ALBERT

Description: In recent years, much progress has been made in observing CO2 distributions from space. However, the use of these observations to infer source/sink distributions in inversion studies continues to be complicated by difficulty in quantifying atmospheric transport model errors. We will present results from several different experiments designed to quantify different aspects of transport error using the Goddard Earth Observing System, Version 5 (GEOS-5) Atmospheric General Circulation Model (AGCM). In the first set of experiments, an ensemble of simulations is constructed using perturbations to parameters in the model s moist physics and turbulence parameterizations that control sub-grid scale transport of trace gases. Analysis of the ensemble spread and scales of temporal and spatial variability among the simulations allows insight into how parameterized, small-scale transport processes influence simulated CO2 distributions. In the second set of experiments, atmospheric tracers representing model error are constructed using observation minus analysis statistics from NASA's Modern-Era Retrospective Analysis for Research and Applications (MERRA). The goal of these simulations is to understand how errors in large scale dynamics are distributed, and how they propagate in space and time, affecting trace gas distributions. These simulations will also be compared to results from NASA's Carbon Monitoring System Flux Pilot Project that quantified the impact of uncertainty in satellite constrained CO2 flux estimates on atmospheric mixing ratios to assess the major factors governing uncertainty in global and regional trace gas distributions.

Description: Convective transport is one of the dominant factors in determining the composition of the troposphere. It is the main mechanism for lofting constituents from near-surface source regions to the middle and upper troposphere, where they can subsequently be advected over large distances. Gases reaching the upper troposphere can also be injected through the tropopause and play a subsequent role in the lower stratospheric ozone balance. Convection codes in climate models remain a great source of uncertainty for both the energy balance of the general circulation and the transport of constituents. This study uses the Goddard Earth Observing System Chemistry-Climate Model (GEOS CCM) to perform a controlled experiment that isolates the impact of convective transport of constituents from the direct changes on the atmospheric energy balance. Two multi-year simulations are conducted. In the first, the thermodynamic variable, moisture, and all trace gases are transported using the multi-plume Relaxed-Arakawa-Schubert (RAS) convective parameterization. In the second simulation, RAS impacts the thermodynamic energy and moisture in this standard manner, but all other constituents are transported differently. The accumulated convective mass fluxes (including entrainment and detrainment) computed at each time step of the GCM are used with a diffusive (bulk) algorithm for the vertical transport, which above all is less efficient at transporting constituents from the lower to the upper troposphere. Initial results show the expected differences in vertical structure of trace gases such as carbon monoxide, but also show differences in lower stratospheric ozone, in a region where it can potentially impact the climate state of the model. This work will investigate in more detail the impact of convective transport changes by comparing the two simulations over many years (1996-2010), focusing on comparisons with observed constituent distributions and similarities and differences of patterns of inter-annual variability caused by the convective transport algorithm. In particular, the impact on lower stratospheric composition will be isolated and the subsequent feedbacks of ozone on the climate forcing and tropopause structure will be assessed.

Description: Sub-grid transport, by convection and turbulence, is known to play an important role in lofting pollutants from their source regions. Consequently, the long-range transport and climatology of simulated atmospheric composition are impacted. This study uses the Goddard Earth Observing System, Version 5 (GEOS-5) atmospheric model to study pollutant transport. The baseline model uses a Relaxed Arakawa-Schubert (RAS) scheme that represents convection through a sequence of linearly entraining cloud plumes characterized by unique detrainment levels. Thermodynamics, moisture and trace gases are transported in the same manner. Various approximate forms of trace-gas transport are implemented, in which the box-averaged cloud mass fluxes from RAS are used with different numerical approaches. Substantial impacts on forward-model simulations of CO (using a linearized chemistry) are evident. In particular, some aspects of simulations using a diffusive form of sub-grid transport bear more resemblance to space-biased CO observations than do the baseline simulations with RAS transport. Implications for transport in the real atmosphere will be discussed. Another issue of importance is that many adjoint/inversion computations use simplified representations of sub-grid transport that may be inconsistent with the forward models: implications will be discussed. Finally, simulations using a complex chemistry model in GEOS-5 (in place of the linearized CO model) are underway: noteworthy results from this simulation will be mentioned.

Description: The Houtman Abrolhos Islands, situated at the western passive margin of the Australian continent, consist of a series of shelf-edge coral reefs. The central platforms of the reefs are Late Pleistocene in age and are generally some 3–5 m above present sea level. The uppermost part of the Last Interglacial reefs normally has an upward-shallowing sequence, consisting of coral framestone, coralline algal bindstone and skeletal grainstone to rudstone. This sequence represents deposition in water depths of less than 2 m, and provides a good indicator of sea level. High-precision mass-spectrometric dates of corals from the Abrolhos reefs, including dates obtained from drill cores, arological, isotopic and stratigraphic criteria are established for the selection of suitable samples for dating and for assessing the reliability of dates. Using the screened dates and the stratigraphic evidence, the timing and character of the sea level variations of the Last Interglacial in the Abrolhos region are examined. The data show that sea level of the Last Interglacial in the Abrolhos was 4 m below its present height by ca. 134 ka BP and probably reached about 2 m above present height at ca. 133 ka BP. The exact time at which sea level reached its peak (6 m above present sea level) cannot be determined from our data. But it is clear that the sea level high stand of the Last Interglacial lasted until ca. 116 ka BP and that for much of the Last Interglacial sea level at the Abrolhos was at a height of about 4 m above its present level

Description: Ningaloo Reef, situated on the central west coast, is Australia's largest fringing reef system extending southward from 22°S for over 200 km. Its narrow lagoon is backed by a coastal plain, which is largely composed of an emergent Last Interglacial reef on the flank of folded Tertiary limestones. The west-facing reef is exposed to strong oceanic swells across a narrow (8 km) continental shelf. Climatic aridity, cyclones, tsunamis, and the poleward flowing Leeuwin Current all influence the reef system. Seismic profiling and a coring and dating program along a transect through a reef pass indicate two periods of reef development in the northern part of the reef: Holocene and Last Interglacial. Seaward of the crest, the Holocene reef forms either a prominent 500 m-wide bulge with 10 m of relief and an abrupt seaward slope, or a series of discrete patch reefs. Holocene reef development is limited to depths of less than 30 m and reaches a maximum thickness of ca. 10–15 m below the reef crest. U/Th TIMS dates from distal parts of the Last Interglacial section between −18 and −36 m give ages toward the end of the high stand (120–115 ka). Last Interglacial reef growth was more extensive of the two, filling much of the available accommodation space, perhaps as a result of a stronger Leeuwin Current. This substrate subsequently provided an antecedent foundation for Holocene reef development.

Description: The Houtman Abrolhos reefs, situated on the western continental margin of Australia, occupy a transitional position between cool-water shelf carbonate sediments to the south and more tropical environments to the north. Their existence at the outer limits of the geographical range for coral reef growth is a result of the warm, poleward-flowing Leeuwin Current. Though the modern reefs differ ecologically from tropical reefs, their geological characteristics have been little known until recently. Each of the three island groups in the Abrolhos consists of a central platform of Last Interglacial reefs, about which windward and leeward Holocene reefs have developed asymmetrically. In the Easter Group the subtidal windward reef in the west is ca. 10 m thick and is backed by a leeward-prograding, lagoon sand sheet which is 0–3 m thick. The emergent parts of the leeward reefs in the east consist of an upward-shallowing sequence comprising reef facies, peritidal rudstone facies, and coral rubble storm ridges. This is underlain by over 26 m of Holocene reef facies. Coring and dating of the Holocene reefs (using both TIMS and 14C methods) in the Easter Group has shown significantly different lithofacies in the windward and leeward reefs, and has allowed reconstruction of Holocene reef growth and sea-level history. Coralline algal bindstones and interbedded coral framestone facies characterise the relatively slow-growing windward Holocene reefs, whereas the fast-growing leeward reefs consist of coral framestone facies which are dominated by Acropora. The leeward reefs commenced growth about 10,000 years ago and the Morley reef grew to 0.3 m above present sea level by 6400 years B.P., recording a relative high sea-level event. This generated Holocene constructional topography characterised by “blue-hole” terrain. Windward Holocene reef growth commenced after 8200 years B.P. following erosion of the windward part of the Last Interglacial platform. High wave energy and competition with macroalgae limited coral growth, and the coralline algal-dominated windward reefs grew more slowly to sea level. The Holocene sea-level record provided by dates from the 26 m core of the Morley reef (a “keep-up” reef) is the first such record from the western continental margin of Australia.

Description: U-series ages measured by thermal ionisation mass spectrometry (TIMS) are reported for a Last Interglacial (LI) fossil coral core from the Turtle Bay, Houtman Abrolhos islands, western Australia. The core is 33.4 m long the top of which is approximately 5 m a.p.s.l. (above present sea level). From the232Th concentrations and the reliability of the U-series ages, two sections in the core can be distinguished. Calculated U/Th ages in core section I (3.3 m a.p.s.l to 11 m b.p.s.l) vary between 124±1.7 ka BP (3.3 m a.p.s.l.) and 132.5±1.8 ka (4 m b.p.s.l., i.e. below present sea level), and those of section II (11–23 m b.p.s.l.) between 140±3 and 214±5 ka BP, respectively. The ages of core section I are in almost perfect chronological order, whereas for section II no clear age-depth relationship of the samples can be recognised. Further assessments based on the ϖ234U(T) criteria reveal that none of the samples of core section II give reliable ages, whereas for core section I several samples can be considered to be moderately reliable within 2 ka. The data of the Turtle Bay core complement and extend our previous work from the Houtman Abrolhos showing that the sea level reached a height of approximately 4 m b.p.s.l at approximately 134 ka BP and a sea level highstand of at least 3.3 m a.p.s.l. at approximately 124 ka BP. Sea level dropped below its present position at approximately 116 ka BP. Although the new data are in general accord with the Milankovitch theory of climate change, a detailed comparison reveals considerable differences between the Holocene and LI sea level rise as monitored relative to the Houtman Abrolhos islands. These observation apparently add further evidence to the growing set of data that the LI sea level rise started earlier than recognised by SPECMAP chronology. A reconciliation of these contradictionary observations following the line of arguments presented by Crowley (1994) are discussed with respect to the Milankovitch theory.

Description: U/Th (TIMS) and 14C (AMS) measurements are presented from two coral cores from the Easter group of the Houtman Abrolhos Islands between 28°S and 29°S on the western continental margin of Australia. The U/Th measurements on the Morley core from Morley Island cover a depth interval from 0.2 m above present sea level to 24.4 m below present sea level and comprise eleven samples. The ages vary between 6320 ± 50 a, at 0.2 m above sea level, and 9809 ± 95 a, at 24.4 m below sea level (all errors are 2σ). The mean growth rate is 7.1 ± 0.9 m/ka. The 14C dates of selected Morley core corals show that the 14C ages are ∼ 1000 a younger than their corresponding U/Th ages, which agrees with previous results. The main purpose of our 14C measurements is to be able to compare them precisely with other coral cores where no U/Th measurements are available. The U/Th measurements of the Suomi core from Suomi Island cover a depth interval from 0.05 m to 14.2 m below present sea level and consist of four samples. The ages vary between 4671 ± 40 a, at 0.05 m below sea level, and 7102 ± 82 a, at 14.2 m below sea level, with a mean growth rate of 5.8 ± 0.2 m/ka. The growth history of both cores is explained by a simple model in which the growth rates of the Morley core can be interpreted as reflecting local rates of sea level rise, whereas the Suomi core is interpreted as reflecting lateral growth during the past ∼ 6000 a. Our results indicate that sea level relative to the western margins of the Australian continent was about 24 m lower than present at about 9800 a B.P. (14C gives a date of 8500 a B.P.). Sea level then rose and reached a highstand, slightly higher than the present position at about 6300 a B.P (14C date: 5500 a). This highstand declined but was still higher than present at 4600 a B.P. This is in agreement with previous observations along the Australian coastal margins and with observations from the Huon peninsula (Papua New Guinea). Our results are very similar to theoretical numerical models, which take into consideration water loading and isostatic compensation and viscous mantle flow. In contrast, coral cores from Barbados show that corals with a 14C age of ∼ 5500 a B.P. are some ∼ 10 m b.p.s.l. We interpret the difference between the Barbados core and the Morley core as resulting from additional “flooding” of Barbados by water redistribution, due to changes in the Earth's geoid but not reflecting global sea level rise or major addition of melt waters over the past ∼ 6000 a. The difference in the geoid at Barbados between ∼ 6000 a B.P. and the present will require a refinement in the geophysical models. Precise230Th (TIMS) measurements on continental coasts will be required to provide an adequate data base for modelling deformation, flow of mantle material and sea-level height

Description: A core from a coral colony of Porites lutea was analysed for stable oxygen isotopic composition*. A 200-year proxy record of sea surface temperatures from the Houtman Abrolhos Islands off west Australia was obtained from coral δ18O. At 29′S, the Houtman Abrolhos are the southernmost major reef complex of the Indian Ocean. They are located on the path of the Leeuwin Current, a southward flow of warm, tropical water, which is coupled to Indonesian throughflow. Coral δ18O primarily reflects local oceanographic and climatic variability, which is largely determined by spatial variability of the Leeuwin Current. However, coherence between coral δ18O and the current strength itself is relatively weak. Evolutionary spectral and singular spectrum analyses of coral δ18O demonstrate a high variability in spectral composition through time. Oscillations in the 5–7-y, 14–15-y, and quasi-biennial bands reflect teleconnections of local sea surface temperature (SST) to tropical Pacific climate variability. Deviations between local (coral-based) and regional (instrument) SST contain a cyclic component with a period of 15 y. Coral δ18O suggests a rise in SST by 0.6 ′C since AD 1944, consistent with available instrumental SST records. A long-term warming by 1.4 ′C since AD 1795 is inferred from the coral record

Description: Wheat, rice, maize, and soybean provide two-thirds of human caloric intake. Assessing the impact of global temperature increase on production of these crops is therefore critical to maintaining global food supply, but different studies have yielded different results. Here, we investigated the impacts of temperature on yields of the four crops by compiling extensive published results from four analytical methods: global grid-based and local point-based models, statistical regressions, and field-warming experiments. Results from the different methods consistently showed negative temperature impacts on crop yield at the global scale, generally underpinned by similar impacts at country and site scales. Without CO2 fertilization, effective adaptation, and genetic improvement, each degree-Celsius increase in global mean temperature would, on average, reduce global yields of wheat by 6.0%, rice by 3.2%, maize by 7.4%, and soybean by 3.1%. Results are highly heterogeneous across crops and geographical areas, with some positive impact estimates. Multimethod analyses improved the confidence in assessments of future climate impacts on global major crops and suggest crop- and region-specific adaptation strategies to ensure food security for an increasing world population.