Climate Change Trend Using Descriptive Time Series Technique in Machine Learning: A case of Jimma Zone, Southwestern Ethiopia

Climate Change Trend Using Descriptive Time Series Technique in Machine Learning: A case of Jimma Zone, Southwestern Ethiopia | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Climate Change Trend Using Descriptive Time Series Technique in Machine Learning: A case of Jimma Zone, Southwestern Ethiopia Wendafiraw Abdisa Gemmechis This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4421058/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Understanding climate variability and monitoring time-series trends of temperature and rainfall is crucial for the sustainable development of our planet. This study utilized historical data from the Global Historical Climatology Network-Monthly (GHCN-M) provided by the National Centers for Environmental Information (NCEI) to analyze the temperature and rainfall data from 2015 to 2022. The analysis was conducted using Python 3.1.1 on Anaconda Jupyter Notebook and the package matplotlib 3.2.1 was used for data visualization. The results revealed a pattern of maximum rainfall between March to May for the years 2020, 2021, and 2022, while for the years 2017, 2018, and 2019, the maximum rainfall was recorded in October, December, and November. Additionally, the annual maximum rainfalls were recorded in the years 2020 and 2022, and the annual maximum temperatures for all study years were recorded in January, February, and March months. On the other hand, the annual minimum temperatures for all study years occurred in June, July, August, and September months. Similarly, annual average temperatures were recorded in January, February, and March months. This study emphasizes the importance of monitoring climate change and its impacts on our planet. By understanding climate variability and time-series trends, we can better prepare for the future and work towards a sustainable world. Climate Analysis and Modeling Time Series Analysis Precipitation temperature rainfall Jupyter Notebook Matplotlip Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 1. Introduction Climate change represents a significant addition to the spectrum of environmental health hazards that humankind faces. Its global scale generates unfamiliarity, though most of its health impacts are increases or decreases in familiar effects of climatic variation on human biology and health[ 1 ].. Climate variability and change have become major environmental challenges of the 21st century, threatening both food security and sustainable development, as well as the totality of human existence [ 2 ]. In addition according to Belay, F.E., et al., climate change is a global concern as it severely affects the livelihoods of the world community in general and agricultural production as well as food security of rural community in particular. Modeling the variations of surface temperature and making dependable forecasts underlie the foundation of sound environmental policies [ 3 ]. Especially, time series analysis and modeling of mean monthly temperature and total monthly precipitation are two most important climatic variables that easily obtained from weather station records [ 4 ]. However, time series (TS) analysis is a frequent and challenging topic of both today and future research in remote sensing (RS). But these problems are solved using technology like global earth observation (EO) programs that provide extensive temperature and precipitation data archives dating back several years or even more decades (e.g., Sentinel, Landsat etc.). On other hand on a country-wide scale, aerial images, LiDAR and photogrammetric point clouds are acquired in two to three year cycles [ 5 ]. Importantly, time series analysis has always been an important topic in remote sensing it is applied by open access to archives of Earth Observation missions, periodically acquired nation-wide aerial imagery and LiDAR point clouds, and various research datasets, together with open tools for time series data processing, it have brought new possibilities in Remote Sensing research but also challenges in education [ 5 ]. With an increasingly supported and accepted of these open data policy free access to these data sources has become a well - recognized practice. Specific sites are monitored on a regular basis (weekly, monthly, yearly) for diverse research purposes. Time series analyses are used frequently to analyze exposures associated with short-term variability of climate. Time series analyses can take account of cyclical patterns, such as seasonal patterns, when evaluating longitudinal trends in disease rates in one geographically defined population [ 1 ]. Seasonal patterns may be due to the seasonality of climate or to other factors, such as the school year [ 1 ]. Two generally used approaches to time series analyses are generalized additive models (GAM) and generalized estimating equations (GEE). The generalized additive model entails the application of a series of semi-parametric Poisson models that use smoothing functions to capture long-term patterns and seasonal trends from data. However, no a priori smoothing is performed for the time series. Instead, the GEE model allows for the removal of long-term patterns in the data by adjusting for over dispersion and autocorrelation. Autocorrelation needs to be controlled for in time series data of weather measurements because today’s weather is correlated with weather on the previous and subsequent days [ 1 ]. For example a study by Hay et al. demonstrates one approach for looking at longer-term variability (48). The authors used spectral analysis to investigate periodicity in both climate and epidemiological time series data of dengue hemorrhagic fever (DHF) in Bangkok, Thailand. DHF exhibits strong seasonality, with peak incidences in Bangkok occurring during the months of July, August and September. This seasonality has been attributed to temperature variations. Spectral analysis (or Fourier analysis) uses stationary sinusoidal functions to deconstruct time series into separate periodic components. A broad band of two to four year periodic components was identified as well as a large seasonal periodicity. One limitation of this method is that it is applicable only for stationary time series in which the periodic components do not change [ 1 ]. Specifically, Rainfall forecasting is one of the most crucial functions of meteorological departments for all countries. According to Waqas M. et al forecasting rainfall helps to prevent flooding to reduce loss of lives and property and it facilitates the management of water resources [ 6 ]. In other case, temperature data are usually given as monthly means in an equally spaced time series. Instrumental records used in computing these mean values globally have only been available for the past ∼150 years [ 7 ]. Many efforts have been made in the statistical modeling of temperature variations using these records (e.g. [ 8 , 9 ]). Among them, the univariate time-series models have gained relative popularity in recent years, partly due to the complexity of mainstream climate models, which are strongly constrained by the current knowledge of the physical climate system [ 10 ]. One subcategory of the univariate models, namely the structural time-series models [ 11 ], has become quite popular due to its trend-detecting capability [ 12 ]. Many time series forecasting methods are based on the analysis of historical data. They assume that past patterns in the data can be used to forecast future events [ 13 ]. This study was conducted to answer the following questions using time series analysis method 1) what is the monthly minimum temperature between 2015 and 2022 years using time series analysis? 2) What is the monthly maximum temperature between 2015 and 2022 years using time series analysis? 3) What is the monthly precipitation or rainfall between 2015 and 2022 years using time series analysis? 4) How are the trends of average temperature between the stipulated periods using time series analysis? 5) Is there any correlation between Average temperature and rainfall by using ARIMA model? This study aims to investigate climate variability and time-series trends over the past two decades. The study will use time series analysis and the ARIMA model to analyze temperature and precipitation data obtained from the Global Historical Climatology Network-Monthly (GHCN-M) dataset provided by the National Centers for Environmental Information (NCEI). The study focuses on the Jimma area station and covers only one station. The study aims to answer several questions about monthly minimum and maximum temperature, monthly precipitation or rainfall, trends in average temperature, and the correlation between average temperature and rainfall using the ARIMA model. The study was conducted in Jupiter Notebook using Pyspark software and Matpilotlib software. 2. Methods and Materials 2.1. Description of the Study Area: This study took place in the Jimma Zone of the Oromia National Regional State, a region renowned for its cultivation of cash crops like Buna. Situated in the northwestern part of Ethiopia, the study area can be found at 7.6670 latitude and 36.8330 longitude, a distance of 450 km from the capital city of Addis Ababa. With its unique location and rich agricultural heritage, the Jimma Zone is an ideal location for in-depth research and analysis. 2.2. Research Methods The study employed the descriptive time-series analysis method and ARIMA model to forecast the annual and monthly temperature and rainfall of the study area. Both traditional Python-Jupyter Notebook and ARIMA model toolsets in MATLAB, using the Python-Jupyter Notebook platform, were utilized for this purpose. The results of seasonal time-series trends and annual analysis were subjected to descriptive statistics to filter out trends and seasonalities. 2.2.2. Temperature and Rainfall Time Series Analysis Method In a study comparing the links between weather variables and hospitalizations for viral pneumonia, including influenza, during normal weather periods and El Niño events in three regions of California, time series analysis was employed. The study utilized temperature variables and precipitation, which were analyzed using time series analysis techniques. In theory, time series data should be stationary to enable accurate forecasting[ 14 ]. Stationarity refers to the covariance of the variable of interest not being dependent on time but rather on lag. Statistical time series methods and even modern machine learning methods benefit from the clearer signal in the data. When dealing with large input data volumes and greater precision is required, traditional time series methods may fail, and deep learning methods can be used to find the best model of unknown nonlinear relationships in time series data. Deep learning LSTM algorithms can be utilized to take advantage of the extra analytics power of mixed properties data, where both stationary and non-stationary features coexist, for better performance. [ 15 ]. 2.2.3. Augmented Adfuller test In 1976, Box and Jenkin from the University of Pennsylvania introduced the autoregressive integrated moving average (ARIMA) model for predicting stationary time series without seasonal variation [ 16 ]. The ARIMA model has three parameters: p, d, and q. the order of auto-regression is denoted by p, the order of differencing needed to make the time series stationary is represented by d, and q is the order of moving averages. Auto-regression is a technique that predicts a variable by comparing the recent pattern to its predecessors. On the other hand, a regression model employs errors from a previous forecast to predict the outcome of a subsequent forecast. Out of several unit root tests, Augmented Adfuller is one of the widely used tests. It uses an autoregressive model and optimizes an information criterion across different lag values[ 17 ]. The null hypothesis (H0) of the test is that the time series can be represented by a unit root and that it is not stationary. The alternate hypothesis (H1) rejects the null hypothesis and suggests that the time series is stationary and does not have time-dependent structure. We interpret the result using the p-value from the test. A p-value below a threshold (10%, 5%, or 1%) indicates that we reject the null hypothesis (stationary). In contrast, a p-value above the threshold indicates that we fail to reject the null hypothesis (non-stationary). If the p-value is greater than 0.05, we fail to reject the null hypothesis (H0), meaning the data has a unit root and is non-stationary. However, if the p-value is less than or equal to 0.05, we reject the null hypothesis (H0), indicating that the data does not have a unit root and is stationary. The stats models library provides the fuller () function that implements the test. In the case of this study, the Adfuller test results showed that the p-value was less than 0.05, so the data is stationary. I ran this test for the study area datasets, and the results of the Adfuller test are presented in Table 1 below. Table 1 Result of Adfuller Test. Adfuller Test Data ADF Statistic -31.58 p-value 0.047 #lag Used 0.047 number of observations used 96 Critial Values: 1%, -3.512 Critial Values: 5%, -2.897 Critial Values: 10%, -2.586 The ADF test's null hypothesis is that the data is not stationary. In this study, the researchers evaluated the test statistics for the study area's data. They found that it was greater than the critical value (at 10%) and the p-value was less than the significant value of 0.05%. Therefore, they failed to reject the null hypothesis and considered the data as stationary. Consequently, the researchers made the data stationary by differencing and decomposing it. The correlation (C) and partial correlation (PC) are used to determine if the data is autoregressive or not. If C tails off gradually and cuts off after p lags, it's an AR(p) model. If C cuts off after q lags tails off gradually, the model is MA(q). Finally, if both C tails off gradually, the model should be ARIMA (p, q). In this study, the sample data showed that C tailed off gradually, which is why the AR model was used in the ARIMA model. 2.2.4. Seasonal ARIMA/SARIMA Model In a SARIMA model, a non-seasonal ARIMA (p, d, q) is used in combination with seasonal components (P, D, Q) to represent several time steps that make up a period m. Seasonal AR, season moving average, and P, Q signify seasonal differencing, with D [ 17 ]. The acronym ARIMA stands for Auto-Regressive Integrated Moving Average. Algebraically, the ARIMA model can be defined by: yt = µ + b1 yt − 1 +. .. + bp yt − p + ǫt − a1 ǫt − 1 +. .. − aq ǫt − q………………equation(1) At time t = 1,. . ., n, where ǫt − j (j = 0, 1,. . ., q) are the lagged forecast errors. The p + q + 1 unknown parameters µ, b1,. .. bp and a1,. .. aq are determined by minimizing the squared residuals. In the first part of the right-hand side of (1), the dependent variable is predicted based on its values at earlier periods. This is the autoregressive (AR) part of the Eq. (1). In the second part, the dependent variable y also depends on the values of the residuals at earlier periods, which may be regarded as prior random shocks. This is the moving average (MA) part of the equation. The researcher analyzing a given time series must find the relevant parameters of the ARIMA (p, d, q) model, where p is the number of autoregressive terms, d is the number of non-seasonal differences, and q is the number of lagged forecast errors in the prediction equation. Two goals must be met, namely to find the most effective model and to restrict the number of parameters. The residuals should also fulfill the conditions of independence and normality [ 18 ]. As for time series prediction, historical data should be stationary where the covariance of the variable of importance is a function of lag, not of time. The datasets are non-stationary, as demonstrated by descriptive and inferential statistical Adfuller tests (Table 2 ), which indicate that both datasets had monthly trends and seasonal changes in temperature and rainfall (Fig. 2 –9). Stationary data with random fluctuations of radon levels is suitable for assigning to an ARIMA (Auto-Regressive Integrated Moving Average) model that predicts future trends (Fig. 10). 3. Results and discussions 3.1. Time Series Variation of Monthly and Annual Precipitation and Temperature The graph presented in Figure 2 below illustrates the time series variation of precipitation for the study station. The results of the precipitation data reveal some interesting patterns. Firstly, it is observed that the maximum rainfall occurred from March to May for the years 2022, 2021, and 2020. This indicates that these months experienced the highest amount of rainfall during these years. Additionally, the precipitation data identified that the maximum rainfall was recorded for the years 2017, 2018, and 2019 during the months of October to November. This suggests that during these months, there was a significant amount of rainfall as compared to the rest of the year. Overall, the data highlights the seasonal variability of precipitation for the study station and provides useful insights into the rainfall patterns for the past few years. The study conducted a thorough analysis of the annual maximum precipitation data spanning from the years 2015 to 2022. The results of the analysis showed that the years 2020 and 2022 had the highest annual maximum precipitation, with precipitation peaks occurring specifically in the month of April. In contrast, the years 2019 and 2021 recorded the lowest maximum precipitation in April. Furthermore, it was found that the year 2019 experienced its highest precipitation in the month of October, while the year 2021 saw its highest precipitation in May. These findings provide a detailed understanding of the patterns of precipitation over the years, which can be valuable in various fields such as agriculture, hydrology, and environmental studies. 3.2. Time Series Variation of Monthly and Annual Maximum Temperature The graph displayed in Fig. 3 below illustrates the maximum temperature time series on a monthly and annual basis. The analysis of the maximum temperature data reveals that over the years, the highest temperatures were recorded in May, December, and November. However, starting from the year 2019, there has been a noticeable decline in the maximum temperature results recorded in December, while the maximum temperature results in April have been on an upward trend. Furthermore, the maximum temperature recorded in June and July during 2018 and 2019 showed unexpected results with a significant decrease. The maximum temperature throughout the year was recorded in January, February, and March, while the annual maximum temperature in April, May, June, and July was comparatively lower than the other months. This was due to the onset of the spring season in Ethiopia, which brings the first rainfall of the year. 3.3. Time Series Variation of Monthly and Annual Minimum Temperature Figure 4 below presents a map of the time series for minimum temperatures on a monthly and annual basis. The minimum temperature for all years was recorded in December, November, and January. However, in 2016 and 2019, the minimum temperature results for December, November, and January have shown an increasing trend. Furthermore, the monthly minimum temperature recorded in April for all years was relatively higher than that of the other months. Throughout the years, it has been observed that the minimum annual temperature tends to occur during the months of June, July, August, and September, while the maximum annual minimum temperature is typically recorded in January, February, March, and April. These results can be attributed to the fact that these months fall within the winter season in Ethiopia, during which the sun is situated in the northern hemisphere. 3.4. Time Series Variation of Monthly and Annual Average Temperature The graph in Fig. 5 below shows the monthly and annual average temperature time series. The highest average temperatures for all years were recorded in February and March. Additionally, in 2018 and 2020, the average temperature results decreased in June and July, while in 2017 and 2019, the average temperature increased during those months. In April, the monthly minimum temperature recorded was higher than in other months for all study years. Furthermore, between 2015 to 2017 and 2021 to 2022, the average temperatures for all months identified the maximum average temperature respectively. Based on the findings of the annual temperature study conducted in 2016, 2021, and 2022, there was a sudden decrease in the average temperature during May, June, and July. On the other hand, the average temperature showed an increase during January, February, and March for all years under study. 3.5. Correlation between Precipitation and Average Temperature An analysis was conducted to determine the correlation between the average temperature and precipitation. This was done by creating a scatter plot diagram, which is presented in Fig. 6 . The highest correlation was found when the rainfall was less than 60mm and the average temperature ranged between 34°C to 39°C. Table 2 below presents the correlation matrix that includes all the available data. As per the findings of this study, the correlation between precipitation and the average temperature is observed to be very low, with a value of -0.08806. This indicates that the correlation is either close to zero or in the negative territory. On the other hand, the maximum temperature and the average temperature exhibit the most significant correlation, with a value of 0.962518. This is followed by the correlation between minimum temperature and average temperature. Table 2 Overall data correlation matrix. MaxTemp (Maximum Temperature), MinTemp (Minimum Temperatre), Averg (Average Temperature), and Percp (Percipitation). Month Year MaxTemp MinTemp Averg Percp Month 1 0.014487 -0.36693 -0.25032 -0.40006 0.000884 Year 0.014487 1 -0.1021 -0.05973 -0.10856 0.113805 MaxTemp -0.36693 -0.1021 1 0.22302 0.962518 -0.20703 MinTemp -0.25032 -0.05973 0.22302 1 0.479047 0.353547 Averg -0.40006 -0.10856 0.962518 0.479047 1 -0.08806 Percp 0.000884 0.113805 -0.20703 0.353547 -0.08806 1 4. Conclusion The two most critical climatic variables for analyzing climate patterns are the mean monthly temperature and total monthly precipitation, which can be easily obtained from weather station records [ 4 ]. For this study, temperature and precipitation data were collected from the Global Historical Climatology Network-Monthly (GHCN-M) dataset provided by the National Centers for Environmental Information (NCEI) during the study periods from 2015 to 2022. The Python 3.1.1 on Anaconda Jupyter Notebook and the matplotlib 3.2.1 package for data visualization were used for monthly and annual temperature and precipitation time series analysis. The results indicate that the maximum rainfall occurred between March and May for 2020, 2021, and 2022, while for 2017, 2018, and 2019, the maximum rainfall was recorded in October, December, and November. The annual maximum rainfall occurred in 2020 and 2022. The annual maximum temperature for all years was recorded in January, February, and March, while the annual minimum temperature was recorded in June, July, August, and September. Additionally, the annual average temperatures occurred in January, February, and March. The analysis of temperature and rainfall time series is essential for monitoring climate change and sustainable development, providing valuable insights into temperature and rainfall trends for governmental and private sectors working on climate change. The researcher suggests that this paper or data can be used by others to study the prediction of temperature and rainfall using different climate models. Declarations Ethical Consideration The research didn’t involve human subject due that it is not applicable Consent for Publication I have agree to submit journal of climate change and approved the manuscript for submission Competing interests The author declare that he have no competing interests Author Contributions As the author of the research is only one, all of the work is done by him Funding information No funding information Availability of data and materials The data is included in the manuscript References McMichael, A.J., et al., eds. Climate change and human health Risks and Responses. World Health Organization Geneva . 2003. Belay, F.E., et al., Spatio-Temporal Variability and Time Series Trends of Monthly and Seasonal Rainfall Over Northwestern Ethiopia. . Global Journal of Environmental Research, 2020. 14 (2): p. 45-54. Ye L. M., G.X., et al., Time-series modeling and prediction of global monthly absolute temperature for environmental decision making. . Adv. Atmos. Sci., , 2013. 30 (2): p. 382-396. Mulomba, P.M. and C. González-García, Time Series Analysis of Climatic Variables in Peninsular Spain. Trends and Forecasting Models for Data between 20th and 21st Centuries. Climate, 2021. 9 (119). Markéta, P., et al., E-TRAINEE: OPEN E-LEARNING COURSE ON TIME SERIES ANALYSIS IN REMOTE SENSING. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences,, 2023. XLVIII-1/W2-2023 . Waqas, M., et al., Potential of Artificial Intelligence-Based Techniques forRainfall Forecasting in Thailand. A Comprehensive Review. Water 2023. 15 (2979). Jones, P.D. and A. Moberg, 2003: Hemispheric and large-scale surface air temperature variations: . An extensive revision and an update to 2001. Climate, 2001. 16 : p. 206-223. Hansen, J.M., et al., Global temperature change. . Proc. National Academy of Sciences USA,, 2006. 103 : p. 14288-14293. Rahmstorf, A.S., et al., Recent climate observations compared to projections. . Science,, 2007. 316 . IPCC, Detection of climate change and attribution of causes. in Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change, J. T. Houghton et al., Eds., Cambridge University Press. 2001. p. 44. Lee, H.J. and K.-T. Sohn, Prediction of monthly mean surface air temperature in a region of China. . Adv. Atmos. Sci.,, 2007. 24 : p. 503-508. Ye, X.G.M.L., et al., Time-series modeling and prediction of global monthly absolute temperature for environmental decision making. Adv. Atmos. Sci.,, 2013. 30 (2): p. 382-396. Murat, M., et al., Forecasting Daily Meteorological Time Series Using ARIMA and Regression Models. Int. Agrophys., 2018. 32 . Ebi, K.L.e.a., Association of normal weather periods and El Niño events with viral pneumonia hospitalizations in females, California 1983–1998. A. merican Journal of Public Health 91, 2001: p. 1200-1208. Khan, S.M., Application of Deep Learning LSTM and ARIMA Models in Time Series Forecasting: A Methods Case Study analyzing Canadian and Swedish Indoor Air Pollution Data. . Austin J Med Oncol., 2022. 9 (1): p. 1073. Box, P.E.G. and M.G. Jenkins, Times Series Analysis, Forecasting and Control; Taylor & Francis on behalf of the American Statistical Association (USA). San Francisco, CA, USA,, (1976). Manzoor, A. and A. Mansaf, An Intelligent IoT-Cloud-Based Air Pollution Forecasting Model Using Univariate Time-Series Analysis. . Arabian Journal for Science and Engineering., 2023. Hecke, T.V., Time series analysis to forecast temperature change. University College Ghent, 2000. Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4421058","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":302493108,"identity":"a07b7b93-7ea8-4a8c-8a35-c84d79d19d30","order_by":0,"name":"Wendafiraw Abdisa Gemmechis","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+klEQVRIiWNgGAWjYJACCQaGA4xtELYNEDM2HiBFSxpISwNxWhog7MNgEq8Wg+O9B298qLkj28d/+OHjgorzdmvbDwNtqbGJxqnlzLlkyxnHnhm3SaQZG884czt525lEoJZjabkNOLSY3cgxk+ZhO5zYJsFgJs3bdjvZ7ABQC2PDYdxa7r8xk/7zD6iF//g3ad5/55LNzj8koOUGj5k0YxtQCwPQOt6GA3ZmNwjYYn8mx9iytw/kl5xiY55jyQlmN4C2JODxi2T7GcMbP77dkZ3ff3zjY54aO3uz8+kPH3yoscGpBQMkglUmEKsc7FJSFI+CUTAKRsHIAADSfmi3fOJBsAAAAABJRU5ErkJggg==","orcid":"","institution":"Addis Ababa University","correspondingAuthor":true,"prefix":"","firstName":"Wendafiraw","middleName":"Abdisa","lastName":"Gemmechis","suffix":""}],"badges":[],"createdAt":"2024-05-14 18:54:43","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-4421058/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4421058/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":56679182,"identity":"00fe04a8-1c23-41a4-ad15-4d3dba23ce10","added_by":"auto","created_at":"2024-05-17 16:41:11","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":269759,"visible":true,"origin":"","legend":"\u003cp\u003eLocational description of study area\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-4421058/v1/c3df832ffe25f88bcd333729.jpeg"},{"id":56679176,"identity":"a25f4b43-5f27-468c-9c82-0c395ceffaeb","added_by":"auto","created_at":"2024-05-17 16:41:10","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":289050,"visible":true,"origin":"","legend":"\u003cp\u003eMonthly and annual precipitation variation over the years (2015–2022) at Jimma station.\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4421058/v1/58e8c09bed4fffddce95de6c.jpg"},{"id":56679177,"identity":"91ab4c93-abe0-4a32-8d93-3b911feb1d7e","added_by":"auto","created_at":"2024-05-17 16:41:10","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":468377,"visible":true,"origin":"","legend":"\u003cp\u003eMonthly and annual Maximum Temperature variation over the years (2015–2022) at Jimma station.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-4421058/v1/8ee0d4d9e90dbc6e416fe9af.png"},{"id":56679180,"identity":"d6ce31bf-0f3e-4423-852c-32db99b7a543","added_by":"auto","created_at":"2024-05-17 16:41:10","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":325308,"visible":true,"origin":"","legend":"\u003cp\u003eMonthly and annual Minimum Temperature variation over the years (2015–2022).\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-4421058/v1/c6f50dd2837b132f1cd7cec4.png"},{"id":56679179,"identity":"afca3aa3-285c-42d5-9e41-e9d9fec699cd","added_by":"auto","created_at":"2024-05-17 16:41:10","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":309987,"visible":true,"origin":"","legend":"\u003cp\u003eMonthly and annual Average Temperature variation over the years (2015–2022) at Jimma station.\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4421058/v1/4f12f5dda775f471b511ece8.jpg"},{"id":56679181,"identity":"3575318b-9830-4205-b19c-b2bb1a782baf","added_by":"auto","created_at":"2024-05-17 16:41:10","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":45038,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation between precipitation and Annual Average Temperature variation over the years (2015–2022).\u003c/p\u003e","description":"","filename":"6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-4421058/v1/5072b839a2f41ad9e4750863.jpeg"},{"id":56679573,"identity":"566fef65-e61f-478a-9aa5-d8d1d21b5fa6","added_by":"auto","created_at":"2024-05-17 16:49:11","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2006259,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4421058/v1/317c7ada-3e5f-4bdb-8f7b-cf2078d51113.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eClimate Change Trend Using Descriptive Time Series Technique in Machine Learning: A case of Jimma Zone, Southwestern Ethiopia\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eClimate change represents a significant addition to the spectrum of environmental health hazards that humankind faces. Its global scale generates unfamiliarity, though most of its health impacts are increases or decreases in familiar effects of climatic variation on human biology and health[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e].. Climate variability and change have become major environmental challenges of the 21st century, threatening both food security and sustainable development, as well as the totality of human existence [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. In addition according to Belay, F.E., et al., climate change is a global concern as it severely affects the livelihoods of the world community in general and agricultural production as well as food security of rural community in particular.\u003c/p\u003e \u003cp\u003eModeling the variations of surface temperature and making dependable forecasts underlie the foundation of sound environmental policies [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Especially, time series analysis and modeling of mean monthly temperature and total monthly precipitation are two most important climatic variables that easily obtained from weather station records [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. However, time series (TS) analysis is a frequent and challenging topic of both today and future research in remote sensing (RS). But these problems are solved using technology like global earth observation (EO) programs that provide extensive temperature and precipitation data archives dating back several years or even more decades (e.g., Sentinel, Landsat etc.). On other hand on a country-wide scale, aerial images, LiDAR and photogrammetric point clouds are acquired in two to three year cycles [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Importantly, time series analysis has always been an important topic in remote sensing it is applied by open access to archives of Earth Observation missions, periodically acquired nation-wide aerial imagery and LiDAR point clouds, and various research datasets, together with open tools for time series data processing, it have brought new possibilities in Remote Sensing research but also challenges in education [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. With an increasingly supported and accepted of these open data policy free access to these data sources has become a well - recognized practice. Specific sites are monitored on a regular basis (weekly, monthly, yearly) for diverse research purposes.\u003c/p\u003e \u003cp\u003eTime series analyses are used frequently to analyze exposures associated with short-term variability of climate. Time series analyses can take account of cyclical patterns, such as seasonal patterns, when evaluating longitudinal trends in disease rates in one geographically defined population [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Seasonal patterns may be due to the seasonality of climate or to other factors, such as the school year [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eTwo generally used approaches to time series analyses are generalized additive models (GAM) and generalized estimating equations (GEE). The generalized additive model entails the application of a series of semi-parametric Poisson models that use smoothing functions to capture long-term patterns and seasonal trends from data. However, no a priori smoothing is performed for the time series. Instead, the GEE model allows for the removal of long-term patterns in the data by adjusting for over dispersion and autocorrelation. Autocorrelation needs to be controlled for in time series data of weather measurements because today\u0026rsquo;s weather is correlated with weather on the previous and subsequent days [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eFor example a study by Hay et al. demonstrates one approach for looking at longer-term variability (48). The authors used spectral analysis to investigate periodicity in both climate and epidemiological time series data of dengue hemorrhagic fever (DHF) in Bangkok, Thailand. DHF exhibits strong seasonality, with peak incidences in Bangkok occurring during the months of July, August and September. This seasonality has been attributed to temperature variations. Spectral analysis (or Fourier analysis) uses stationary sinusoidal functions to deconstruct time series into separate periodic components. A broad band of two to four year periodic components was identified as well as a large seasonal periodicity. One limitation of this method is that it is applicable only for stationary time series in which the periodic components do not change [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Specifically, Rainfall forecasting is one of the most crucial functions of meteorological departments for all countries. According to Waqas M. et al forecasting rainfall helps to prevent flooding to reduce loss of lives and property and it facilitates the management of water resources [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn other case, temperature data are usually given as monthly means in an equally spaced time series. Instrumental records used in computing these mean values globally have only been available for the past \u0026sim;150 years [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Many efforts have been made in the statistical modeling of temperature variations using these records (e.g. [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]). Among them, the univariate time-series models have gained relative popularity in recent years, partly due to the complexity of mainstream climate models, which are strongly constrained by the current knowledge of the physical climate system [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. One subcategory of the univariate models, namely the structural time-series models [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], has become quite popular due to its trend-detecting capability [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Many time series forecasting methods are based on the analysis of historical data. They assume that past patterns in the data can be used to forecast future events [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThis study was conducted to answer the following questions using time series analysis method 1) what is the monthly minimum temperature between 2015 and 2022 years using time series analysis? 2) What is the monthly maximum temperature between 2015 and 2022 years using time series analysis? 3) What is the monthly precipitation or rainfall between 2015 and 2022 years using time series analysis? 4) How are the trends of average temperature between the stipulated periods using time series analysis? 5) Is there any correlation between Average temperature and rainfall by using ARIMA model?\u003c/p\u003e \u003cp\u003eThis study aims to investigate climate variability and time-series trends over the past two decades. The study will use time series analysis and the ARIMA model to analyze temperature and precipitation data obtained from the Global Historical Climatology Network-Monthly (GHCN-M) dataset provided by the National Centers for Environmental Information (NCEI). The study focuses on the Jimma area station and covers only one station. The study aims to answer several questions about monthly minimum and maximum temperature, monthly precipitation or rainfall, trends in average temperature, and the correlation between average temperature and rainfall using the ARIMA model. The study was conducted in Jupiter Notebook using Pyspark software and Matpilotlib software.\u003c/p\u003e"},{"header":"2. Methods and Materials","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Description of the Study Area:\u003c/h2\u003e \u003cp\u003eThis study took place in the Jimma Zone of the Oromia National Regional State, a region renowned for its cultivation of cash crops like Buna. Situated in the northwestern part of Ethiopia, the study area can be found at 7.6670 latitude and 36.8330 longitude, a distance of 450 km from the capital city of Addis Ababa. With its unique location and rich agricultural heritage, the Jimma Zone is an ideal location for in-depth research and analysis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2. Research Methods\u003c/h2\u003e \u003cp\u003eThe study employed the descriptive time-series analysis method and ARIMA model to forecast the annual and monthly temperature and rainfall of the study area. Both traditional Python-Jupyter Notebook and ARIMA model toolsets in MATLAB, using the Python-Jupyter Notebook platform, were utilized for this purpose. The results of seasonal time-series trends and annual analysis were subjected to descriptive statistics to filter out trends and seasonalities.\u003c/p\u003e \u003cdiv id=\"Sec5\" class=\"Section3\"\u003e \u003ch2\u003e2.2.2. Temperature and Rainfall Time Series Analysis Method\u003c/h2\u003e \u003cp\u003eIn a study comparing the links between weather variables and hospitalizations for viral pneumonia, including influenza, during normal weather periods and El Ni\u0026ntilde;o events in three regions of California, time series analysis was employed. The study utilized temperature variables and precipitation, which were analyzed using time series analysis techniques. In theory, time series data should be stationary to enable accurate forecasting[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Stationarity refers to the covariance of the variable of interest not being dependent on time but rather on lag. Statistical time series methods and even modern machine learning methods benefit from the clearer signal in the data. When dealing with large input data volumes and greater precision is required, traditional time series methods may fail, and deep learning methods can be used to find the best model of unknown nonlinear relationships in time series data. Deep learning LSTM algorithms can be utilized to take advantage of the extra analytics power of mixed properties data, where both stationary and non-stationary features coexist, for better performance. [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section3\"\u003e \u003ch2\u003e2.2.3. Augmented Adfuller test\u003c/h2\u003e \u003cp\u003eIn 1976, Box and Jenkin from the University of Pennsylvania introduced the autoregressive integrated moving average (ARIMA) model for predicting stationary time series without seasonal variation [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. The ARIMA model has three parameters: p, d, and q. the order of auto-regression is denoted by p, the order of differencing needed to make the time series stationary is represented by d, and q is the order of moving averages. Auto-regression is a technique that predicts a variable by comparing the recent pattern to its predecessors. On the other hand, a regression model employs errors from a previous forecast to predict the outcome of a subsequent forecast. Out of several unit root tests, Augmented Adfuller is one of the widely used tests. It uses an autoregressive model and optimizes an information criterion across different lag values[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. The null hypothesis (H0) of the test is that the time series can be represented by a unit root and that it is not stationary. The alternate hypothesis (H1) rejects the null hypothesis and suggests that the time series is stationary and does not have time-dependent structure. We interpret the result using the p-value from the test. A p-value below a threshold (10%, 5%, or 1%) indicates that we reject the null hypothesis (stationary). In contrast, a p-value above the threshold indicates that we fail to reject the null hypothesis (non-stationary). If the p-value is greater than 0.05, we fail to reject the null hypothesis (H0), meaning the data has a unit root and is non-stationary. However, if the p-value is less than or equal to 0.05, we reject the null hypothesis (H0), indicating that the data does not have a unit root and is stationary. The stats models library provides the fuller () function that implements the test. In the case of this study, the Adfuller test results showed that the p-value was less than 0.05, so the data is stationary. I ran this test for the study area datasets, and the results of the Adfuller test are presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e below.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eResult of Adfuller Test.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAdfuller Test\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eData\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eADF Statistic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-31.58\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.047\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e#lag Used\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.047\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003enumber of observations used\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e96\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCritial Values: 1%,\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-3.512\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCritial Values: 5%,\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-2.897\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCritial Values: 10%,\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-2.586\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe ADF test's null hypothesis is that the data is not stationary. In this study, the researchers evaluated the test statistics for the study area's data. They found that it was greater than the critical value (at 10%) and the p-value was less than the significant value of 0.05%. Therefore, they failed to reject the null hypothesis and considered the data as stationary. Consequently, the researchers made the data stationary by differencing and decomposing it. The correlation (C) and partial correlation (PC) are used to determine if the data is autoregressive or not. If C tails off gradually and cuts off after p lags, it's an AR(p) model. If C cuts off after q lags tails off gradually, the model is MA(q). Finally, if both C tails off gradually, the model should be ARIMA (p, q). In this study, the sample data showed that C tailed off gradually, which is why the AR model was used in the ARIMA model.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e2.2.4. Seasonal ARIMA/SARIMA Model\u003c/h2\u003e \u003cp\u003eIn a SARIMA model, a non-seasonal ARIMA (p, d, q) is used in combination with seasonal components (P, D, Q) to represent several time steps that make up a period m. Seasonal AR, season moving average, and P, Q signify seasonal differencing, with D [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. The acronym ARIMA stands for Auto-Regressive Integrated Moving Average. Algebraically, the ARIMA model can be defined by: yt\u0026thinsp;=\u0026thinsp;\u0026micro;\u0026thinsp;+\u0026thinsp;b1 yt\u0026thinsp;\u0026minus;\u0026thinsp;1 +. .. + bp yt\u0026thinsp;\u0026minus;\u0026thinsp;p + ǫt\u0026thinsp;\u0026minus;\u0026thinsp;a1 ǫt\u0026thinsp;\u0026minus;\u0026thinsp;1 +. .. \u0026minus; aq ǫt\u0026thinsp;\u0026minus;\u0026thinsp;q\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;\u0026hellip;equation(1) At time t\u0026thinsp;=\u0026thinsp;1,. . ., n, where ǫt\u0026thinsp;\u0026minus;\u0026thinsp;j (j\u0026thinsp;=\u0026thinsp;0, 1,. . ., q) are the lagged forecast errors. The p\u0026thinsp;+\u0026thinsp;q\u0026thinsp;+\u0026thinsp;1 unknown parameters \u0026micro;, b1,. .. bp and a1,. .. aq are determined by minimizing the squared residuals. In the first part of the right-hand side of (1), the dependent variable is predicted based on its values at earlier periods. This is the autoregressive (AR) part of the Eq.\u0026nbsp;(1). In the second part, the dependent variable y also depends on the values of the residuals at earlier periods, which may be regarded as prior random shocks. This is the moving average (MA) part of the equation. The researcher analyzing a given time series must find the relevant parameters of the ARIMA (p, d, q) model, where p is the number of autoregressive terms, d is the number of non-seasonal differences, and q is the number of lagged forecast errors in the prediction equation. Two goals must be met, namely to find the most effective model and to restrict the number of parameters. The residuals should also fulfill the conditions of independence and normality [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAs for time series prediction, historical data should be stationary where the covariance of the variable of importance is a function of lag, not of time. The datasets are non-stationary, as demonstrated by descriptive and inferential statistical Adfuller tests (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), which indicate that both datasets had monthly trends and seasonal changes in temperature and rainfall (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e\u0026ndash;9). Stationary data with random fluctuations of radon levels is suitable for assigning to an ARIMA (Auto-Regressive Integrated Moving Average) model that predicts future trends (Fig.\u0026nbsp;10).\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"3. Results and discussions","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\n\u003ch2\u003e3.1. Time Series Variation of Monthly and Annual Precipitation and Temperature\u003c/h2\u003e\n\u003cp\u003eThe graph presented in Figure 2 below illustrates the time series variation of precipitation for the study station. The results of the precipitation data reveal some interesting patterns. Firstly, it is observed that the maximum rainfall occurred from March to May for the years 2022, 2021, and 2020. This indicates that these months experienced the highest amount of rainfall during these years. Additionally, the precipitation data identified that the maximum rainfall was recorded for the years 2017, 2018, and 2019 during the months of October to November. This suggests that during these months, there was a significant amount of rainfall as compared to the rest of the year. Overall, the data highlights the seasonal variability of precipitation for the study station and provides useful insights into the rainfall patterns for the past few years. The study conducted a thorough analysis of the annual maximum precipitation data spanning from the years 2015 to 2022. The results of the analysis showed that the years 2020 and 2022 had the highest annual maximum precipitation, with precipitation peaks occurring specifically in the month of April. In contrast, the years 2019 and 2021 recorded the lowest maximum precipitation in April. Furthermore, it was found that the year 2019 experienced its highest precipitation in the month of October, while the year 2021 saw its highest precipitation in May. These findings provide a detailed understanding of the patterns of precipitation over the years, which can be valuable in various fields such as agriculture, hydrology, and environmental studies.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\n\u003ch2\u003e3.2. Time Series Variation of Monthly and Annual Maximum Temperature\u003c/h2\u003e\n\u003cp\u003eThe graph displayed in Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e below illustrates the maximum temperature time series on a monthly and annual basis. The analysis of the maximum temperature data reveals that over the years, the highest temperatures were recorded in May, December, and November. However, starting from the year 2019, there has been a noticeable decline in the maximum temperature results recorded in December, while the maximum temperature results in April have been on an upward trend. Furthermore, the maximum temperature recorded in June and July during 2018 and 2019 showed unexpected results with a significant decrease.\u003c/p\u003e\n\u003cp\u003eThe maximum temperature throughout the year was recorded in January, February, and March, while the annual maximum temperature in April, May, June, and July was comparatively lower than the other months. This was due to the onset of the spring season in Ethiopia, which brings the first rainfall of the year.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n\u003ch2\u003e3.3. Time Series Variation of Monthly and Annual Minimum Temperature\u003c/h2\u003e\n\u003cp\u003eFigure \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e below presents a map of the time series for minimum temperatures on a monthly and annual basis. The minimum temperature for all years was recorded in December, November, and January. However, in 2016 and 2019, the minimum temperature results for December, November, and January have shown an increasing trend. Furthermore, the monthly minimum temperature recorded in April for all years was relatively higher than that of the other months.\u003c/p\u003e\n\u003cp\u003eThroughout the years, it has been observed that the minimum annual temperature tends to occur during the months of June, July, August, and September, while the maximum annual minimum temperature is typically recorded in January, February, March, and April. These results can be attributed to the fact that these months fall within the winter season in Ethiopia, during which the sun is situated in the northern hemisphere.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\n\u003ch2\u003e3.4. Time Series Variation of Monthly and Annual Average Temperature\u003c/h2\u003e\n\u003cp\u003eThe graph in Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e below shows the monthly and annual average temperature time series. The highest average temperatures for all years were recorded in February and March. Additionally, in 2018 and 2020, the average temperature results decreased in June and July, while in 2017 and 2019, the average temperature increased during those months. In April, the monthly minimum temperature recorded was higher than in other months for all study years. Furthermore, between 2015 to 2017 and 2021 to 2022, the average temperatures for all months identified the maximum average temperature respectively.\u003c/p\u003e\n\u003cp\u003eBased on the findings of the annual temperature study conducted in 2016, 2021, and 2022, there was a sudden decrease in the average temperature during May, June, and July. On the other hand, the average temperature showed an increase during January, February, and March for all years under study.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\n\u003ch2\u003e3.5. Correlation between Precipitation and Average Temperature\u003c/h2\u003e\n\u003cp\u003eAn analysis was conducted to determine the correlation between the average temperature and precipitation. This was done by creating a scatter plot diagram, which is presented in Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e. The highest correlation was found when the rainfall was less than 60mm and the average temperature ranged between 34\u0026deg;C to 39\u0026deg;C.\u003c/p\u003e\n\u003cp\u003eTable\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e below presents the correlation matrix that includes all the available data. As per the findings of this study, the correlation between precipitation and the average temperature is observed to be very low, with a value of -0.08806. This indicates that the correlation is either close to zero or in the negative territory. On the other hand, the maximum temperature and the average temperature exhibit the most significant correlation, with a value of 0.962518. This is followed by the correlation between minimum temperature and average temperature.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n\u003ctable id=\"Tab2\" border=\"1\"\u003e\u003ccaption\u003e\n\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n\u003cdiv class=\"CaptionContent\"\u003e\n\u003cp\u003eOverall data correlation matrix. MaxTemp (Maximum Temperature), MinTemp (Minimum Temperatre), Averg (Average Temperature), and Percp (Percipitation).\u003c/p\u003e\n\u003c/div\u003e\n\u003c/caption\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eMonth\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eYear\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eMaxTemp\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eMinTemp\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eAverg\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003ePercp\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eMonth\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.014487\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.36693\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.25032\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.40006\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.000884\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eYear\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.014487\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.1021\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.05973\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.10856\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.113805\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eMaxTemp\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.36693\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.1021\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.22302\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.962518\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.20703\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eMinTemp\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.25032\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.05973\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.22302\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.479047\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.353547\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eAverg\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.40006\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.10856\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.962518\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.479047\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e\u003cstrong\u003e-0.08806\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003ePercp\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.000884\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.113805\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.20703\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e0.353547\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e-0.08806\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e1\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003c/div\u003e\n\u003c/div\u003e"},{"header":"4. Conclusion","content":"\u003cp\u003eThe two most critical climatic variables for analyzing climate patterns are the mean monthly temperature and total monthly precipitation, which can be easily obtained from weather station records [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. For this study, temperature and precipitation data were collected from the Global Historical Climatology Network-Monthly (GHCN-M) dataset provided by the National Centers for Environmental Information (NCEI) during the study periods from 2015 to 2022. The Python 3.1.1 on Anaconda Jupyter Notebook and the matplotlib 3.2.1 package for data visualization were used for monthly and annual temperature and precipitation time series analysis. The results indicate that the maximum rainfall occurred between March and May for 2020, 2021, and 2022, while for 2017, 2018, and 2019, the maximum rainfall was recorded in October, December, and November. The annual maximum rainfall occurred in 2020 and 2022. The annual maximum temperature for all years was recorded in January, February, and March, while the annual minimum temperature was recorded in June, July, August, and September. Additionally, the annual average temperatures occurred in January, February, and March. The analysis of temperature and rainfall time series is essential for monitoring climate change and sustainable development, providing valuable insights into temperature and rainfall trends for governmental and private sectors working on climate change. The researcher suggests that this paper or data can be used by others to study the prediction of temperature and rainfall using different climate models.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthical Consideration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe research didn\u0026rsquo;t involve human subject due that it is not applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for Publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eI have agree to submit journal of climate change and approved the manuscript for submission\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe author declare that he have no competing interests\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs the author of the research is only one, all of the work is done by him\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNo funding information\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data is included in the manuscript\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eMcMichael, A.J., et al., eds. \u003cem\u003eClimate change and human health Risks and Responses. World Health Organization Geneva\u003c/em\u003e. 2003.\u003c/li\u003e\n\u003cli\u003eBelay, F.E., et al., \u003cem\u003eSpatio-Temporal Variability and Time Series Trends of Monthly and Seasonal Rainfall Over Northwestern Ethiopia. .\u003c/em\u003e Global Journal of Environmental Research, 2020. \u003cstrong\u003e14\u003c/strong\u003e(2): p. 45-54.\u003c/li\u003e\n\u003cli\u003eYe L. M., G.X., et al., \u003cem\u003eTime-series modeling and prediction of global monthly absolute temperature for environmental decision making. .\u003c/em\u003e Adv. Atmos. Sci., , 2013. \u003cstrong\u003e30\u003c/strong\u003e(2): p. 382-396.\u003c/li\u003e\n\u003cli\u003eMulomba, P.M. and C. Gonz\u0026aacute;lez-Garc\u0026iacute;a, \u003cem\u003eTime Series Analysis of Climatic Variables in Peninsular Spain. Trends and Forecasting Models for Data between 20th and 21st Centuries. \u003c/em\u003eClimate, 2021. \u003cstrong\u003e9\u003c/strong\u003e(119).\u003c/li\u003e\n\u003cli\u003eMark\u0026eacute;ta, P., et al., \u003cem\u003eE-TRAINEE: OPEN E-LEARNING COURSE ON TIME SERIES ANALYSIS IN REMOTE SENSING.\u003c/em\u003e The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences,, 2023. \u003cstrong\u003eXLVIII-1/W2-2023\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eWaqas, M., et al.,\u003cem\u003e Potential of Artificial Intelligence-Based Techniques forRainfall Forecasting in Thailand.\u003c/em\u003e A Comprehensive Review. Water 2023. \u003cstrong\u003e15\u003c/strong\u003e(2979).\u003c/li\u003e\n\u003cli\u003eJones, P.D. and A. Moberg,\u003cem\u003e 2003: Hemispheric and large-scale surface air temperature variations: .\u003c/em\u003e An extensive revision and an update to 2001. Climate, 2001. \u003cstrong\u003e16\u003c/strong\u003e: p. 206-223.\u003c/li\u003e\n\u003cli\u003eHansen, J.M., et al., \u003cem\u003eGlobal temperature change. .\u003c/em\u003e Proc. National Academy of Sciences USA,, 2006. \u003cstrong\u003e103\u003c/strong\u003e: p. 14288-14293.\u003c/li\u003e\n\u003cli\u003eRahmstorf, A.S., et al., \u003cem\u003eRecent climate observations compared to projections. .\u003c/em\u003e Science,, 2007. \u003cstrong\u003e316\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eIPCC, \u003cem\u003eDetection of climate change and attribution of causes. \u003c/em\u003ein \u003cem\u003eClimate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change, J. T. Houghton et al., Eds., Cambridge University Press.\u003c/em\u003e 2001. p. 44.\u003c/li\u003e\n\u003cli\u003eLee, H.J. and K.-T. Sohn, \u003cem\u003ePrediction of monthly mean surface air temperature in a region of China. .\u003c/em\u003e Adv. Atmos. Sci.,, 2007. \u003cstrong\u003e24\u003c/strong\u003e: p. 503-508.\u003c/li\u003e\n\u003cli\u003eYe, X.G.M.L., et al., \u003cem\u003eTime-series modeling and prediction of global monthly absolute temperature for environmental decision making.\u003c/em\u003e Adv. Atmos. Sci.,, 2013. \u003cstrong\u003e30\u003c/strong\u003e(2): p. 382-396.\u003c/li\u003e\n\u003cli\u003eMurat, M., et al., \u003cem\u003eForecasting Daily Meteorological Time Series Using ARIMA and Regression Models.\u003c/em\u003e Int. Agrophys., 2018. \u003cstrong\u003e32\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eEbi, K.L.e.a., \u003cem\u003eAssociation of normal weather periods and El Ni\u0026ntilde;o events with viral pneumonia hospitalizations in females, California 1983\u0026ndash;1998. A.\u003c/em\u003e \u003cem\u003emerican Journal of Public Health \u003c/em\u003e91, 2001: p. 1200-1208.\u003c/li\u003e\n\u003cli\u003eKhan, S.M., \u003cem\u003eApplication of Deep Learning LSTM and ARIMA Models in Time Series Forecasting: A Methods Case Study analyzing Canadian and Swedish Indoor Air Pollution Data. .\u003c/em\u003e Austin J Med Oncol., 2022. \u003cstrong\u003e9\u003c/strong\u003e(1): p. 1073.\u003c/li\u003e\n\u003cli\u003eBox, P.E.G. and M.G. Jenkins, \u003cem\u003eTimes Series Analysis, Forecasting and Control; Taylor \u0026amp; Francis on behalf of the American Statistical Association (USA).\u003c/em\u003e San Francisco, CA, USA,, (1976).\u003c/li\u003e\n\u003cli\u003eManzoor, A. and A. Mansaf,\u003cem\u003e An Intelligent IoT-Cloud-Based Air Pollution Forecasting Model Using Univariate Time-Series Analysis. .\u003c/em\u003e Arabian Journal for Science and Engineering., 2023.\u003c/li\u003e\n\u003cli\u003eHecke, T.V., \u003cem\u003eTime series analysis to forecast temperature change.\u003c/em\u003e University College Ghent, 2000.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Addis Ababa University","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Time Series Analysis, Precipitation, temperature, rainfall, Jupyter Notebook, Matplotlip","lastPublishedDoi":"10.21203/rs.3.rs-4421058/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4421058/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eUnderstanding climate variability and monitoring time-series trends of temperature and rainfall is crucial for the sustainable development of our planet. This study utilized historical data from the Global Historical Climatology Network-Monthly (GHCN-M) provided by the National Centers for Environmental Information (NCEI) to analyze the temperature and rainfall data from 2015 to 2022. The analysis was conducted using Python 3.1.1 on Anaconda Jupyter Notebook and the package matplotlib 3.2.1 was used for data visualization. The results revealed a pattern of maximum rainfall between March to May for the years 2020, 2021, and 2022, while for the years 2017, 2018, and 2019, the maximum rainfall was recorded in October, December, and November. Additionally, the annual maximum rainfalls were recorded in the years 2020 and 2022, and the annual maximum temperatures for all study years were recorded in January, February, and March months. On the other hand, the annual minimum temperatures for all study years occurred in June, July, August, and September months. Similarly, annual average temperatures were recorded in January, February, and March months. This study emphasizes the importance of monitoring climate change and its impacts on our planet. By understanding climate variability and time-series trends, we can better prepare for the future and work towards a sustainable world.\u003c/p\u003e","manuscriptTitle":"Climate Change Trend Using Descriptive Time Series Technique in Machine Learning: A case of Jimma Zone, Southwestern Ethiopia","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-05-17 16:41:05","doi":"10.21203/rs.3.rs-4421058/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7881a807-b3e5-4da9-9919-6f9261629936","owner":[],"postedDate":"May 17th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":31918443,"name":"Climate Analysis and Modeling"}],"tags":[],"updatedAt":"2024-05-17T16:41:05+00:00","versionOfRecord":[],"versionCreatedAt":"2024-05-17 16:41:05","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4421058","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4421058","identity":"rs-4421058","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}