Assessment type : Final report Weight : 60% Length : 6,000 words. Instructions Prepare a written final report based on your completed project. The final report must be structured as follows: ·...

FINAL REPORT
Introduction
    Due to over population and increasing demand for resources, we are facing a major problem regarding climate change. Climate change not only affects the seasonal patterns but also includes major issues such as global warming and other environmental effects. It has been researched that climate changes occur due to the emission of greenhouse gases and other gases such as CO2 and methane. As greenhouse gases rise up, the global surface temperature rises too. It is on record that last decade i.e. 2011-2021 has been the warmest on the record. We can observe that land areas are seeing more hot waves and thus people are reluctant to go outdoors. According to WHO, they call climate change one of the global threats in the 21st century. Thus climate affects mostly all the lives which starts from the food sources to the transportation and what not. It has a huge impact on our livelihoods, our health and future. Climate change is one of the topics that determine the expectations of having the future climatic patterns. By using the latitude figure, it is sort of easy to determine the chances of having the snow and hail which are reaching the Earth’s surface. This topic is actually becoming the scientific study of climatic changes over a period of time. According to WHO (World Health Organization), they call climate change one of the global threats in the 21st century. Thus climate affects mostly all the lives which starts from the food sources to the transportation and what not. It has a huge impact on our livelihoods, our health and future. Climate change is one of the topics that determine the expectations of having the future climatic patterns.
    We will be analyzing the dataset that is taken up from the kaggle which focus on the trends of climatic change over a period of time. Earlier this data is being collected by technicians using the mercury thermometers where there are variations observed in the time measurements. Later on , in 1940s, airports got constructed leading to many migration of weather stations and further on in the 1980s, there was an introduction of electronic thermometers which are based on the cooling bias.
    Given such a complexity in the measurements done by these thermometers, many range of organizations work together to check on the climatic issues. Data sets such as NOAA’s MLOST, UK’s HadCrut etc plays a major role in citing the land and ocean temperatures over the period of time. Here, finally we have repackaged the whole dataset to form a new combination which is given by the Berkeley Earth, that is affiliated with the Lawrence Berkeley National Laboratory. This laboratory studied the earth surface temperature which combines all the 1.6 billion temperature reports that are taken up by 16 pre-existing folders. This packaging of data is done in such a way that interesting sets can be obtained on the basis of country, city and state. This source data is published and made some transformations through different research. We will be using the same dataset to check the study of earth surface temperature and can get weather observations in different perspective.
    In this dataset, we have several files such as global land and ocean land temperature called as GlobalTemperatures.csv where the following depicts the attributes present in this dataset which are as follows:
1. Date: It majorly starts in 1750 for depicting an average land temperature and 1850 is considered as max and min land and global ocean and land temperature
2. LandAverageTemperature: It depicts global average land temperature is measured in Celsius
3. LandAverageTemperatureUncertainty: It depicts the land average temperature with 95% confidence interval
4. LandMaxTemperature: It is depicting the global average maximum temperature related to land measured in Celsius
5. LandMaxTemperatureUncertainty: It is depicting the land temperature which shows maximum 95% confidence claim
6. LandMinTemperature: It is depicting the minimum land temperature measured in Celsius
7. LandMinTemperatureUncertainty: It is depicting the minimum land temperature with claim of 95% confidence claim
8. LandAndOceanAverageTemperature: It is depicting the global ocean and land temperature measured in Celsius
9. LandAndOceanAverageTemperatureUncertainty: It is depicting global ocean and land temperature with 95% confidence interval claim
We have files such as global land temperature by country, state, city and major city and this is taken up from the Berkeley data set with earth’s surface temperature. We will be using the regression analysis to check the visualization scenario. We will complete the whole analysis using the Python language and its libraries such as Matplotlib, Numpy, pandas , seaborn etc We will calculate the descriptive statistics of the variables and clearly analyze the trends appearing the data.
Institutional Context

Research Question: Why and how earth temperature is significantly changing based on different parameters such as country, city and state and impacting climatic changes?
        We will be reviewing the literature context of climatic change by using some of the reports and current highest records. It has been seen that the world’s longest unbreakable record for the carbon dioxide emissions is from one of the observatory called as Mauna Loa Observatory popularly known as the Keeling curve. Many of the scientists made the atmospheric measurements in different remote locations to get the sample air and study them to derive the insights. The Berkeley Earth Surface Temperature Study takes all the temperature reports which are in 1.6 billion sizes which are extracted from the pre existing folders. Based on these records, we can visualize the weather observations for different countries. We will be checking all the interpretations to get the statistical measurement of the entire atmospheric figure.
PREVIOUS RESEARCH
·     We have a funnel approach by introducing data set by studying different interpretations such as XCDAT approach which determine the special features with xarray that is actually an extension of having a climatic data analysis on different structure grids. It includes features such as temporal averaging stuff like average of time series obtained with unweighted or weighted climatic parameters and optional seasonal configuration figures. It supports a rectilinear grid for geospatial details and is developed by the developers from the Energy Exascale Earth System Model which sponsors and works on the analysis of climatic change. A. Kuchar proposed an Orographic Atmospheric Wave analysis that is based on the visualization of the model that takes the repository datasets for predicting and deriving the insights of Orographic Gravity Wave Dynamics
Haley Egan and his friends proposed Natural Disasters Over Time , an analytical report on the dataset that is being drawn from the International Disaster Database which includes technological and other hazardous disasters such as chemical spills and wind speed to check the climatic conditions over the natural disasters and its being researched that natural disasters are increasing on a daily basis due to the changes made in the climatic conditions

·      Victor Schmidt proposed to visualize the results of the climatic conditions represented as ClimateGAN which comprise of images and labels from the 3D virtual world where we evaluate the performance of the model by using the flood masks and its decoder to get the details and thus visualization of floods for all the urban, suburban and rural areas. This ClimateGAN may have some limitations regarding the height of the floods as there is no dataset to check the metric height of the street. The National Snow and Ice Data Center provides support towards researching the topic of the world's frozen things such as snow, frozen ground, ice, glaciers etc. and see its interaction with climatic patterns by using Daily Sea Ice Extent Data. By performing regression analysis on the dataset provided, it is concluded that the global mean sea level is increasing on a yearly basis and the problem needs to be addressed quickly. Greenhouse Gas Inventory provides substantial information from 1990 till the latest year regarding the anthropogenic emissions by different sources and the process of removal of greenhouse gases such as carbon dioxide, methane, nitrous oxide, hydrofluorocarbons, nitrogen trifluoride etc. those are mostly not controllable by the Montreal protocol. It has been found out those countries such as the USA, Russia and Japan are mostly responsible for greenhouse gas emissions. It has been achieved through the use of International GreenHouse Gas Emissions dataset.
By using the references, we will now look into the Berkeley Earth Surface Temperature dataset that will give more insights regarding the climatic changes where information regarding temperature and average of ocean temperature is analyzed properly to work on subsets of data that differs based on country wise.
Data Source and Empirical Approach
    For the data source, we will be using a combination of datasets that suggest climatic changes. We have used the combination of the land and ocean temperature data sets such as NASA’s GISTEMP, NOAA’s MLOST and UK’s HadCrut. These datasets are packaged and a new variation has been put together by the Berkeley Earth, which is having an affiliation with the Lawrence Berkeley National Laboratory. It has combined up to 1.6 billion temperature reports from the existing archive data and is distinguished based on the name of the country. They have published the data and used different methodologies to check the weather observations.
    We will be using Python, as a programming language and a tool with import of different libraries such as Matplotlib, Pandas, Numpy, seaborn to visualize and interpret better insights to check different global land and ocean temperature with respect to year, country, city and states.
    
For this dataset, we have used several files which are as follows:
1.     Global Temperature csv file : This file comprised of all the details for the global land and the ocean-and-land temperature whose features are described as follows:
i. Date: This column specifies the date period for the average land and maximum and minimum land temperature for global ocean bodies. For example: It starts with the 1750 and maximum is 1850.
ii. LandAverageTemperature : This feature suggest the global average land temperature measured in degree Celsius
iii. LandAverageTemperatureUncertainty: This feature provides a claim of 95% confidence interval over the feature named as land average temperature
iv. LandMaxTemperature: This feature provides the maximum value specified for the land temperature
v. LandMinTemperature: This feature provides the minimum value specified for the land temperature
vi. LandMinTemperatureUncertainty: This feature provides a claim of 95% confidence interval over the feature named as land minimum temperature
vii. LandMaxTemperatureUncertainty: This feature provides a claim of 95% confidence interval over the feature named as land maximum temperature
viii. LandAndOceanAverageTemperature: This feature provides global land and ocean average temperature measured in degree Celsius
ix. LandAndOceanAverageTemperatureUncertainty: This feature provides a claim of 95% confidence interval over the feature named as land and ocean average temperature
We have same features prescribed for the different countries, states, major cities and cities and they are named as GlobalLandTemperaturesByState, GlobalLandTemperaturesByCountry,...