The coverage uncertainty estimate has also been upgraded to capture some of the seasonal cycle in the uncertainty.
Not exact matches
One interesting feature of the HadCRUT4 paper is the use of the NCEP / NCAR reanalysis data to obtain an
estimate for the
coverage uncertainty.
The remaining nine events (i.e., the 8 severe storm events and wildfire) have lower potential
uncertainty surrounding their
estimate due to more complete insurance
coverage and data availability.
A better approach would be to use a window of months about the current month to obtain a time - dependent
estimate of both the bias and the
uncertainty due lack of
coverage.
In 2017, seven of the sixteen billion - dollar events (i.e., the 2 inland flooding events, drought, freeze and hurricanes Harvey, Irma and Maria) have higher potential
uncertainty values around the loss
estimates due to less
coverage of insured assets and data latency.
Lyman and colleagues combined different ocean monitoring groups» data sets, taking into account different sources of bias and
uncertainty — due to researchers using different instruments, the lack of instrument
coverage in the ocean, and different ways of analyzing data used among research groups — and put forth a warming rate
estimate for the upper ocean that it is more useful in climate models.
Given that there is greater
uncertainty associated with the HadCRUT data prior to 1900 due to fewer stations and sparser global
coverage, and that the TCR constrained by 1901 - 2000 data better matches the IPCC central TCR
estimates, their higher TCR (approximately 1.7 to 2.5 °C) seems more likely to be correct.
The hunch was that the
uncertainty estimates are too small in the nineteenth century and suggested that this was due to incomplete
coverage uncertainties being underplayed.
UHA is also an
estimate using different instruments,
estimating different figures, with different
coverage, and also carries with it different
uncertainties.
My main critique of their
uncertainty analysis is the issue of structural
uncertainty and specifically their
estimate of the
uncertainty associated with incomplete
coverage.
Spatial sampling
uncertainties were
estimated by simulating poorly sampled periods (e.g. 1753 to 1850) with modern data (1960 to 2010) for which the Earth
coverage was better than 97 % complete, and measuring the departure from the full site average when using only the limited spatial regions available at early times.
The station data
coverage in this region is poor for the E-OBS, which contributes to a relatively large
uncertainty in precipitation and temperature
estimates for this region in the E-OBS dataset.
Many historical in situ marine data still remain to be digitised and incorporated into the database, to improve
coverage and reduce the
uncertainties in our
estimates of marine climatic variations.
It should be noted that their
estimate of
uncertainty relates mostly to the spatial sampling, or sparse data
coverage.