Speaker
Description
In order to estimate the global temperature response and climate sensitivity to radiative forcing, volcanic activity is an important testbed. This work uses a non-parametric method for estimating the temperature response due to volcanic forcing from a simulation of the NorESM model [1]. In addition, this response will be convolved with reconstructed volcanic forcing and compared to temperature recordings. The volcanic forcing and the temperature response data sets are given in [2] and [3] and cover the last two millennia. Most previous attempts of acquiring the response function have been by use of parametric models and parameter fits based on historical data (e.g. [4] and [5]).
Under the assumption that the temperature responds linearly to the forcing, the system is modelled using the filtered Poisson process (FPP) such that the Richardson-Lucy deconvolution algorithm can be applied. The FPP is a methodological model that describes a signal as superposed pulses. The sum of pulses can be written as a convolution as $T=G*F$, where $T$ is the temperature, $F$ is the forcing and $G$ is some response function. Using the Richardson-Lucy deconvolution algorithm we can get back $G$ given $F$ and $T$, that is, we get back the response from knowing the volcanic forcing and the temperature recording [6]–[8]. This approach is thus a non-parametric method of acquiring the response function, and because of the superposition of pulses it is also insensitive to pulse overlap.
Further, the response function obtained may be used within the same framework to predict the temperature evolution when different forcing scenarios are run, for example a doubling of CO2 scenario. This is, of course, still assuming a linear dependence between the forcing and temperature response.
References
- M. Bentsen et al., “The norwegian earth system model, noresm1-m – part 1: Description and basic evaluation of the physical climate,” Geoscientific Model Development, 2013, vol. 6, no. 3, pp. 687–720.
- P. D. Jones and M. E. Mann, “Climate over past millennia,” Reviews of Geophysics, 2004, vol. 42, no. 2, pp. 1–42.
- PAGES 2k Consortium. et al., “Consistent multidecadal variability in global temperature reconstructions and simulations over the common era,” Nature Geoscience, Aug. 2019, vol. 12, no. 8, pp. 643–649.
- G. J. Boer, M. Stowasser, and K. Hamilton, “Inferring climate sensitivity from volcanic events,” Climate Dynamics, Feb. 2007, vol. 28, no. 5, pp. 481–502.
- F. A. Bender, A. M. L. Ekman, and H. Rodhe, “Response to the eruption of Mount Pinatubo in relation to climate sensitivity in the CMIP3 models,” Climate dynamics, Oct. 2010, vol. 35, no. 5, pp. 875–886.
- W. H. Richardson, “Bayesian-Based Iterative Method of Image Restoration*,” J. Opt. Soc. Am., Jan. 1972, vol. 62, no. 1, pp. 55–59.
- L. B. Lucy, “An iterative technique for the rectification of observed distributions,” The Astronomical Journal, 1974, vol. 79, no. 6, p. 745.
- F. Benvenuto, R. Zanella, L. Zanni, and M. Bertero, “Nonnegative least-squares image deblurring: Improved gradient projection approaches,” Inverse Problems, 2010, vol. 26, no. 2.
