Global temperature deviations as a random walk
- Estonian Academy of Sciences, Toravere (Estonia). Inst. of Astrophysics and Atmospheric Physics
Surface air temperature is the main parameter to represent the earth`s contemporary climate. Several historical temperature records on a global/monthly basis are available. Time-series analysis shows that they can be modelled via autoregressive moving average models closely connected to the classical random walk model. Fitted models emphasize a nonstationary character of the global/monthly temperature deviation from a certain level. The nonstationarity explains all trends and periods, found in the last century`s variability of global mean temperature. This means that the short-term temperature trends are inevitable and may have little in common with a currently increasing carbon dioxide amount. The calculations show that a reasonable understanding of the contemporary global mean climate is attainable, assuming random forcing to the climate system and treating temperature deviation as a response to it. The forcings occur due to volcanic eruptions, redistribution of cloudiness, variations in snow and ice covered areas, changes in solar output, etc. Their impact can not be directly estimated from changes of the earth`s radiation budget at the top of the atmosphere, because actual measurements represent mixture of the forcings and responses. Thus, it is impossible empirically to separate the impact of one particular forcing (e.g., that due to increase of CO{sub 2} amount) from the sequence of all existing forcings in the earth climate system. More accurate modelling involving main feedback loops is necessary to ease such a separation.
- OSTI ID:
- 470993
- Report Number(s):
- CONF-960420-; ISBN 0-884736-02-5; TRN: IM9722%%116
- Resource Relation:
- Conference: 7. global warming international conference and exposition, Vienna (Austria), 1-3 Apr 1996; Other Information: PBD: 1996; Related Information: Is Part Of The 7. global warming international conference and expo: Abstracts; PB: 154 p.
- Country of Publication:
- United States
- Language:
- English
Similar Records
STITCHES: creating new scenarios of climate model output by stitching together pieces of existing simulations
Global, Hemispheric, and Zonal Temperature Deviations Derived From a 63-Station Radiosonde Network