective economic conditions, as seen in several Asian countries that have strong economies but tend to report low levels of public welfare. This study differs from previous global studies in that it focuses on Asia, a region with high cultural and religious diversity. This phenomenon has prompted studies on the role of non-economic factors, such as religiosity, in shaping financial perceptions. This study aims to analyse the relationship between religiosity and sociodemographic variables on financial satisfaction in Asian countries using multilevel modeling. A multilevel approach was used because the data has hierarchical structure data (individuals nested within countries). Using a multilevel approach, this study is able to capture the influence of religiosity at both individual and country levels. This study uses secondary data from the seventh wave of the World Values Survey (WVS) (2017–2022), which includes 30,878 respondents from 20 Asian countries. The multilevel models constructed are the null model, the random intercept model, and the random slope model. Religiosity is measured through two indicators: the importance of God in life and belief in God. The results of this study indicate that the random intercept model is the best model. The importance of God, age, being female, higher education, and income positively influence financial satisfaction, while belief in God negatively influences financial satisfaction at both the individual and national levels. These findings confirm that financial satisfaction in Asian countries is not only influenced by economic factors but also by religious values and individual psychological well-being.
The world is experiencing rapid changes, making it more complex and difficult to predict. This situation is called the Volatility, Uncertainty, Complexity, and Ambiguity (VUCA) Era. Therefore, the demand for rapid adaptation through innovation is greater. However, data from the World Intellectual Property Organization shows that Indonesia's innovation index in 2022 still ranked 75th of 132 countries. It indicates that Indonesia still faces challenges and weaknesses in managing the innovation cycle, particularly in government innovation. Consequently, it is crucial to conduct research on local government innovation by identifying the determinants of the regional innovation index (RII) in 2022. In this research, robust regression with MM-estimation was used to address outlier issues in the data. The result shows that the digital skills score, degree of fiscal decentralization, and innovation capability scores have a significant positive effect on the RII in 2022. Meanwhile, the high school education completion rate and the ratio of students and lecturers have a significant negative effect on the RII in 2022.
Ascorbic acid, a water-soluble vitamin, is prone to degradation during the storage of fruits and vegetables, necessitating fast and reliable detection techniques for quality control. This study presents a digital instrument employing the eddy current principle for the sensor, designed with a slit-structured copper (Cu) electrode coated with a TiO₂ nanocomposite, to quantify ascorbic acid levels in orange, pineapple, and bilimbi juice. The system incorporates an Arduino Uno microcontroller and an inductive proximity sensor to monitor voltage shifts induced by interactions between ascorbic acid molecules and the electrode surface. Calibration was carried out using standard solutions ranging from 1 to 9 ppm, with UV-Vis spectrophotometry serving as the validation method. The sensor exhibited a strong linear correlation (R² = 0.991), with output voltage decreasing from 6.12 V at 1 ppm to 4.86 V at 9 ppm. Measured concentrations were 8.93 ppm (orange), 7.07 ppm (pineapple), and 4.59 ppm (bilimbi), closely aligning with UV-Vis results. The mean relative error was 3.25%, indicating an accuracy of 96.74%. Statistical analysis using linear regression and the t-test confirmed no significant difference from the reference method. These results highlight the sensor's potential as a rapid and precise tool for ascorbic acid detection in the agri-food industry.
Geosesarma noduliferum is a semi-terrestrial freshwater crab endemic to Java, exhibits behavioral activity potentially regulated by daily light-dark cycles. This study was aimed to analyze the circadian rhythm of this crab under laboratory conditions and examine habitat preferences during diurnal and nocturnal periods. Four individuals of 2 males and 2 female crabs were collected from their natural habitat and observed in simulated habitat tanks over three days. Behavioral observations were conducted using focal sampling during six daily sessions (three in diurnal and three nocturnal), and all activities were categorized in active or passive behaviour. The results showed that active behaviors—such as exploration, feeding, and social interaction—were more frequent and prolonged during the nocturnal period, peaking at 11:00 PM. However, passive behaviors such as resting and burial were more prevalent during diurnal. Kruskal-Wallis and Dunn tests indicated significant behavioral differences between day and night periods (p<0.05). Habitat preference also varied; both sexes favoring terrestrial areas at night; during diurnal, males predominantly occupied aquatic areas while females showed a balanced distribution. This study suggests that G. noduliferum exhibits a predominantly nocturnal activity pattern.
Microbes form infectious biofilms to avoid antibiotic mechanisms, leading to the rise of antibiotic resistance. This phenomenon demands massive development of novel antibiofilm agents. Coffee plants have been investigated as antibiofilm agents despite their value as a popular export commodity. Gayo Arabica coffee, a signature product of Aceh, represents a sustainable and locally abundant natural resource. The aim of this study was to determine the antibiofilm activity of Gayo Arabica coffee bean extract against two of the most well-known biofilm producers, namely Pseudomonas aeruginosa and Candida albicans. The minimum inhibitory concentrations of the 96% ethanol-macerated coffee beans were identified against both microbes. The antibiofilm activity was evaluated based on inhibition and degradation assays. Chemical profiling of the extract was performed using Gas Chromatography-Mass Spectrometry, followed by molecular docking to assess interactions between compounds and biofilm-associated proteins in P. aeruginosa and C. albicans. The extract demonstrated antibiofilm activities against both microbes, which may be associated with the dominant compounds, particularly caffeine, hexadecanoic acid, and cis-13,16-docosadienoic acid. Caffeine had the strongest binding affinity to target proteins based on the docking analysis. This study exhibits the potent antibiofilm activity of Gayo Arabica coffee bean extract against P. aeruginosa and C. albicans. Keywords: biofilm, resistance, Pseudomonas aeruginosa, Candida albicans, Gayo coffee
ABSTRACT: Uji aktivitas antioksidan dengan metode DPPH ekstrak heksana, etil asetat (EtOAc), dan methanol (MeOH) biji kelor telah dilakukan. Metode ekstraksi yang digunakan adalah ekstraksi sokletasi. Metode ekstraksi sokletasi merupakan jenis ekstraksi dengan pelarut cair organik yang dilakukan secara kontinyu pada suhu dan jumlah pelarut tertentu. Rendemen dari masing-masing ekstrak adalah 26.76%, 9.86% dan 39.02%. Tingkat aktivitas antioksidan ekstrak biji kelor diukur dengan nilai IC50. Sebagai kontrol positif digunakan Vitamin C dimana mempunyai nilai IC50 42,78 ppm. Aktivitas antioksidan ekstrak heksana, EtOAc, dan MeOH masing-masing adalah sebesar 406.09, 399.03 dan 135.83 ppm. Tingkat aktivitas antioksidan tertinggi adalah ekstrak MeOH, dimana termasuk dalam kategori tingkat aktivitas sedang. ABSTRACT: Uji aktivitas antioksidan dengan metode DPPH ekstrak heksana, etil asetat (EtOAc), dan methanol (MeOH) biji kelor telah dilakukan. Metode ekstraksi yang digunakan adalah ekstraksi sokletasi. Metode ekstraksi sokletasi merupakan jenis ekstraksi dengan pelarut cair organik yang dilakukan secara kontinyu pada suhu dan jumlah pelarut tertentu. Rendemen dari masing-masing ekstrak adalah 26.76%, 9.86% dan 39.02%. Tingkat aktivitas antioksidan ekstrak biji kelor diukur dengan nilai IC50. Sebagai kontrol positif digunakan Vitamin C dimana mempunyai nilai IC50 42,78 ppm. Aktivitas antioksidan ekstrak heksana, EtOAc, dan MeOH masing-masing adalah sebesar 406.09, 399.03 dan 135.83 ppm. Tingkat aktivitas antioksidan tertinggi adalah ekstrak MeOH, dimana termasuk dalam kategori tingkat aktivitas sedang. Kata kunci: Antioksidan, Biji Kelor, DPPH, Ekstrak Heksana, Ekstrak EtOAc, Ekstrak MeOH
Kemasan galon berbahan polikarbonat (PC) diketahui mengandung senyawa Bisfenol-A (BPA) yang berpotensi bermigrasi ke dalam air dan menimbulkan risiko kesehatan, terutama apabila disimpan pada suhu tinggi. Penelitian ini bertujuan untuk menganalisis kadar migrasi BPA dari kemasan galon PC No. 7 ke dalam larutan simulan etanol 20% yang beredar di Kabupaten Manokwari. Sampel diinkubasi pada suhu 60°C selama dua durasi berbeda, yaitu 2 jam dan 10 hari. Analisis dilakukan menggunakan metode High-Performance Liquid Chromatography (HPLC) dengan detektor fluoresens. Hasil penelitian menunjukkan bahwa waktu inkubasi berpengaruh terhadap tingkat migrasi BPA. Migrasi BPA tidak terdeteksi setelah inkubasi selama 2 jam, namun terdeteksi setelah 10 hari inkubasi dengan konsentrasi BPA berkisar antara 0,007–0,017 ppm. Selain itu, uji terhadap 18 sampel galon PC No. 7 yang beredar di Manokwari menunjukkan bahwa seluruh sampel mengandung BPA dengan kadar antara 0,005–0,185 ppm. Seluruh kadar BPA yang terdeteksi masih berada di bawah ambang batas maksimum yang ditetapkan dalam Peraturan BPOM Nomor 20 Tahun 2019, yaitu 0,6 ppm.
One of the secondary infections observed among COVID-19 patients is a Urinary Tract Infection (UTI). The presence of bacteria and fungi causes UTI, which can certainly occur after a urine culture. Urinalysis is one of the critical examinations to diagnose UTIs and assess functional disorders in the urinary tract. This study examines the characteristics of microbes isolated from urine specimens and urinalysis results in COVID-19 patients at Dr. Zainoel Abidin Hospital, Banda Aceh. This observational analytic study used secondary data from urine culture and urinalysis results of COVID-19 patients by implementing the total sampling technique. This study involved 110 urine culture data from confirmed COVID-19 patients. Comparative analysis of parameters between positive and negative urine culture groups with urinalysis results using categorical variables through the Chi-square test and Fisher's Exact test. The findings suggest that Escherichia coli, Enterococcus spp., and Candida spp. are the predominant uropathogens in COVID-19 patients, with urinalysis frequently indicating leukocytosis as a marker of urinary tract involvement. Urine culture is the gold standard for quantitatively diagnosing UTIs by determining bacterial density and identifying specific pathogens. Urinalysis, which checks for leukocytes in the urine, can support these results.
PT Bumi Bara Makmur Mandiri is a mining company engaged in coal mining, located in Hajran Village, Bathin XXIV District, Batang Hari Regency, Jambi Province. The company employs an open-pit mining method, which results in the formation of depressions that act as rainwater catchment areas, creating pools (sumps). To manage these sump areas effectively for mining activities, a mine dewatering drainage system is implemented. Over the past ten years, rainfall data indicates a catchment area of 246.02 hectares, with an estimated rainfall of 280.88 mm for a return period of ten years. The groundwater that accumulates in the pooling area has an inflow rate of 0.11 m³/s, leading to a total water discharge entering PIT A of 39.12 m³/second. Currently, there is one pump available with a power capacity of 10.83 kW to remove this water. Due to the significant water discharge, it is essential to design an economical open flow that can handle a flow rate of 2.96 m³/s. Furthermore, a design for a holding pond is required, as the existing capacity is insufficient to accommodate the incoming water discharge, which hampers the sedimentation process. To address this, a redesigned settling pond with dimensions of 46 m x 36 m x 7 m and six compartments has been proposed. This solution will enhance the effectiveness of the mine drainage process
Upland rice is a type of rice that can be grown in Ultisol. The obstacles to cultivating plants in Ultisol are low pH and high aluminum content which hinders seed germination. Seed priming is a technology that can overcome problems with Ultisol. The aim of this research was to determine the effect of priming on the germination of upland rice seeds in acid soil conditions. The experiment conducted on non-factorial Completely Randomized Design (CRD) consisting of 8 treatments with 3 replications. The data were analyzed for variance and followed by Honest Significant Differences (HSD) at α 5% using the Statistic R Program. Seed priming treatments tested were (1) Untreatment; (2) Hydropriming; (3) priming GA3 25 ppm; (4) priming GA3 50 ppm; (5) priming PEG 6000 10%; (6) priming PEG 6000 20%; (7) 0.5% KNO3 priming, and (8) 1% KNO3 priming. Rice seed are soaked in priming for 24 hours. The upland rice seeds of the Inpago 13 Fortiz variety were planted in Ultisol soil media with a pH of 4.45, Al content of 0.44%, and Fe 1.37%. The results showed that the priming treatment increased the germination and themost effective treatment was priming GA3 50 ppm, each value of showed germination (92.38%), germination speed (19.71% day-1), vigor index (83.81%), and time of appearance of plumule (2.96 day).
This study reports the synthesis and characterization of a monolithic activated carbon adsorbent modified with alumina and chitosan (Al-Chit/OAC), derived from oil palm empty fruit bunches (OPEFB). The adsorbent was fabricated through pyrolysis, followed by alumina incorporation and chitosan impregnation. FTIR analysis confirmed the presence of functional groups including O–H stretching (3640 cm⁻¹), C–H stretching (2920 cm⁻¹), C–N/C–O stretching (1055–1031 cm⁻¹), and Al–O vibrations (693, 522, 495 cm⁻¹), indicating successful surface modification. TGA revealed two major stages of thermal degradation, with a total mass loss of 17.4% and a final residue of 17.55%, reflecting the presence of thermally stable inorganic components. SEM imaging showed a heterogeneous and porous surface with agglomerated particles and interparticle voids, suggesting enhanced surface accessibility. Even though we didn't test how well it absorbs substances, the physical and chemical properties of the composite show it could be very useful for cleaning up acid mine drainage (AMD) in the future. Further studies are recommended to validate its adsorption performance.
Accurate poverty mapping at the district and municipal levels remains challenging due to small sample sizes in household surveys, which often result in unstable direct estimates. To address this issue, this study employs microdata from the 2023 National Socioeconomic Survey (SUSENAS) to estimate household-level poverty proportions across 27 districts and municipalities in West Java Province using a binomial Generalized Linear Mixed Model (GLMM) combined with the Empirical Best Predictor (EBP) and Simultaneous Confidence Intervals (SCI). The GLMM framework captures household characteristics and random area effects to account for spatial heterogeneity. Three SCI approaches—Bonferroni correction, Bootstrap-t, and the Simes procedure—were implemented to evaluate EBP uncertainty while controlling the family-wise error rate. Results reveal substantial disparities, with Tasikmalaya (21.7%), Bandung Barat (15.5%), and Cianjur (12.8%) consistently above the provincial average of (6.8%), while urban areas such as Cimahi, Bekasi, and Depok report poverty rates below 2%. All methods achieved full empirical coverage (ECP = 100%), although interval widths differed: Bonferroni produced the widest intervals (AIW = 44.99), Bootstrap-t yielded the narrowest and most efficient (AIW = 29.16), and Simes provided intermediate but highly consistent results (AIW = 33.24). These findings underscore the methodological importance of integrating GLMM, EBP, and SCI for small area estimation while offering practical insights for evidence-based policy development and poverty reduction strategies in Indonesia.
Many scientific fields, including the geosciences, have successfully employed machine learning to address numerous significant issues. Current studies show that the application of machine learning within the geosciences is still in its early stages, and there is a huge potential for this technique that need to be explored. This research focuses on the Late Permian Beekeeper Formation from the Perth Basin, Australia. It aims to improve our understanding of the application of machine learning to characterise subsurface rock formations. The objectives of this study are threefold: (1) to conduct cutting, crossplot, and modern machine learning analyses on a mixed carbonate-siliciclastic reservoir; (2) to compare the results from the aforementioned analyses and to interpret the electrofacies and lithofacies; and (3) to understand the degree of accuracy of the application of machine learning in the characterisation of the subsurface rock formations. Cutting, crossplotting, and modern machine learning analyses have been conducted to achieve the aim and objectives of this study. Seven electrofacies, associated with nine lithofacies, were identified within the studied data, and these were classified into carbonate-dominated facies group, siliciclastic-dominated facies group, and mixed carbonate-siliciclastic facies group. Results also show the presence of stratal and compositional mixing within the Beekeeper Formation. A combination of cutting, crossplot, and machine learning analyses can provide a better, more accurate, and more reliable interpretation of the facies of the Beekeeper Formation. This study is expected to advance our understanding of the application of machine learning in geosciences.
Wickerhamomyces anomalus and Pichia kudriavzevii have high potential to produce bioethanol under high stress condition, due to their stress-tolerant properties. To elucidate and develop an efficient and sustainable bioethanol production, characterization of ethanol fermentation reactions is highly substantial. Ethanol fermentation employs key enzyme ADH1 encoded by ADH1 gene, important for conversion of acetaldehyde to ethanol. However, structural studies about alcohol dehydrogenase1 from these genera of yeasts are limited. This study aimed to detect the alcohol dehydrogenase 1 gene from Pichia spp. Using computational-bioinformatics approaches. The adh1 gene was amplified by PCR, visualized by electrophoresis, and analysed for sequence homology by BlastN and BlastP. The enzyme structure was constructed by SWISS-MODEL and I-TASSER with validation by Ramachandran plot, QMEAN4, and Local Quality Estimate. The Similarity and homology analysis of ADH1 genes and their corresponding protein sequence of yeast isolates showed that the ADH1gene was successfully detected. Multiple sequence alignment (MSA) and phylogenetic tree revealed that W. anomalus BT1-BT6 has close evolutionary relationship with ADH1 from Saccharomyces cerevisiae sequence while P. kudriavzevii IP4 showed different pattern. The ADH 1 enzyme model, generated using the SWISS-MODEL web server, demonstrated the best stereochemical quality, with a Ramachandran plot value of 100% for W. anomalus BT1 and 99.3% for P. kudriavzevii IP4. Superimposition of 3D-predicted model of ADH1 from W. anomalus BT1 and P. kudriavzevii 1P4 showed an exact match with amino acid in Zn2+ binding sites, confirming the ADH1 metaloenzyme properties. These findings provide structural insights about ADH1 genes and protein properties which can be used further for the development of efficient and high productivity of bioethanol productions through genetic and protein engineering.
Extreme daily rainfall in rapidly urbanizing tropical cities frequently overwhelms drainage and disrupts critical services, yet station-scale forecasting remains limited by convective variability and sparse observations. This motivates lightweight, interpretable machine-learning tools that operate on routine station data. We propose and evaluate a station-scale framework to classify heavy-rainfall days (≥50 mm) in a humid tropical setting. Using 1,796 daily observations from the Soekarno-Hatta Meteorological Station (2018–2022), we engineered lag-informed predictors (e.g., previous-day rainfall, 3-day sums/means) and compared three representative classifiers, Logistic Regression (LR), Support Vector Machine (SVM), and Random Forest (RF). Class imbalance was addressed with class-weighted training, and models were assessed on a held-out test set using precision, recall, F1, and Receiver Operating Characteristic – Area Under the Curve (ROC-AUC). LR achieved the highest recall (0.429), indicating moderate sensitivity to rare heavy-rainfall events, whereas RF yielded the best probabilistic discrimination (AUC = 0.619) but failed to flag positives at the default threshold; SVM displayed near-random behavior. Feature analyses highlighted humidity, temperature, and recent rainfall accumulation as the most influential predictors, consistent with tropical convective processes. Despite severe class imbalance, simple, station-based classifiers can extract actionable signals for rare-event screening in data-limited tropical regions. Operational value is likely to improve through probability calibration and threshold tuning, ensemble integration, and spatial generalization to multi-station settings.
Mesona palustris Bl. has considerable potential. Still, it has yet to be matched by its utilization in nanotechnology, especially nanoparticles (1-100 nm), which have superior physicochemical properties and activities. This laboratory study aims to determine the potential of of Mesona palustris Bl. leaf water extract as a bioreductant and capping agent in the biosynthesis of silver nanoparticles (AgNPs), as well as to evaluate the antioxidant activity of the synthesized AgNPs using the DPPH assay. The biosynthesis method produced optimum AgNPs with a ratio of 1% Mesona palustris Bl. leaf water extract, AgNO3, and dispersing medium 0.75:7:5 (v/v) synthesized for 60 minutes at 55°C. The maximum wavelength of AgNPs produced was 433 nm with an absorbance of 0.677. FT-IR spectrophotometric characterization showed that phenolic compounds in the Mesona palustris Bl. leaf water extract is thought to act as a bioreductors and capping agent in the synthesis of AgNPs. SEM results showed that the AgNPs were spherical. Meanwhile, the PSA test results showed the average size of AgNPs was 81 nm, and the polydispersity index was 0.323 (moderately polydispersed). The IC50 value of AgNPs synthesized with Mesona palustris Bl. leaf water extract against DPPH free radicals was 71.501 ± 1.347 μg/mL, which is included in the potent antioxidant and better than Mesona palustris Bl. leaf water extract.
This study aims to determine the effect of administering Jamblang Stem Bark Ethanol Extract (Syzygium cumini L.) (JSBEE) on the activity and phagocytosis capacity of macrophages in mice (Mus musculus) infected with Staphylococcus aureus bacteria. The research method used a completely randomized design (CRD) with five treatments and five repetitions, conducted in-vitro and in vivo test. The treatments for in vitro involved administering distilled water (T0), Stimuno (T1), and JSBEE at concentrations of 10 ppm (T2), 100 ppm (T3), and 1000 ppm (T4). Subsequently, in vivo treatment was conducted using distilled water (T0), stimuno (T1), JSBEE at 10 mg/kg (T2), 100 mg/kg (T3), and 1000 mg/kg (T4). Initially, mice underwent in vivo administration of JSBEE, administered orally via catheter tip, with a dosage of 1 mL per 10 g of body weight. JSBEE was administered orally for 10 days, followed by infection with S. aureus on the 11th day. The in vitro tests were conducted by isolating macrophage cells from the intraperitoneal fluid, to which S. aureus and JSBEE were added. Intraperitoneal fluid collected from the mice was used to prepare smears using the thin blood smear method and Giemsa staining. Macrophage phagocytosis activity was observed and assessed based on the percentage activity formula, and the phagocytic capacity of macrophages was measured by the number of S. aureus cells phagocytosed. The results indicated that JSBEE significantly affected (P0.05) both the activity and phagocytic capacity of macrophages in vivo and in-vitro. The best concentration of JSBEE for increasing both the activity value and phagocytic capacity of macrophages was 100 ppm (T3).
Climate change is happening worldwide, so global climate conditions are a major concern. In densely populated urban areas such as Jakarta, it is impossible to avoid the impacts of climate change, particularly the daily changes in air temperature. Therefore, a sophisticated and efficient approach is needed to find inconsistencies in daily air temperature data to provide critical information for sustainable urban planning and efforts to reduce risks. This research will combine two innovative approaches for hybrid anomaly detection. The method combines generative methods and can extract complex features, such as variational autoencoder (VAE), along with the temporal coding capabilities of long-short-term memory (LSTM), a type of Recurrent Neural Network (RNN). The data used in this study is the average daily air temperature data in Jakarta, obtained from the Kemayoran Meteorological Station and provide by the Meteorology, Climatology, and Geophysics Agency (BMKG). The data used is daily from April 2000 to December 2023. The threshold used to detect anomalies was 229.5, which resulted in excellent performance, namely F1-Score 0.985, Recall 1.000, and Precision 0.971. The VAE-LSTM model identified all dates with significant temperature anomalies, including January 21, 2014, February 22, 2014, November 12, 2014, and February 9, 2015. These dates are significant as they represent extreme weather events that can have severe implications for urban planning and climate change adaptation. The anomalies fall into the categories of point and contextual anomalies. This study contributes to climate research by providing evidence of the effectiveness of deep learning-based hybrid models in detecting complex and context-sensitive temperature anomalies.
