6 - Glacial Processes Flashcards
(275 cards)
3 ways of measuring glaciers
glaciological, geodetic, gravimetric
Three approaches to calculating past
contributions of glaciers & ice caps to
sea level
Statistical area-weighted
extrapolations of locally- or
regionally-derived glaciological &
geodetic data over most glaciers on
earth
Geodetic approach only, using repeat
DEMs from ASTER satellite
Gravimetric approach using GRACE
data
DEM stands for
digital elevation model
4 essential ingredients of a model
- A spatial domain to run your model. Do you want to model conditions at
a single point? Or investigate two-dimensional spatial patterns? - A Mathematical description of the system – either empirical (i.e., derived
from data), or based on physical principles (e.g., Newton’s Laws). - Suitable inputs. For glacier mass balance, we need meteorological data
(e.g., precipitation & temperature) to calculate accumulation & ablation.
Real-world observations to calibrate &
validate the mode
3 main options for spatial domain of glacier model
0 dimensional applications that treat glaciers as a single bulk entity
1 dimensional applications that split glaciers into a series of bulk elevation bands
2 dimensional applications that are fully spatially distributed
equation for SMB
Bn = c + a + R
where c is accumulation, a is ablation (defined negatively), R is refreezing
three main types of glacier mass balance model
- Degree Day (also known as temperature index) models
- Energy Balance (also known as physically-based) models
- Hybrid (also known as enhanced degree day / temperature index) models
+ves and -ives of degree day model
Advantages
- Quick & easy
- Only requires air temperature as input - easy to apply in data sparse,
remote regions - Good for predicting melt in future if future air temperatures modelled
well
Disadvantages
* Model is empirical - extrapolating over space & time is uncertain
* Not physically-based - offers limited insight into processes
* Limited information on spatial patterns across glacier, or temporal
information at sub-seasonal scales.
+ves and -ves of energy balance model
Advantages
- Physically-based – offers good insight into the processes occurring
- Good for investigating spatial patterns, and short, sub-daily, time periods
- Does not depend on empirical relations, so easy to apply to other
glaciers & different time periods
Disadvantages
- Computationally expensive
- Requires many meteorological inputs so application often limited to
locations with lots of observations - Arguably less useful for future predictions where some of the climate
drivers, e.g. cloudiness, humidity, wind speeds are less well predicted
than, e.g. air temperature
Advantages & Disadvantages of Hybrid Models
- Account for spatial patterns
- Variant 1 requires no extra field measurements; Variant 2 requires Gs measurements
- Depends on empirical relations, so not easy to apply to other glaciers & time periods
- More computationally expensive than classic DD method but less than EB approach
- Variant 1 could be useful for future predictions
outline the maths behind c (accumulation)
Precipitation usually treated as
linear function of elevation
* Air temperature threshold (e.g. 1oC)
used to distinguish between snow or
rain
* Works well on a glacier-average
scale over whole summers
* Works less well at smaller spatial
and temporal scales
* Does not account for local
topographic features or snow
redistribution by wind.
outline the traditional approach to monitoring meteorological inputs to glaciers
Automatic weather station (AWS) on
or near glacier.
* Meteorological variables then
extrapolated over the DEM.
* e.g. temperature assuming a
standard atmospheric lapse rate (6.5
˚C per km).
* e.g. precipitation assuming
regionally measured gradient.
* BUT… most glaciers are in remote
locations & so it is often difficult to get
local measurements.
outline the new approach ~(last 20 years ) to monitoring meteorological inputs to glaciers
Climate reanalysis (e.g. ERA-40, ERA
Interim, ERA 20C, JRA55, NOAA 20CR).
* Produced by ‘reanalysing’ observations
using a weather/climate model.
* Global fields of meteorological variables
(e.g., temperature, precipitation) on a
moderate resolution grid (typically 1
degree lat./long.). So we can model any
glacier!
* BUT… reanalyses are produced using
models so are subject to uncertainty (e.g.
biases)
outline some parameters that must be estimated that are used in a model
- DDFs for snow and ice (for a DD model) * snow and ice albedo (for an EB model) * temperature lapse rate
- precipitation gradient
- threshold temperature for rain / snow
what is calibration
we adjust uncertain model parameters so model output agrees
well with real-world observations
what is validation
we test our calibrated model to see how well it performs against real-world observations.
what do we do if we do not have have measurements of parameters
we have to
calibrate (or ‘tune’ or ‘optimize’) them.
what does DD model stand for
degree day
what is DDF stand for
degree day factor
outline the work of Orleans and Fortune 1992
- Applied energy balance model to 12 glaciers around the world
- Examined sensitivity to a 1 ̊C increase in air temperature.
Key conclusion:
Maritime glaciers are more sensitive to air temperature changes than continental glaciers..
why are Maritime glaciers are more sensitive to air temperature changes than continental glaciers..
relationship between temperature & melt is exponential. So perturbing temperature has greater impact in warmer maritime regions.
- More precipitation will fall as rain in warmer maritime regions compared to sub- zero continental regions
- A positive feedback loop: melting lowers albedo, which increases melting
what is perturbing temperature
term used to describe a departure from the regular flow of atmospheric currents
outline hock et al 2007
compared 5 models of varying complexity ( 3 DD and 2 EB).
Applies to Storglaciaren, calibrated using ERA 40 reanalysis
used regional climate model output to predict MB up to 2100