Some say climate change is the biggest threat of our age while others say it’s a myth based on dodgy science. We are turning some of the data over to you so you can form your own view.

Even more than with other data sets that Kaggle has featured, there’s a huge amount of data cleaning and preparation that goes into putting together a long-time study of climate trends. Early data was collected by technicians using mercury thermometers, where any variation in the visit time impacted measurements. In the 1940s, the construction of airports caused many weather stations to be moved. In the 1980s, there was a move to electronic thermometers that are said to have a cooling bias.

Given this complexity, there are a range of organizations that collate climate trends data. The three most cited land and ocean temperature data sets are NOAA’s MLOST, NASA’s GISTEMP and the UK’s HadCrut.

We have repackaged the data from a newer compilation put together by the Berkeley Earth, which is affiliated with Lawrence Berkeley National Laboratory. The Berkeley Earth Surface Temperature Study combines 1.6 billion temperature reports from 16 pre-existing archives. It is nicely packaged and allows for slicing into interesting subsets (for example by country). They publish the source data and the code for the transformations they applied. They also use methods that allow weather observations from shorter time series to be included, meaning fewer observations need to be thrown away.

In this dataset, we have include several files:

Global Land and Ocean-and-Land Temperatures (GlobalTemperatures.csv):

• Date: starts in 1750 for average land temperature and 1850 for max and min land temperatures and global ocean and land temperatures
• LandAverageTemperature: global average land temperature in celsius
• LandAverageTemperatureUncertainty: the 95% confidence interval around the average
• LandMaxTemperature: global average maximum land temperature in celsius
• LandMaxTemperatureUncertainty: the 95% confidence interval around the maximum land temperature
• LandMinTemperature: global average minimum land temperature in celsius
• LandMinTemperatureUncertainty: the 95% confidence interval around the minimum land temperature
• LandAndOceanAverageTemperature: global average land and ocean temperature in celsius
• LandAndOceanAverageTemperatureUncertainty: the 95% confidence interval around the global average land and ocean temperature

Other files include:

• Global Average Land Temperature by Country (GlobalLandTemperaturesByCountry.csv)
• Global Average Land Temperature by State (GlobalLandTemperaturesByState.csv)
• Global Land Temperatures By Major City (GlobalLandTemperaturesByMajorCity.csv)
• Global Land Temperatures By City (GlobalLandTemperaturesByCity.csv)

The raw data comes from the Berkeley Earth data page.

Annual and monthly global mean surface temperature anomalies in degrees Celsius, from NASA GISTEMP and UK Met Office HadCRUT5, covering 1850–present. Anomalies are relative to source-specific base periods.

The table Global Temperatures by Country is part of the dataset Climate Change: Earth Surface Temperature Data, available at https://columbia.redivis.com/datasets/1e0a-f4931vvyg. It contains 577462 rows across 4 variables.

Since 1979, NOAA satellites have been carrying instruments which measure the natural microwave thermal emissions from oxygen in the atmosphere. The intensity of the signals these microwave radiometers measure at different microwave frequencies is directly proportional to the temperature of different, deep layers of the atmosphere. Every month, John Christy and Dr. Roy Spencer update global temperature datasets that represent the piecing together of the temperature data from a total of fifteen instruments flying on different satellites over the years.

Copy Right Information

Copyright 2022 Roy Spencer, Ph. D. - All Rights Reserved

The table Global Temperatures by Major City is part of the dataset Climate Change: Earth Surface Temperature Data, available at https://columbia.redivis.com/datasets/1e0a-f4931vvyg. It contains 239177 rows across 7 variables.

This is the official UK temperature record for the entire planet going back 170+ years. Instead of reporting actual temperatures, it shows how much warmer or cooler each month was compared to the average from 1961 to 1990. That baseline is just a reference point, so you can see the warming trend clearly. The data includes uncertainty ranges because measurements aren't perfect, especially from the 1800s. Scientists use this alongside similar records from the US and other organizations to confirm what's happening with global warming.

Since the 1980s, the annual temperature departure from the average has been consistently positive. In 2024, the global land and ocean surface temperature anomaly stood at 1.29 degrees Celsius above the 20th-century average, the largest recorded across the displayed period. What are temperature anomalies? Temperature anomalies represent the difference from an average or baseline temperature. Positive anomalies show that the observed temperature was warmer than the baseline, whereas a negative anomaly indicates that the observed temperature was lower than the baseline. Land surface temperature anomalies are generally higher than ocean anomalies, although the exact reasons behind this phenomenon are still under debate. Temperature anomalies are generally more important in the study of climate change than absolute temperature, as they are less affected by factors such as station location and elevation. A warming planet The warmest years have been recorded over the past decade, with the highest anomaly in 2024. Global warming has been greatly driven by increased emissions of carbon dioxide and other greenhouse gases into the atmosphere. Climate change is also evident in the declining extent of sea ice in the Northern Hemisphere. Weather dynamics can affect regional temperatures, and therefore, the level of warming can vary around the world. For instance, warming trends and ice loss are most obvious in the Arctic region compared to Antarctica.

Global Mean Temperature by Country (1950–2024)

Global Mean Temperature by Country (1950–2024) is a country-level climate dataset containing annual mean surface air temperatures for 196 countries from 1950 to 2024.

The dataset provides a consistent time series of national average temperatures, making it useful for climate analysis, global warming research, time-series modeling, and environmental data science.

Dataset Structure

Each row represents the average annual temperature for a specific country in a given year.

Columns:

• iso_code – ISO 3166-1 alpha-3 country code
• country – Country name
• year – Year of observation (1950–2024)
• mean_temperature – Mean annual surface air temperature in °C

Dataset Size

• Countries: 196
• Years: 1950–2024
• Total observations: 14,700
• Missing values: None

Potential Use Cases

This dataset can be used for:

• climate change research
• global temperature trend analysis
• time-series forecasting
• machine learning models
• environmental data visualization
• country-level climate comparisons

Source

Temperature values are derived from the CRU TS (Climatic Research Unit Time Series) dataset, which aggregates global weather station data to produce gridded climate observations.

Institution:

• Climatic Research Unit, University of East Anglia

Data accessed through:

• World Bank Climate Change Knowledge Portal

This dataset is deprecated and will be removed on 05/08/26 as part of our ongoing review of the data we publish to ArcGIS. Our current focus is on UK data and scenarios most relevant to risk assessment, but we are looking to enhance the global climate data we publish to ArcGIS and would love your feedback on what you would like to see. In the meantime, a wide selection of climate data is available via the IPCC interactive atlas What does the data show?

This data shows the monthly averages of surface temperature (°C) for 2040-2069 using a combination of the CRU TS (v. 4.06) and UKCP18 global RCP2.6 datasets. The RCP2.6 scenario is an aggressive mitigation scenario where greenhouse gas emissions are strongly reduced.

The data combines a baseline (1981-2010) value from CRU TS (v. 4.06) with an anomaly from UKCP18 global. Where the anomaly is the change in temperature at 2040-2069 relative to 1981-2010.

The data is provided on the WGS84 grid which measures approximately 60km x 60km (latitude x longitude) at the equator.

Limitations of the data

We recommend the use of multiple grid cells or an average of grid cells around a point of interest to help users get a sense of the variability in the area. This will provide a more robust set of values for informing decisions based on the data.

What are the naming conventions and how do I explore the data?

This data contains a field for each month’s average over the period. They are named 'tas' (temperature at surface), the month and ‘upper’ ‘median’ or ‘lower’. E.g. ‘tas Mar Lower’ is the average of the daily average temperatures in March throughout 2040-2069, in the second lowest ensemble member.

To understand how to explore the data, see this page: https://storymaps.arcgis.com/stories/457e7a2bc73e40b089fac0e47c63a578

Please note, if viewing in ArcGIS Map Viewer, the map will default to ‘tas Jan Median’ values.

What do the ‘median’, ‘upper’, and ‘lower’ values mean?

Climate models are numerical representations of the climate system. To capture uncertainty in projections for the future, an ensemble, or group, of climate models are run. Each ensemble member has slightly different starting conditions or model set-ups. Considering all of the model outcomes gives users a range of plausible conditions which could occur in the future.

To select which ensemble members to use, the monthly averages of surface temperature for the period 2040-2069 were calculated for each ensemble member and they were then ranked in order from lowest to highest for each location.

The ‘lower’ fields are the second lowest ranked ensemble member. The ‘upper’ fields are the second highest ranked ensemble member. The ‘median’ field is the central value of the ensemble.

This gives a median value, and a spread of the ensemble members indicating the range of possible outcomes in the projections. This spread of outputs can be used to infer the uncertainty in the projections. The larger the difference between the lower and upper fields, the greater the uncertainty.

Data source

CRU TS v. 4.06 - (downloaded 12/07/22)

UKCP18 v.20200110 (downloaded 17/08/22)

Useful links

Further information on CRU TS Further information on the UK Climate Projections (UKCP) Further information on understanding climate data within the Met Office Climate Data Portal

This version has been superseded by a newer version. It is highly recommended for users to access the current version. Users should only access this superseded version for special cases, such as reproducing studies. If necessary, this version can be accessed by contacting NCEI. The NOAA Global Surface Temperature Dataset (NOAAGlobalTemp) is derived from two independent analyses: the Extended Reconstructed Sea Surface Temperature (ERSST) analysis and the land surface temperature (LST) analysis using the Global Historical Climatology Network (GHCN) temperature database. The data is merged into a monthly global surface temperature dataset dating back from 1880 to the present, updated monthly, in gridded (5 degree x 5 degree) and time series formats. This data set is used in climate monitoring assessments of near-surface temperatures on a global scale. The changes from version 3.5.4 to version 4.0.0 include an update to the primary input dataset (ERSST) now at version 4.0.0 and GHCN-Monthly now at version 3.3.0. This dataset is formerly known as Merged Land-Ocean Surface Temperature (MLOST).

Data are included from the GISS Surface Temperature (GISTEMP) analysis and the global component of Climate at a Glance (GCAG). Two datasets are provided: 1) global monthly mean and 2) annual mean temp...

If you've heard that 'the world is warming,' this is the data behind it. Instead of telling you the actual temperature (which varies wildly by season and location), it shows how much hotter or colder each country was in a given year compared to what was normal from 1991 to 2020. A value of +1 means that year was about 1 degree Celsius warmer than the baseline. A value of -0.5 means it was about half a degree cooler. You can see which countries are warming fastest, whether warming is happening everywhere, and how temperatures have shifted over the past 120 years.

HadCRUT5 is a gridded dataset of global historical surface temperature anomalies relative to a 1961-1990 reference period. Data are available for each month from January 1850 to December 2018 (updates will be available in time), on a 5 degree grid. The dataset is a collaborative product of the Met Office Hadley Centre and the Climatic Research Unit at the University of East Anglia.

This dataset provides high-resolution gridded temperature and precipitation observations from a selection of sources. Additionally the dataset contains daily global average near-surface temperature anomalies. All fields are defined on either daily or monthly frequency. The datasets are regularly updated to incorporate recent observations. The included data sources are commonly known as GISTEMP, Berkeley Earth, CPC and CPC-CONUS, CHIRPS, IMERG, CMORPH, GPCC and CRU, where the abbreviations are explained below. These data have been constructed from high-quality analyses of meteorological station series and rain gauges around the world, and as such provide a reliable source for the analysis of weather extremes and climate trends. The regular update cycle makes these data suitable for a rapid study of recently occurred phenomena or events. The NASA Goddard Institute for Space Studies temperature analysis dataset (GISTEMP-v4) combines station data of the Global Historical Climatology Network (GHCN) with the Extended Reconstructed Sea Surface Temperature (ERSST) to construct a global temperature change estimate. The Berkeley Earth Foundation dataset (BERKEARTH) merges temperature records from 16 archives into a single coherent dataset. The NOAA Climate Prediction Center datasets (CPC and CPC-CONUS) define a suite of unified precipitation products with consistent quantity and improved quality by combining all information sources available at CPC and by taking advantage of the optimal interpolation (OI) objective analysis technique. The Climate Hazards Group InfraRed Precipitation with Station dataset (CHIRPS-v2) incorporates 0.05° resolution satellite imagery and in-situ station data to create gridded rainfall time series over the African continent, suitable for trend analysis and seasonal drought monitoring. The Integrated Multi-satellitE Retrievals dataset (IMERG) by NASA uses an algorithm to intercalibrate, merge, and interpolate “all'' satellite microwave precipitation estimates, together with microwave-calibrated infrared (IR) satellite estimates, precipitation gauge analyses, and potentially other precipitation estimators over the entire globe at fine time and space scales for the Tropical Rainfall Measuring Mission (TRMM) and its successor, Global Precipitation Measurement (GPM) satellite-based precipitation products. The Climate Prediction Center morphing technique dataset (CMORPH) by NOAA has been created using precipitation estimates that have been derived from low orbiter satellite microwave observations exclusively. Then, geostationary IR data are used as a means to transport the microwave-derived precipitation features during periods when microwave data are not available at a location. The Global Precipitation Climatology Centre dataset (GPCC) is a centennial product of monthly global land-surface precipitation based on the ~80,000 stations world-wide that feature record durations of 10 years or longer. The data coverage per month varies from ~6,000 (before 1900) to more than 50,000 stations. The Climatic Research Unit dataset (CRU v4) features an improved interpolation process, which delivers full traceability back to station measurements. The station measurements of temperature and precipitation are public, as well as the gridded dataset and national averages for each country. Cross-validation was performed at a station level, and the results have been published as a guide to the accuracy of the interpolation. This catalogue entry complements the E-OBS record in many aspects, as it intends to provide high-resolution gridded meteorological observations at a global rather than continental scale. These data may be suitable as a baseline for model comparisons or extreme event analysis in the CMIP5 and CMIP6 dataset.

About Dataset

Global Temperature Anomalies (1880-2025) tracks how much hotter (or cooler) Earth has been compared to average temperatures from 1951-1980 🌡️.

Simple Dataset Breakdown

This beginner-friendly CSV has 1,751 rows of monthly data from April 1880 to August 2025. Three easy columns: 1. time ⏲️(like 1880.04 for April 1880).

1. station 🚉(global temperature change in °C)

3.**land** (just land areas in °C)

Great for Learning

• Plot simple line charts to see trends (use matplotlib in jupyter notebook).
• Practice time series basics or basic math like averages.
• Compare global vs. land temperatures side-by-side.

Give your reviews for my first dataset on this platform , I will take any advice from you guys on how to improve this dataset 😀.

This version has been superseded by a newer version. It is highly recommended for users to access the current version. Users should only access this superseded version for special cases, such as reproducing studies. If necessary, this version can be accessed by contacting NCEI. The NOAA Global Surface Temperature Dataset (NOAAGlobalTemp) is a blended product from two independent analysis products: the Extended Reconstructed Sea Surface Temperature (ERSST) analysis and the land surface temperature (LST) analysis using the Global Historical Climatology Network (GHCN) temperature database. The data is merged into a monthly global surface temperature dataset dating back from 1850 to the present. The monthly product output is in gridded (5 degree x 5 degree) and time series formats. The product is used in climate monitoring assessments of near-surface temperatures on a global scale. Changes to the data in version 5.1 included: removing the EOT filtering; filling in data gaps over the polar regions; and extending the beginning data coverage from 1880 to 1850.

In 2024, the global ocean surface temperature was 0.97 degrees Celsius warmer than the 20th-century average. Oceans are responsible for absorbing over 90 percent of the Earth's excess heat from global warming. Departures from average conditions are called anomalies, and temperature anomalies result from recurring weather patterns or longer-term climate change. While the extent of these temperature anomalies fluctuates annually, an upward trend has been observed over the past several decades. Effects of climate change Since the 1980s, every region of the world has consistently recorded increases in average temperatures. These trends coincide with significant growth in the global carbon dioxide emissions, greenhouse gas, and a driver of climate change. As temperatures rise, notable decreases in the extent of arctic sea ice have been recorded. Outlook An increase in emissions from the use of fossil fuels is projected for the coming decades. Nevertheless, global investments in clean energy have increased dramatically since the early 2000s.

Mature Support Notice: This item is in mature support as of April 2026. A replacement item has not been identified at this time. Esri recommends updating your maps and apps to phase out use of this item. Measurements of surface air and ocean temperature are compiled from around the world each month by NOAA’s National Centers for Environmental Information and are analyzed and compared to the 1971-2000 average temperature for each location. The resulting temperature anomaly (or difference from the average) is shown in this feature service, which includes an archive going back to 1880. The 2025 values are the last that will be included and no future updates are planned. The NOAAGlobalTemp dataset is the official U.S. long-term record of global temperature data and is often used to show trends in temperature change around the world. It combines thousands of land-based station measurements from the Global Historical Climatology Network (GHCN) along with surface ocean temperature from the Extended Reconstructed Sea Surface Temperature (ERSST) analysis. These two datasets are merged into a 5-degree resolution product. A report that summarizes the data is released each month (and end of the year) by NOAA NCEI is available here. GHCN monthly mean averages for temperature and precipitation for the 1981-2010 period are also available in Living Atlas here. What can you do with this layer? Visualization: This layer can be used to plot areas where temperature was higher or lower than the historical average for each month going back to 1880. Be sure to configure the time settings in your web map to view the time series correctly. Analysis: This layer can be used as an input to a variety of geoprocessing tools, such as Space Time Cubes and other trend analyses. A version showing just the most recent month is available here.

In April 2024, the global surface air temperature anomaly reached ****°C, relative to the 1991-2020 average for that month. This was the hottest April on record, with an average surface air temperature of *****°C.

Q: Is annual sea surface temperature warmer or cooler than usual? A: Colors on this map show where and by how much annual sea surface temperature differed from a long-term average (1985-1993, details from Coral Reef Watch). Red and orange areas were warmer than average, and blue areas were cooler than average. The darker the color, the larger the difference from the long-term average. White and very light areas were near average. Q: Where do these measurements come from? A: Monthly measurements are made from NOAA's CoralTemp sea surface temperature (SST) data. Every day, instruments on eight satellites in two different orbits (geostationary and polar) measure sea surface temperature by checking how much energy is radiated by the ocean at different wavelengths. Computer programs plot these measurements on a gridded map and then merge and smooth the data into a gap-free product using mathematical filters. Each grid point covers an area approximately 5 x 5 km. Daily temperatures at each grid point are averaged together to calculate monthly average temperature. To calculate the difference-from-average temperatures shown here, a computer program takes the monthly average temperature at each grid point, and subtracts the long-term average for that month. Monthly measurements are averaged together to generate an annual image. If the result is a positive number, the sea surface was warmer than the long-term average. A negative result from the subtraction means the sea surface was cooler than usual. Q: What do the colors mean? A: Shades of blue show locations where sea surface temperature was cooler than its long-term average. Locations shown in shades of orange and red are where the sea’s surface was warmer than the long-term average. The darker the shade of red or blue, the larger the difference from the long-term average or “usual” sea surface temperature. Locations that are white or very light show where sea surface temperature was the same as or very close to its long-term average. Q: Why do these data matter? A: Water covers more than 70% of our planet's surface, so gathering data on ocean temperatures gives us a better picture of global temperatures. Tracking the temperature of the sea’s surface helps scientists understand how much heat energy is in the ocean and how it changes over time. Sea surface temperatures can have dramatic impacts on weather, including weather patterns such as El Niño-Southern Oscillation (ENSO) that travel hundreds of miles inland. Sea surface temperatures also play a significant role in the extent and thickness of Arctic and Antarctic sea ice, which serve as our planet’s built-in air-conditioning system. And sea surface temperatures have significant effects on marine life. The upwelling of cold water, for instance, provides nutrients to phytoplankton, the base of the marine food chain. In contrast, warm ocean surface waters deprive phytoplankton of nutrients, sometimes with devastating effects up the chain. Q: How did you produce these snapshots? A: Data Snapshots are derivatives of existing data products: to meet the needs of a broad audience, we present the source data in a simplified visual style. NOAA's Environmental Visualization Laboratory (NNVL) produces the Sea Surface Temperature Anomaly files. To produce our images, we run a set of scripts that access these NNVL source files, re-project them into a Hammer-Aitoff globe, and output them in a range of sizes. References NOAA NNVL Sea Surface Temperature Anomaly (SSTA) NOAA NNVL SSTA FTP access NOAA Coral Reef Watch CoralTemp data CoralTemp climatology (long-term average) CoralTemp climatology methodology Source: https://www.climate.gov/maps-data/data-snapshots/data-source/sst-global-yearly-difference-average This upload includes two additional files:* SST - Global, Yearly Difference from Average _NOAA Climate.gov.pdf is a scre