4 ms·

From the data source at [0]: > Data are currently available from September 1, 1981. Seems like a short time range to analyze, but I guess most weather data is

by haolez 3y ago

From the data source at [0]: > Data are currently available from September 1, 1981. Seems like a short time range to analyze, but I guess most weather data is recent-ish. [0] https://www.ncei.noaa.gov/products/optimum-interpolation-sst https://www.ncei.noaa.gov/products/optimum-interpolation-sst

dredmorbius 3y ago
Data spans vary considerably. There are detailed contemporary human observations dating back a couple of centuries, on both land (at specific locations) and sea (mostly via ships logs). These can provide extensive direct measurement and are useful in calibrating other sources. Remote satellite sensing of course dates back only to the late 1950s, and for the most part useful measurements to the 1970s and 80s. These are useful in being global and detailed, but of course offer limited timespans. Proxy measures from an incredible array of biological and geological factors and indicators can provide strong indicia of extant conditions ranging from 100s to 1,000s to 1,000,000s of years. Among some that I'm aware of are isotope ratios in trapped gasses (found in ice cores, ranging back ~800,000 years), mineralisation of seashells and seashell-derived limestones (dating back 100s of millions of years), tree-ring data and pollen deposition in lakes and bogs (both dating back 10s of thousands of years), and again isotopic patterns in geological strata (cruder but ranging back billions of years), as well as signs of glaciation (indicating sub-freezing surface temperatures), of ancient seas, lakes, or rivers (indicating sea levels and precipitation patterns), dust deposition in seafloor strata (drill cores from the Mid-Atlantic Ridge were pivotal in determining periods of desertification and temperate climates in the Sahara / North African region), amongst others. It's of course tricky to directly compare two different data series, but it's possible, and the narrative which emerges is both generally consistent and points to the critical role of atmospheric CO2 and other greenhouse gasses. <https://skepticalscience.com/surface-temperature-measurements.htm https://skepticalscience.com/surface-temperature-measurement...> <https://skepticalscience.com/big-picture.html https://skepticalscience.com/big-picture.html> <https://scitechdaily.com/66-million-years-of-earths-climate-history-uncovered-puts-current-changes-in-context/?expand_article=1 https://scitechdaily.com/66-million-years-of-earths-climate-...>
photochemsyn 3y ago
I think this is the more comprehensive long-term dataset: https://www.ncei.noaa.gov/products/extended-reconstructed-sst https://www.ncei.noaa.gov/products/extended-reconstructed-ss... Note also the most reliable data on this is probably from the Argo floats, there are about 4000 of them deployed globally (collected data besides SST, such as mixed-layer depth, salinity etc.), since 2000: https://argo.ucsd.edu/about/status/ https://argo.ucsd.edu/about/status/

by haolez 3y ago

dredmorbius 3y ago

Data spans vary considerably. There are detailed contemporary human observations dating back a couple of centuries, on both land (at specific locations) and sea (mostly via ships logs). These can provide extensive direct measurement and are useful in calibrating other sources. Remote satellite sensing of course dates back only to the late 1950s, and for the most part useful measurements to the 1970s and 80s. These are useful in being global and detailed, but of course offer limited timespans. Proxy measures from an incredible array of biological and geological factors and indicators can provide strong indicia of extant conditions ranging from 100s to 1,000s to 1,000,000s of years. Among some that I'm aware of are isotope ratios in trapped gasses (found in ice cores, ranging back ~800,000 years), mineralisation of seashells and seashell-derived limestones (dating back 100s of millions of years), tree-ring data and pollen deposition in lakes and bogs (both dating back 10s of thousands of years), and again isotopic patterns in geological strata (cruder but ranging back billions of years), as well as signs of glaciation (indicating sub-freezing surface temperatures), of ancient seas, lakes, or rivers (indicating sea levels and precipitation patterns), dust deposition in seafloor strata (drill cores from the Mid-Atlantic Ridge were pivotal in determining periods of desertification and temperate climates in the Sahara / North African region), amongst others. It's of course tricky to directly compare two different data series, but it's possible, and the narrative which emerges is both generally consistent and points to the critical role of atmospheric CO2 and other greenhouse gasses. <https://skepticalscience.com/surface-temperature-measurements.htm https://skepticalscience.com/surface-temperature-measurement...> <https://skepticalscience.com/big-picture.html https://skepticalscience.com/big-picture.html> <https://scitechdaily.com/66-million-years-of-earths-climate-history-uncovered-puts-current-changes-in-context/?expand_article=1 https://scitechdaily.com/66-million-years-of-earths-climate-...>

photochemsyn 3y ago

I think this is the more comprehensive long-term dataset: https://www.ncei.noaa.gov/products/extended-reconstructed-sst https://www.ncei.noaa.gov/products/extended-reconstructed-ss... Note also the most reliable data on this is probably from the Argo floats, there are about 4000 of them deployed globally (collected data besides SST, such as mixed-layer depth, salinity etc.), since 2000: https://argo.ucsd.edu/about/status/ https://argo.ucsd.edu/about/status/