Our Common Future Under Climate Change

International Scientific Conference 7-10 JULY 2015 Paris, France

Menu
  • Home
  • Zoom Interactive Programme
Cliquer pour fermer

Tuesday 7 July - 17:00-18:30 UPMC Jussieu - ROOM 105 - Block 24/34

2245 - Modelling the complexities of the Earth System

Parallel Session

Chair(s): H. Le Treut (Université Paris Diderot, Paris, France)

17:00

Added value and limitations of earth system models (invited)

J.L. Dufresne, (CNRS, Paris, France)

Abstract details
Added value and limitations of earth system models (invited)
Abstract content
17:20

Mitigation delay sensitivity of temperature, sea level and ocean acidification

P. Pfister (Physics Institute, University of Bern, Bern, Switzerland), T. Stocker (Physics Institute, University of Bern, Bern, Switzerland)

Abstract details
Mitigation delay sensitivity of temperature, sea level and ocean acidification

P. Pfister (1) ; T. Stocker (1)
(1) Physics Institute, University of Bern, Climate and environmental physics, Bern, Switzerland

Abstract content

With annually increasing anthropogenic carbon emissions, the commitment to climate change and its impacts continues to grow. A policy-relevant metric for this growth is the mitigation delay sensitivity (MDS), measuring the additional commitment per decade until emissions start to decrease. For the peak temperature increase ΔT, MDS can be estimated analytically using idealized emission scenarios (Allen and Stocker, 2014; Stocker, 2013). Here, we use an Earth System Model of Intermediate Complexity to evaluate the MDS of ΔT, steric sea level rise (SSLR), and two ocean acidification impacts following Steinacher et al. (2013). We examine the dependance of the MDS on the rate of increase (r) and subsequent decrease (s) in annual emissions, for three different equilibrium climate sensitivities (ECS) of 1.5, 3.0 and 4.5 K. The modeled MDS is in good agreement with analytical estimates for ΔT and SSLR, except for scenarios with very high cumulative emissions. For standard parameters (ECS = 3.0 K, r = 2% and s ranging from 5% down to 0.5%), MDS amounts to 0.3-0.7 K/decade for ΔT and 7-20 cm/decade for SSLR by 3000 AD. With regard to ocean acidification, we find that partial Aragonite undersaturation of the Southern Ocean surface (by 2100 AD) may be avoided with sufficiently early and stringent mitigation. MDS is 15-17% of the Southern Ocean area per decade for standard parameters, but for s>3% no undersaturation occurs if emissions start to decrease before 2030. Further loss of strongly supersaturated ocean surface areas (associated with coral reef habitats) is virtually unavoidable, but can be substantially reduced by mitigation. For standard parameters, we are already committed to a loss of roughly 30-90% of these areas by 2100 AD, even with immediate mitigation (with respect to preindustrial); MDS amounts to an additional loss of 22-66% of the remaining area per decade.

References:

Allen, M. R. & Stocker, T. F. Impact of delay in reducing carbon dioxide emissions. Nature Clim. Change 4, 23--26 (2014).

Stocker, T. The closing door of climate targets. Science 339, 280--282 (2013).

Steinacher, M., Joos, F. & Stocker, T. F. Allowable carbon emissions lowered by multiple climate targets. Nature 499, 197--201 (2013).

17:40

Do ENSO modeling discrepancies affect the climate-carbon feedback?

A. Bastos (Instituto Dom Luiz - Universidade de Lisboa, Lisbon, Portugal), C. Gouveia (Instituto Dom Luiz - Universidade de Lisboa, Lisbon, Portugal), R. Trigo (Instituto Dom Luiz - Universidade de Lisboa, Lisbon, Portugal)

Abstract details
Do ENSO modeling discrepancies affect the climate-carbon feedback?

A. Bastos (1) ; C. Gouveia (1) ; R. Trigo (1)
(1) Instituto Dom Luiz - Universidade de Lisboa, Lisbon, Portugal

Abstract content

The El-Niño/Southern Oscillation controls most of global climate variability on inter-annual time-scales. In particular, ENSO affects temperature and precipitation patterns in South America and Southeast Asia, although its influence extends to remote regions and latitudes.

ENSO is also known to play an important role in the carbon-cycle, influencing a large fraction of variability in atmospheric CO2 growth rate, mainly due to its impact on terrestrial ecosystems in the Southern Hemisphere. During El-Niño events, lower than average precipitation and warmer temperatures are registered in most of northern and central South America, South Africa and most of Australia, leading to reduced CO2 uptake by vegetation, and to increased fire activity. During the cold phases (La-Niña), the patterns are approximately reversed and the land-sink is enhanced.

Although Earth-System Models (ESMs) in CMIP5 do include representation of the ocean-atmosphere coupling processes that produce ENSO, they still present difficulties in modeling some of the most relevant features of the phenomenon, such as the temporal evolution and recurrence periods of warm and cold phases .

It has been shown that the difference between model estimates of future CO2 concentration are related to their biases in representing present atmospheric CO2. Moreover, the larger disagreements of future CO2 uptake are due to uncertainty about the future response of the land-sink to climate change. It has been further shown that part of the uncertainties in future estimates of CO2 storage on terrestrial ecosystems are related to the differences in the climate forcing from the global circulation models (GCMs).

Here we investigate whether the discrepancies between observations and modeled ENSO may explain biases in the modeled present land-sink in CMIP5 ESMs. We evaluate the performance of ESMs in reproducing, for the period 1959-2011, a set of temporal features, such as the intensity, spatial configuration, and spectral properties of ENSO.  Subsequently, the ability of ESMs to reproduce the global land-sink strength and variability was evaluated and compared to the relationship between the modeled ENSO and variability in the modeled land-sink. We find that biases in the land-sink are related to the way ESMs simulate the response of terrestrial ecosystems to ENSO variability and that this, in turn, is mainly dependent on the spectral properties of the modeled ENSO. The ESMs that mis-represent the peak periodicity of ENSO in the 3-7 years band tend to present larger errors in the modeled land-sink characteristics. The nature of this relationship between ENSO discrepancies and land-sink biases was studied by evaluating the influence of discrepancies in ENSO frequency on temperature and precipitation patterns within each model.

The analysis was further extended to the future RCP4.5 and RCP8.5 scenarios, by evaluating the influence of the discrepancies of the present terrestrial CO2 uptake on the future land-sink and the role played by ENSO in these discrepancies. 

17:52

Spread of ocean heat uptake efficiency in CMIP models

J. Gregory (NCAS, University of Reading and Met Office Hadley Centre, Reading, United Kingdom), E. Exarchou (IC3, Barcelona, Spain), T. Kuhlbrodt (NCAS, University of Reading and Met Office Hadley Centre, Reading, United Kingdom)

Abstract details
Spread of ocean heat uptake efficiency in CMIP models

E. Exarchou (1) ; J. Gregory (2) ; T. Kuhlbrodt (2)
(1) IC3, Barcelona, Spain; (2) NCAS, University of Reading and Met Office Hadley Centre, Reading, United Kingdom

Abstract content

During transient climate change, the ocean heat uptake efficiency, quantified by the ratio between the rate of ocean heat uptake N and surface global-mean temperature change DT, can quantitatively measure how effectively the heat is vertically transported in the deeper ocean, thus moderating surface transient change. Among models of the Coupled Model Intercomparison Project (CMIP) framework, the spread in the values of k varies by about a factor of 2. The spread in k affects the model spread in projections of ocean heat uptake and thermal expansion, therefore contributing to the uncertainty in model future projections of transient climate change. We investigate potential sources of model spread in kappa by using a set of 23 CMIP models with CO2 increasing at a rate of 1% per year. In the set of models we analyse, in line with previous studies, the models with the strongest AMOC have the strongest CO2-induced AMOC reduction. We find that the models with the strongest AMOC, and consequently the models with the strongest AMOC reduction, have the strongest ocean heat uptake efficiency. Despite the strong correlation with AMOC,  the largest portion of the ocean heat uptake efficiency and of its spread comes from the Southern Ocean. By further analysing the detailed process-based heat budgets in three models, HadGEM2, GFDL-ESM2M and MPIESM-LR, which have k values from the low, middle and high part of the kappa spread, we find that the spread in these three models mainly arises from changes in advection. The dominant part of the advective ocean heat uptake takes place in the Southern Ocean. We explore potential physical mechanisms that link the Southern Ocean advective heat uptake, hence the Southern Ocean overturning, with the AMOC and the ocean heat uptake efficiency. 

18:04

Forest Mortality, Economics, and Climate

L. Hawkins (Oregon State University, Oregon, United States of America), B. Law (Oregon State University, Oregon, United States of America), P. Mote (Oregon State University, Oregon, United States of America), A. Plantinga (University of California Santa Barbara, California, United States of America), J. Hicke (University of Idaho, Idaho, United States of America), M. Allen (University of Oxford, Oxford, United Kingdom), R. Betts (University of Exeter, Exeter, United Kingdom)

Abstract details
Forest Mortality, Economics, and Climate

L. Hawkins (1) ; B. Law (2) ; P. Mote (3) ; A. Plantinga (4) ; J. Hicke (5) ; M. Allen (6) ; R. Betts (7)
(1) Oregon State University, Earth, Ocean, and Atmospheric Science, Oregon, United States of America; (2) Oregon State University, Forest ecosystems & society, Oregon, United States of America; (3) Oregon State University, Earth, ocean, & atmospheric sciences, Oregon, United States of America; (4) University of California Santa Barbara, Natural resource economics and policy, California, United States of America; (5) University of Idaho, Geography, Idaho, United States of America; (6) University of Oxford, School of geography and the environment, Oxford, United Kingdom; (7) University of Exeter, Geography, Exeter, United Kingdom

Abstract content

Forests in western North America provide critical economic and environmental services including watershed protection, timber production, and carbon sequestration.  Recently this region has experienced widespread forest mortality attributed to droughts, fires, and regional warming.  Forest mortality rates doubled in the Pacific Northwest between 1990 and 2010 with bark beetles killing trees across 11 million hectares in B.C and the western U.S. The complex interactions among climate, ecosystem response, and economic factors govern the impacts these stressors will have on terrestrial ecosystems in the future. To examine how major forests will function under future climatic and land use regimes we couple an improved forest mortality model to NCAR’s Community Land Model (CLM4.5) incorporating a predictive model of mountain pine beetle outbreaks and a forest product economic model.  The economic model estimates the values of private forest investment, wood products, recreation, and water. Annual harvest levels, silvicultural investments, and forest management strategies feedback into CLM. This methodology will elucidate feedbacks among climate, land use, forest ecosystems, carbon sequestration and ecosystem services.  To assess uncertainty in the simulated forest response to climate change a dynamic vegetation model (TRIFFID operating within the Hadley Centre Earth System Model) is used to generate a super-ensemble of simulations using crowd sourced computing (climateprediction.net).  This novel form of evaluating intrinsic variability allows for thorough investigation of model sensitivity to parameterization.  Outcomes of this work will provide policy makers and resource managers with tools for developing adaptive strategies for responding to projected warming, drought, insects, and economic factors.

18:16

Modeling and visualization of the marshes vegetation

A. Sadovski (TexasA&M University-Corpus Christi, Corpus Christi, TX, United States of America), P. Montagna, (Texas A&M University-Corpus Christi, Corpus Christi, United States of America), S. King, (Texas A&M University-Corpus Christi, Corpus Christi, United States of America)

Abstract details
Modeling and visualization of the marshes vegetation

A. Sadovski (1) ; P. Montagna, (2) ; S. King, (3)
(1) TexasA&M University-Corpus Christi, Corpus Christi, TX, United States of America; (2) Texas A&M University-Corpus Christi, Harte research institute for the gulf of mexico, Corpus Christi, United States of America; (3) Texas A&M University-Corpus Christi, School of engineering and computing sciences, Corpus Christi, United States of America

Abstract content

This paper deals with a mathematical models of the vegetation growth in the marshes of the Gulf of Mexico. It is a spatial-temporal model of multi-species competition presented in the form of the system of partial differential equations. Analysis of the behavior of such a system is studied through simulation and visualization tools. The objective of the study is to find the impact of fresh water releases and precipitations on the vegetation cover of marshes so it can be used for future preservation of the marsh ecological system.

Coastal marshes are important ecosystems that provide many benefits to human health and well-being including: protecting the inland areas from storm surge, storing water, removing nutrients that flow in from watersheds, and providing nursery habitat for key commercial and recreational fisheries.  Yet, marshes are under extreme pressure from development and 50% of the marshes nationwide have disappeared since the founding of the United States. 

The Texas coast is flat, hot, and windy; which makes coastal marshes very susceptible to effects of climate change and water resource development.  Climate change can have three effects: sea-level rise, water cycle alterations, and temperature alterations.  The main effects will be to drown marshes during rising water levels or dry them out as evapotranspiration rates increase during droughts.  Water resource development has decreased water delivery to marshes in the Nueces Delta by 45% over the last 40 years, which has led to marsh degradation.

There is a need to understand the dynamics and the interactive roles of climate and water cycle changes in order to predict changes in salt marshes in the future.  This information is critical for resource management.  However, few tools exist to forecast effects of human activities on marsh function.  Results and models of this ongoing research could be extended to coastal estuaries in other regions of the world.