The Instrumental Record Flashcards
ESD
the instrumental record
refers to pieces of evidence which can be collated to highlight global climate change
interpolation
process in which statistical methods would be used to fill in gaps within the climate record -> uncertainty (Gulev et al., 2021)
CO2 concentrations -> Mauna Loa Observatory (1960->)
increasing overtime -> peak during the 1997/98 El Niño (6th IPCC, Gulev et al., 2021).
Temperature Records -> often divided based on land and SSTs and NH/SH
450 million reports across regions/times -> International Surface Temperature Initiative = 35,000 stations created for land temperatures (6th IPCC, Gulev et al., 2021)
Temperature Record -> HadCRUT4 (1850->) -> collates HADGT3 and CRUTEM4 = 5x5 resolution (Morice et al., 2012)
84% of the data is taken from the surface -> Africa and poles = missing data (Cowtan and Way, 2014)
Gridding -> anomalies at each latitude and longitude are collected -> step-change is analysed in comparison to local stations to determine whether it was anomalous or a climate signal -> produces 12 maps (1/month) and collates them into data points on a time series/year -> uses these to create grid boxes
anomalies -> needed to avoid biases and cover a greater spatial scale -> technicality would appear that the poles are cooling if absolute data taken (Rhein et al., 2013 - 5th IPCC Report)
realisations -> model runs of the climatic conditions to determine the certainty of the data set
statistical means are then taken to represent the magnitude of warming in the grid boxes (Morice et al., 2012)
HadCRUT4 -> does not use interpolation -> leaves regions without data as the average - problematic due to arctic amplification not being represented (were improvements in data sets from HadCRUT3-4 though with Siberia included)
Cowtan and Way (2014) -> filled in the gaps via interpolation/satellite data -> earth has warmed 2x faster in the last 15yrs than the data implies
HadCRUT5 (Met Office) -> 5,000 stations
MSLOT (NOAA) -> 7,000 stations (Pidcock, 2015) -> digitisation has also improves spatial coverage (Morice et al., 2012)
Land/Sea Data Sets -> GISS/GISTEMP (NASA) -> operates at 2x2 resolution and has 99% data so requires less interpolation (Pidcock, 2015; Cowtan and Way, 2014)
MSLOT (NOAA) and Japan Meterological Agency -> different resolution of grid boxes = similar trends though (Pidcock, 2015)
World Meteorological Organisation and Global Climate Observation System
land stations -> smooth the data and share it through the WMO (Morice et al., 2021)
5th IPCC Report -> 0.85 increase in ocean temp and land surface (1880-2021) -> small differences between data sets = reproducibility
clear obvious trend of increasing
1960-1970 -> aerosol emissions led to a plateau
SST interannual/interdecadal variability
ENSO -> 2022 temp drop due to a La Niña -> 1997/98 temp rise due to El Niño (Gulev et al., 2021 - 6th IPCC Report)
Arctic Amplification
poles are warming 2x quicker -> ice-albedo feedback -> melting ice has an albedo of 0.8-0.9 while polar water has one of 0.1 (Simmonds, 2015)
SST records -> Southern Ocean and Atlantic Ocean
massive heat store -> facilitated by AMOC heat transfers (Cheng et al., 2017)
SSTs -> 1,200 drifting buoys and 4,000 ships
1.5million observations monthly
Corrections -> inhomogeneities and urbanisation account for <0.05 (Folland and Parker, 1995)
wooden buckets -> 0.42 correction as water evaporated from buckets and leaked out (Hartmann et al., 2013) -> engine room warmed the buckets but HadCRUT4 used an error model to amend this (Morice et al., 2021)
Corrections -> instrumental data
historical SST -> 0.1 estimated error in dating at the start of the dating -> exacerbated by differences in ground/air observations
Lower troposphere = warming
greenhouse effect enhanced so less heat released to stratosphere
tropospheric ozone increased 30-70% (1970-2010)
tropopause increased in height from 1980 (Gulev et al., 2021)
Asymmetry in tropical upper atmosphere -> warmed 1k (2010-2020)
more SH warming than NH -> due to land asymmetry and differences in ozone concentrations (Ladstädter et al., 2023).
stratosphere = cooling
decrease in ozone = less UV
Montreal Protocol 2013 -> Antarctic ozone hole so called to reduce HCFC production -> more recovery in the NH (Gulev et al., 2021)
radiosondes -> measure atmospheric depth at certain temperatures (Hartmann et al., 2013)
ozonesondes -> measure ozone (Gulev et al., 2021)
Global Navigation Satellite System -> measures the earth’s vertical profile
uses satellites and produces data to a high-resolution (Ladstädter et al., 2023)
atmospheric interannual variability
Quasi-Biennial Oscillation (Ladstädter et al., 2023)
Corrections - instrumental differernces e.g. satellites vs weather balloons (would underestimate temp change and would pop) (Ladstädter et al., 2023).
there was a slowdown in rates of warming in the troposphere -> CMIP5 models were unable to simulate this due to residual error/uncertainties -> highlights complexities (Santer et al., 2017)
sub-surface ocean temps recording started in the 1960s
argofloats used -> 4,000 in 2018 now 3,887 in 2022 (Cheng et al., 2017)
sub-surface ocean temp gaps -> if earth is split into 1x1 -> in 1960 <10% coverage, 2003 > 20% and 2015 <30% (Chen et al., 2016)
warming has occurred for upper-700m from 1871 and occured for 700m-2000m from 1971 (Gulev et al., 2021)
PERSIANN-CDR (NCEI) -> 1982 to present day using IR and Microwave radiation on a 2.5x2.5 resolution -> not sufficient to show regional change
Rain Sphere -> finer resolution and more local observations using IR on geostationary satellites (Nguyen et al., 2018)
Global Rain Gauge -> shares data
but limited global coverage - only sports field worth of area is monitored for precipitation (Kidd, 2017)
Clausius Claperyon -> air can hold more water vapour as it warms (7% increase in temp) = more precipitation (Gulev et al., 2021)
trend complicated however -> often easier to split places into regions to examine -> consider variability e.g. NAO, ENSO = noise
Precipitation increase = N.A, Tropical Africa, C.A, Maritime Continent, part of Eur
precipitation decrease = S.A., west N.A. north Africa, Middle East (Gulev et al., 2021)
5th IPCC Report -> precipitation over the midlatitudes =
increase from 1901 (Hartmann et al., 2013)
corrections precipitation -> big uncertainty between model projections = different variables and parameterisation (simplifies processes)
(Hartmann et al., 2013)
Satellites used to determine precipitation rates -> based on idea that clouds cause preciptation
midlatitudes examines for eddies, tropics examined for deep, convective clouds -> satellites use IR to look at cloud top temperature to inform precipitation rates – but it is harder to get satellite data for the midlatitudes than the tropics as there is deeper convection in the tropics -> can also underestimate precipitation as warmer rain occurs from mountain induced uplift but would not be detected by cloud top temperature.
Sea Level Rise = satellite monitoring via LIDAR altimeters (time it takes for the satellite to reach the surface and back)
e.g. Jason-2 and Poseidon/TOPEX (Nerem et al., 2018)
3mm/year rise in global sea levels from 1933 = contribution of melting glaciers not thermal expansion (Nerem et al., 2018) -> global uniformity.
The 5th IPCC report -> sea levels increased 0.19m from 1901-2010, 1.7mm 1991-1993 and 3.2mm 1993-2010 (Rhein et al., 2013).
SLR -> collected from the 1700s
1800s = tide gauges
1900s = satellites (Rhein et al., 2013)
SLR variability
episodic variation -> Mt Pinatubo 1991 -> drop in sea levels and led to the TOPEX satellite being launched and impacted the start of the data set (Nerem et al., 2018)
SLR correction
tide gauges -> implicated by variability as they are local -> while SLR signal during an ENSO needs eof analysis to make it more accurate (Nerem et al., 2018)
problems with data = noise makes depicting the signal more difficult = corrections are required
calibrations, methods shifting, inhomogeneities etc… complicates the noise to signal ratio
GRACE mission -> LIDAR/RADAR = ice thickness (Nerem et al., 2018)
Operation IceBridge -> more sophisticated tech to analyse ice sheet thickness -> lidar altimeter, snow camera etc (Lindsay and Schweiger, 2015).
ice extent -> submarines and sonar or electromagnetic sensors on helicopters
more limitations due to seasonal variability and limitations in recording data (Lindsay and Schweiger, 2015).
Sea ice extent NH = clear cumulative loss in ice, SH = trend harder to discern
West Antarctica e.g Thwaites Glacier = thinning, East Antarctica = thickening event despite CMIP modelling (Simmonds, 2015)
AR5 highlighted that the Arctic loss around 1.3 to 2.3m of thickness between 1980 and 2008 indicating clear thinning (Gulav et al., 2021).
Antarctic sea ice change has been different, there has been a decrease in the volume of sea ice though this decrease was only by a small amount, with a loss of 1.2-1.8% from 1979 to 2012 -> entire regional change is small for Antarctica due to some regions thinning and others thickening (Gulav et al., 2021).
glacial extent -> global decrease in mass from the 1970s and most in retreat (Gulev et al., 2021)
rates can be variable -> variability e.g. Hindu Kush-Karokoram-Himalaya region with low thinning around Karakoram -> due to the Asian Monsoon and westerly atmospheric circulation pattern (Kaab et al., 2021)
ice extent corrections
need to account for the different techniques being drawn upon e.g. snow influences ULS (upward looking sonar) instruments but not Air-EM (plane-based electromagnetic waves) (Lindsay and Schweiger, 2015).
