Assessing the feasibility of using a neural network to filter Orbiting Carbon Observatory 2 (OCO-2) retrievals at northern high latitudes

Satellite retrievals of XCO2 at northern high latitudes currently have sparser coverage and lower data quality than most other regions of the world. We use a neural network (NN) to filter Orbiting Carbon Observatory 2 (OCO-2) B10 bias-corrected XCO2 retrievals and compare the quality of the filtered data to the quality of the data filtered with the standard B10 quality control filter. To assess the performance of the NN filter, we use Total Carbon Column Observing Network (TCCON) data at selected northern high latitude sites as a truth proxy. We found that the NN filter decreases the overall bias by 0.25 ppm (∼ 50 %), improves the precision by 0.18 ppm (∼ 12 %), and increases the throughput by 16 % at these sites when compared to the standard B10 quality control filter. Most of the increased throughput was due to an increase in throughput during the spring, fall, and winter seasons. There was a decrease in throughput during the summer, but as a result the bias and precision were improved during the summer months. The main drawback of using the NN filter is that it lets through fewer retrievals at the highest-latitude Arctic TCCON sites compared to the B10 quality control filter, but the lower throughput improves the bias and precision.

Spatial distributions of <i>X</i>_CO₂ seasonal cycle amplitude and phase over northern high-latitude regions

Satellite-based observations of atmospheric carbon dioxide (CO2) provide measurements in remote regions, such as the biologically sensitive but undersampled northern high latitudes, and are progressing toward true global data coverage. Recent improvements in satellite retrievals of total column-averaged dry air mole fractions of CO2 (XCO2) from the NASA Orbiting Carbon Observatory 2 (OCO-2) have allowed for unprecedented data coverage of northern high-latitude regions, while maintaining acceptable accuracy and consistency relative to ground-based observations, and finally providing sufficient data in spring and autumn for analysis of satellite-observed XCO2 seasonal cycles across a majority of terrestrial northern high-latitude regions. Here, we present an analysis of XCO2 seasonal cycles calculated from OCO-2 data for temperate, boreal, and tundra regions, subdivided into 5∘ latitude by 20∘ longitude zones. We quantify the seasonal cycle amplitudes (SCAs) and the annual half drawdown day (HDD). OCO-2 SCAs are in good agreement with ground-based observations at five high-latitude sites, and OCO-2 SCAs show very close agreement with SCAs calculated for model estimates of XCO2 from the Copernicus Atmosphere Monitoring Services (CAMS) global inversion-optimized greenhouse gas flux model v19r1 and the CarbonTracker2019 model (CT2019B). Model estimates of XCO2 from the GEOS-Chem CO2 simulation version 12.7.2 with underlying biospheric fluxes from CarbonTracker2019 (GC-CT2019) yield SCAs of larger magnitude and spread over a larger range than those from CAMS, CT2019B, or OCO-2; however, GC-CT2019 SCAs still exhibit a very similar spatial distribution across northern high-latitude regions to that from CAMS, CT2019B, and OCO-2. Zones in the Asian boreal forest were found to have exceptionally large SCA and early HDD, and both OCO-2 data and model estimates yield a distinct longitudinal gradient of increasing SCA from west to east across the Eurasian continent. In northern high-latitude regions, spanning latitudes from 47 to 72∘ N, longitudinal gradients in both SCA and HDD are at least as pronounced as latitudinal gradients, suggesting a role for global atmospheric transport patterns in defining spatial distributions of XCO2 seasonality across these regions. GEOS-Chem surface contact tracers show that the largest XCO2 SCAs occur in areas with the greatest contact with land surfaces, integrated over 15–30 d. The correlation of XCO2 SCA with these land surface contact tracers is stronger than the correlation of XCO2 SCA with the SCA of CO2 fluxes or the total annual CO2 flux within each 5∘ latitude by 20∘ longitude zone. This indicates that accumulation of terrestrial CO2 flux during atmospheric transport is a major driver of regional variations in XCO2 SCA.

Source and variability of formaldehyde (HCHO) at northern high latitude: an integrated satellite, ground/aircraft, and model study

Here we use satellite observations of HCHO vertical column densities (VCD) from the TROPOspheric Monitoring Instrument (TROPOMI), ground-based and aircraft measurements, combined with a nested regional chemical transport model (GEOS-Chem at 0.5° × 0.625° resolution), to understand the variability and sources of summertime HCHO better in Alaska. We first evaluate GEOS-Chem with in-situ airborne measurements during Atmospheric Tomography Mission 1 (ATom-1) aircraft campaign and ground-based measurements from Multi-AXis Differential Optical Absorption Spectroscopy (MAX-DOAS). We show reasonable agreement between observed and modeled HCHO, isoprene and monoterpenes. In particular, HCHO profiles show spatial homogeneity in Alaska, suggesting a minor contribution of biogenic emissions to HCHO VCD. We further examine the TROPOMI HCHO product in Alaska during boreal summer, which is in good agreement with GEOS-Chem model results. We find that HCHO VCDs are dominated by free-tropospheric background in wildfire-free regions. During the summer of 2018, the model suggests that the background HCHO column, resulting from methane oxidation, contributes to 66–80 % of the HCHO VCD, while wildfires contribute to 14 % and biogenic VOC contributes to 5–9 % respectively. For the summer of 2019, which had intense wildfires, the model suggests that wildfires contribute to 40 to 65 %, and the background column accounts for 30 to 50 % of HCHO VCD in June and July. In particular, the model indicates a major contribution of wildfires from direct emissions of HCHO, instead of secondary production of HCHO from oxidation of larger VOCs. We find that the column contributed by biogenic VOC is often small and below the TROPOMI detection limit. The source and variability of HCHO VCD above Alaska during summer is mainly driven by background methane oxidation and wildfires. This work discusses challenges for quantifying HCHO and its precursors in remote pristine regions.

Holocene regional climate change and formation of southern Ontario's largest swamp inferred from a kettle-lake pollen record

Greenock Swamp wetland complex is one of few remaining natural wetlands in the Great Lakes region and, at 89 km2 in areal extent, is currently the largest hardwood swamp in southern Ontario, Canada. We present here pollen and sediment records from a kettle hole (Schmidt Lake) and adjacent Thuja occidentalis swamp to reconstruct regional paleoclimate and vegetation history, and to assess the timing and development of the swamp ecosystem and associated carbon stocks. Pollen-inferred paleoclimate reconstructions show the expected warming in the Early Holocene, and indicate the Mid-Holocene initiation of lake-effect snow. This enhanced snowfall may have maintained high water tables in the adjacent wetland since ca. 8300 years ago, promoting the establishment of a swamp dominated by Thuja occidentalis. Carbon accumulation rates in a >2-m-long peat core collected from a Thuja occidentalis stand adjacent to Schmidt Lake are 30–40 g C/m2/yr, which is higher than the average of northern high-latitude peatlands. Using topographic and hydrological parameters, we estimated that mean swamp peat thicknesses could exceed 2 m. Thus, this study encourages future investigations on temperate swamps from the perspective of hitherto underestimated Holocene carbon sinks and shows the importance of regional hydroclimate in supporting swamp ecosystems.

Earth Ice Age Dynamics: A Bimodal Forcing Hypothesis

The Earth has gone through multiple ice ages in the past million years. Understanding the ice age dynamics is crucial to paleoclimatic study, and is helpful for addressing future climate challenges. Though ice ages are paced by variations in Earth’s orbit geometry, how various climatic system components on the Earth respond to insolation forcing and interact with each other remains unclear. A prevailing view argues that the initial responses occur in the northern high latitudes (i.e. the northern high-latitude hypothesis, NHH). This opinion is challenged by recent reports, such as the lead of climate change in the Southern Hemisphere (SH) relative to that in the Northern Hemisphere (NH), the southern control on Atlantic meridional overturning circulations (AMOC), and the potential significance of Southern Hemisphere (SH). Alternatively, the tropical hypothesis (TH) argues for a leading role of the tropics. Both the NHH and the TH belong to a single-forcing mechanism, and have difficulty in interpreting phenomena, such as the saw-tooth pattern of the ice ages. Here we present a new proposal concerning the Earth’s ice age dynamics: the bimodal forcing hypothesis (BFH). The essential assumption of this hypothesis is that for glacial-interglacial cycles, the cooling (glaciation) starts from the northern high latitudes, whereas the warming (deglaciations) starts from the SH. Particularly, the BFH emphasizes the significance of SH oceans in accumulating and transferring heat for deglaciations. Thus, it is capable to reasonably explain the saw-tooth pattern. We compiled 100 paleotemperature records globally for validation. The BFH is consistent with most of these records, and provides a straightforward and comprehensible way to interpret ice age on Earth.

Timing and structure of early-Holocene climate anomalies inferred from north Chinese stalagmite records

In this paper, a new decadal resolution stalagmite δ18O record covering 10.4–6.5 ka BP from Kulishu cave in Beijing, north China is presented in combination with the published stalagmite δ18O record covering 10.4–14.0 ka BP in the same cave. Five significant monsoon collapses were identified around 11.5, 11.0, 10.0, 9.4, and 8.2 ka BP as well as three smaller ones around 10.3, 9.0, and 8.6 ka BP. The weak monsoon episodes around 8.6 and 8.2 ka BP form the two-step structure of the 8.2 ka event. All monsoon collapses, coeval with the cooling in northern high-latitude records, are correlated with Lakes Agassiz-Ojibway outbursts. Thus, our data support the idea of freshwater forcing of abrupt climate anomalies during the early Holocene. Nevertheless, the decreased irradiance together with freshwater outburst may account for the 9.2/9.3 ka event, which is expressed more significantly in low-latitude records.

Assessing the Feasibility of Using a Neural Network to Filter OCO-2 Retrievals at Northern High Latitudes

Satellite retrievals of XCO2 at northern high latitudes currently have sparser coverage and lower data quality than most other regions of the world. We use a neural network (NN) to filter OCO-2 B10 bias-corrected XCO2 retrievals and compare the quality of the filtered data to the quality of the data filtered with the standard B10 quality control filter. To assess the performance of the NN filter, we use Total Carbon Column Observing Network (TCCON) data at selected northern high latitude sites as a truth proxy. We found that the NN filter decreases the overall bias by 0.25 ppm (~50 %), improves the precision by 0.18 ppm (~12 %), and increases the throughput by 16 % at these sites when compared to the standard B10 quality control filter. Most of the increased throughput was due to an increase in throughput during the spring, fall, and winter seasons. There was a decrease in throughput during the summer, but as a result the bias and precision were improved during the summer months. The main drawback of using the NN filter is that it lets through fewer retrievals at the highest latitude Arctic TCCON sites compared to the B10 quality control filter, but the lower throughput improves the bias and precision.

Past terrestrial hydroclimate driven by Earth System Feedbacks

Geologic evidence suggests drastic reorganizations of subtropical terrestrial hydroclimate during past warm intervals, including the mid-Piacenzian Warm Period (MP, 3.3 to 3.0 Ma). Despite having a similar to present-day atmospheric CO2 level (pCO2), MP featured moist subtropical conditions with high lake levels in Northern Africa, and mesic vegetation and sedimentary facies in subtropical Eurasia. Here, we demonstrate that major loss of the northern high-latitude ice sheets and continental greening, not the pCO2 forcing, are key to generating moist terrestrial conditions in subtropical Sahel and east Asia. In contrast to previous hypotheses, the moist conditions simulated in both regions are a product of enhanced tropospheric humidity and a stationary wave response to the surface warming pattern, both varying strongly in response to land cover changes. These results suggest that past terrestrial hydroclimate states were driven by Earth System Feedbacks, which may outweigh the direct effect of pCO2 forcing.