skip to main content


Title: Hot Air, Hot Lakes, or Both? Exploring Mid‐Holocene African Temperatures Using Proxy System Modeling
Abstract

Climate models predict Africa will warm by up to 5°C in the coming century, stressing African societies. To provide independent constraints on model predictions, this study compares two notable reconstructions of East African temperatures to those predicted by Paleoclimate Model Intercomparison Project (PMIP3) and transient TraCE (Transient Climate Evolution) simulations, focusing on the Mid‐Holocene (MH, 5–8 kyr B.P.). Reconstructions of tropical African temperature derived from lake sedimentary archives indicate 1–2.5°C of warming during the MH relative to the 20th century, but most climate models do not replicate the warming observed in these paleoclimate data. We investigate this discrepancy using a new lake proxy system model, with attention to the (potentially non‐stationary) relationship between lake temperature and air temperature. We find amplified lake surface temperature changes compared to air temperature during the MH due to heightened seasonality and precessional forcing. Lacustrine processes account for some of the warming, and highlight how the lake heat budget leads to a rectification of the seasonal cycle; however, the simulated lake heating bias is insufficient to reconcile the full discrepancy between the models and the proxy‐derived MH warming. We find further evidence of changes in mixing depth over time, potentially driven by changes in cloud cover and shortwave radiative fluxes penetrating the lake surface. This may confound interpretation for glycerol dialkyl glycerol tetraethers (GDGT) compounds which exist in the mixed layer, and suggests a need for independent constraints on mixed layer depth. This work provides a new interpretive framework for invaluable paleoclimate records of temperature changes over the African continent.

 
more » « less
Award ID(s):
1903345 1903348
NSF-PAR ID:
10414194
Author(s) / Creator(s):
 ;  ;  ;  
Publisher / Repository:
DOI PREFIX: 10.1029
Date Published:
Journal Name:
Journal of Geophysical Research: Atmospheres
Volume:
126
Issue:
10
ISSN:
2169-897X
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract

    Quantitative temperature reconstructions from lacustrine organic geochemical proxies including branched glycerol dialkyl glycerol tetraethers (brGDGTs) and alkenones provide key constraints on past continental climates. However, estimation of air temperatures from proxies can be impacted by non‐stationarity in the relationships between seasonal air and water temperatures, a factor not yet examined in strongly seasonal high‐latitude settings. We pair downcore analyses of brGDGTs and alkenones measured on the same samples through the Holocene with forward‐modeled proxy values based on thermodynamic lake model simulations for a western Greenland lake. The measured brGDGT distributions suggest that stable autochthonous (aquatic) production overpowers allochthonous inputs for most samples, justifying the use of the lake model to interpret temperature‐driven changes. Conventional calibration of alkenones (detected only after 5.5 thousand years BP) suggests substantially larger temperature variations than conventional calibration of brGDGTs. Comparison of proxy measurements to forward‐modeled values suggests variations in brGDGT distributions monotonically reflect multi‐decadal summer air temperatures changes, although the length of the ice‐free season dampens the influence of air temperatures on water temperatures. Drivers of alkenone variability remain less clear; potential influences include small changes in the seasonality of proxy production or biases toward specific years, both underlain by non‐linearity in water‐air temperature sensitivity during relevant seasonal windows. We demonstrate that implied temperature variability can differ substantially between proxies because of differences in air‐water temperature sensitivity during windows of proxy synthesis without necessitating threshold behavior in the lake or local climate, and recommend that future studies incorporate lake modeling to constrain this uncertainty.

     
    more » « less
  2. Abstract

    During the mid‐Holocene (MH: ∼6,000 years Before Present) and Last Interglacial LIG (LIG: ∼129,000–116,000 years Before Present) differences in the seasonal and latitudinal distribution of insolation drove Northern Hemisphere high‐latitude warming comparable to that projected for the end of the 21st century in low emissions scenarios. Paleoclimate proxy records point to distinct but regionally variable hydroclimatic changes during these past warm intervals. However, model simulations have generally disagreed on North American regional moisture patterns during the MH and LIG. To investigate how closely the latest generation of models associated with the Paleoclimate Model Intercomparison Project (PMIP4) reproduces proxy‐inferred moisture patterns during recent warm periods, we compare hydroclimate output from 17 PMIP4 models with newly updated compilations of moisture‐sensitive North American proxy records during the MH and LIG. Agreement is lower for the MH, with models producing wet anomalies across the western United States (US) where most proxies indicate increased aridity relative to the preindustrial period. The models that agree most closely with the LIG proxy compilation display relative wetness in the eastern US and Alaska, and dryness in the northwest and central US. An assessment of atmospheric dynamics using an ensemble of the three LIG simulations that best agree with the proxies suggests that weaker winter North Pacific pressure gradients and steeper summer North Pacific and Atlantic gradients drive LIG precipitation patterns. Our updated compilations and proxy‐model comparisons offer a tool for benchmarking climate models and their performance in simulating climate states that are warmer than present.

     
    more » « less
  3. Abstract

    Reconstructing past climates remains a difficult task because pre‐instrumental observational networks are composed of geographically sparse and noisy paleoclimate proxy records that require statistical techniques to inform complete climate fields. Traditionally, instrumental or climate model statistical relationships are used to spread information from proxy measurements to other locations and to other climate variables. Here ensembles drawn from single climate models and from combinations of multiple climate models are used to reconstruct temperature variability over the last millennium in idealized experiments. We find that reconstructions derived from multi‐model ensembles produce lower error than reconstructions from single‐model ensembles when reconstructing independent model and instrumental data. Specifically, we find the largest decreases in error over regions far from proxy locations that are often associated with large uncertainties in model physics, such as mid‐ and high‐latitude ocean and sea‐ice regions. Furthermore, we find that multi‐model ensemble reconstructions outperform single‐model reconstructions that use covariance localization. We propose that multi‐model ensembles could be used to improve paleoclimate reconstructions in time periods beyond the last millennium and for climate variables other than air temperature, such as drought metrics or sea ice variables.

     
    more » « less
  4. Waitt RB, Thackray GD (Ed.)
    Mountain glacier moraine sequences and their chronologies allow us to evaluate the timing and climate conditions that underpin changes in the equilibrium line altitudes (ELAs), which can provide valuable information on the paleoclimatology of understudied regions such as tropical East Africa. However, moraine sequences are inherently discontinuous, and the precise climate conditions that they represent can be ambiguous due to the sensitivity of mountain glaciers to temperature, precipitation, and other environmental variables. Here, we used a two-dimensional (2-D) iceflow and mass-balance model to simulate glacier extents and ELAs in the Rwenzori Mountains in East Africa over the past 31,000 yr (31 k.y.), including the Last Glacial Maximum (LGM), late glacial period, and the Holocene Epoch. We drove the glacier model with two independent, continuous temperature reconstructions to simulate possible glacier length changes through time. Model input paleoclimate values came from branched glycerol dialkyl glycerol tetraether (brGDGT) temperature reconstructions from alpine lakes on Mount Kenya for the last ~31 k.y., and precipitation reconstructions for the LGM came from various East African locations. We then compared the simulated fluctuations with the positions and ages (where known) of the Rwenzori moraines. The simulated glacier extents reached within 1.1 km of the dated LGM moraines in one valley (93% of the full LGM extent) when forced by the brGDGT temperature reconstructions (maximum cooling of 6.1 °C) and a decrease in precipitation (-10% than modern amounts). These simulations suggest that the Rwenzori glaciers required a cooling of at least 6.1 °C to reach the dated LGM moraines. Based on the model output, we predict an age of 12–11 ka for moraines located halfway between the LGM and modern glacier extents. We also predict ice-free conditions in the Rwenzori Mountains for most of the early to middle Holocene, followed by a late Holocene glacier readvance within the last 2000 yr. 
    more » « less
  5. Abstract. Reconstructions of past temperature and precipitation are fundamental to modeling the Greenland Ice Sheet and assessing its sensitivity to climate. Paleoclimate information is sourced from proxy records and climate-model simulations; however, the former are spatially incomplete while the latter are sensitive to model dynamics and boundary conditions. Efforts to combine these sources of information to reconstruct spatial patterns of Greenland climate over glacial–interglacial cycles have been limited by assumptions of fixed spatial patterns and a restricted use of proxy data. We avoid these limitations by using paleoclimate data assimilation to create independent reconstructions of mean-annual temperature and precipitation for the last 20 000 years. Our method uses oxygen isotope ratios of ice and accumulation rates from long ice-core records and extends this information to all locations across Greenland using spatial relationships derived from a transient climate-model simulation. Standard evaluation metrics for this method show that our results capture climate at locations without ice-core records. Our results differ from previous work in the reconstructed spatial pattern of temperature change during abrupt climate transitions; this indicates a need for additional proxy data and additional transient climate-model simulations. We investigate the relationship between precipitation and temperature, finding that it is frequency dependent and spatially variable, suggesting that thermodynamic scaling methods commonly used in ice-sheet modeling are overly simplistic. Our results demonstrate that paleoclimate data assimilation is a useful tool for reconstructing the spatial and temporal patterns of past climate on timescales relevant to ice sheets. 
    more » « less