Welcome to my research page! You can follow the links below (or scroll through this page) to see some summaries of current and past research projects!
Overview
What I do and why I do it!
Atmospheric Coupling
(and how it modifies the answers you’d get from land-only models)
Coupling matters!
Land Surface Impacts on the Atmospheric Radiative Balance
Using radiative kernels to tease apart direct and indirect impacts of land use on shortwave and longwave radiation.
Simple Land Interface Model
Simple models are fun! They’re also quite useful.
Global Effects of Terrestrial Evaporation
(Lessons from Northland)
Maybe it warms. Maybe it cools. Which response “wins” depends on a few things…
Mid-latitude Forest Cover
Trees can trigger cloud feedbacks; that’s pretty neat.
Collaborations
Including:
Exoplanets ∙ Ecoclimate Teleconnections ∙ Deforestation ∙ Plant Response to CO2 ∙ Arctic forest expansion & clouds
There are links to the journal-formatted articles and to open access pre-print pdfs of manuscripts associated with each of these projects below.
Overview
Climate plays an important role in the behavior of the terrestrial portion of the Earth System. Less frequently discussed, however, is the idea that changes in the land surface itself can influence the atmosphere and climate. My work is focused on understanding how changes in the land surface can drive responses in the atmosphere, at both regional and global scales.
In particular, my work aims to identify where the atmosphere is most sensitive to changes in the land surface, which surface properties have the biggest impact on the atmosphere at any given location, and how changes in surface energy and water fluxes modify atmospheric processes.
I use Earth System Models to study how the land and the atmosphere interact. My work has leveraged the Community Earth System Model (CESM), SLIM (an idealized land model which I have developed to run coupled to the CESM framework), and Isca (an idealized global circulation model). I enjoy thinking about idealized representations of the Earth system, in particular to dig into the basic physical processes involved in land-atmosphere coupling. This includes exploring representations of “Earth” with idealized continental configurations.
I am also interested in regional land-atmosphere coupling. I aim to understand how properties of the land surface modulate physical processes in the atmospheric boundary layer and lower troposphere to modify cloud formation and precipitation in different climate regimes (e.g. tropical vs. boreal areas). My goal is to understand what scale of vegetation change in a given biome is required to drive a local, regional, or large-scale response in the atmosphere.
Atmospheric Coupling
Individual land surface properties modify different aspects of the surface energy budget, and thus have distinct effects on surface-atmosphere coupling. During my PhD, I explored how three individual land surface properties (albedo, evaporative resistance, and aerodynamic roughness) modify the terrestrial surface energy budget and land surface climate both in land-only (“offline”) and land-atmosphere (“coupled”) global Earth system model simulations. Atmospheric coupling leads not only to much stronger temperature (and surface energy flux) changes in response to land surface albedo and evaporative resistance, but also modifies the pattern of which land regions respond most strongly to each land surface property.
Annual mean scaled surface temperature Ts response (K) for (a)–(c) coupled simulations and (d)–(f) offline simulations, per (a),(d) 0.04 darkening of the surface albedo, (b),(e) 50 s m−1 increase in evaporative resistance, and (c),(f) 5.0-m decrease in vegetation height. Violet regions (ΔTs < −0.1) indicate regions where the temperature cooled substantially in response to the prescribed surface change. Stippling indicates regions where the slope is not significantly different from zero (p > 0.05).
Uncoupled (land-only) effects of albedo, evaporation, and roughness:
When the atmosphere is not allowed to respond to changes in the land surface, albedo has the larges impact on surface temperatures in places that are dry and sunny (e.g. the Sahara desert – see figure below). The temperature effect is smaller in areas with low insolation (e.g. the high latitudes) because the same change in albedo leads to a smaller increase in absorbed shortwave than in regions with large amount of sun reaching the surface. Making the land surface darker leads to smaller amounts of warming in areas with high soil moisture than in dry areas, as the extra shortwave energy absorbed can be shed from the land surface through latent heat (evaporation), rather than sensible heat or emitted longwave radiation (which increase with surface temperature).
Increasing the evaporative resistance of the land surface (i.e. making it harder to evaporate water) only decreases evaporation in regions with soil moisture. That is, if you consider a region with a near-zero latent heat flux, then make it harder to evaporate water, it will still have a near-zero latent heat flux. In comparison, if you consider a wet region like a rainforest, making it harder to evaporate water has a substantial impact on surface latent heat fluxes. With all else held equal, decreasing latent heat flux means the surface energy budget must shed that energy through either emitted longwave radiation or through sensible heat flux, both of which correspond to an increase in surface temperature. In a land-only simulation, increasing evaporative resistance leads to the most warming in regions with lots of soil moisture.
Changing the aerodynamic resistance of the land surface modifies how efficiently turbulent fluxes (sensible and latent heat) can exchange heat and moisture between the land surface and the boundary layer.
Coupled effects of albedo, evaporation, and roughness:
Coupling with an interactive atmosphere results in much larger temperature changes in response to a decrease in land albedo, as well as a different pattern in the temperature response. The pattern results from changes in atmospheric energy transport, which drives high-latitude warming (which is then amplified by the loss of snow and sea ice). A large portion of the increased magnitue of warming to a decrease in land albedo can be understood through local longwave feedbacks in the atmosphere. In an offline/land-only simulation, decreasing albedo warms the land surface, but doesn’t change the amount of downwelling longwave radiation from the atmosphere. In a coupled simulation, the same change in albedo initially warms the land surface; the overlying atmosphere warms both from increased sensible heating and by absorbing the increased longwave radiation from the warmer surface. The warmer air then radiates more LW back down to the land surface. Now there is more energy being absorbed by the land surface, which in turn results in a larger warming signal when the system reaches equilibrium.
The pattern and magnitude of surface warming as a result of increased land surface evaporative resistance in coupled simulations is largely the result of changes in cloud cover. In particular, low clouds in the CAM5 atmospheric model are particularly sensitive to terrestrial evaporation in parts of the northern mid-latitudes. While decreasing evaporation from the land surface initially leads to warming (from reduced latent heat flux), in some regions this reduction in evaporation leads to sufficiently large decreases in atmospheric water vapour to cause loss of low cloud cover. This results in more shortwave energy reaching the land surface in regions with cloud loss, which drives the increased magnitude (and specific pattern) of warming in coupled simulations.
Changing the aerodynamic roughness of the land surface results in similar surface temperatures in land-only and coupled simulations. However, the 2m air temperature change driven by a change in land surface roughness differs substantially between coupled and land-only simulations. This is a direct result of the parametrization of turbulent energy exchange in land surface models; in land-only models, artificially large air-to-surface temperature gradients are allowed to form, which are mixed out/smoothed in coupled land-atmosphere simulations.
Importance of atmospheric coupling
As shown above, allowing the atmosphere to interact with changes in land surface properties fundamentally alters how a change in the land surface modifies the climate system both locally and on larger scales. While a change in some land surface property can modify surface energy fluxes and surface temperatures in a land-only model, allowing the atmosphere to respond to land surface changes allows for atmospheric feedbacks on the land surface. These feedbacks can be local, such as a change in atmospheric air temperature or cloud cover above a region where the land surface is modified. However, these feedbacks can also be global or far removed from the initial change in the land surface – that is, a change in the land surface in one region can modify atmospheric circulation, advection, etc. and impact surface climate in regions far-removed from the original land surface perturbation. Moreover, in particular when considering changes in land surface roughness, caution is required when using offline (not coupled to an atmosphere) land surface models to consider the effects of vegetation change on surface climate, independent of any complications generated by local or remote atmospheric feedbacks.
Three types of land atmosphere interactions: (a) the direct, local response of the surface to the atmosphere (with no feedbacks); (b) local atmospheric feedbacks, where changes in the atmosphere above a modified land surface occur because of the modified land surface below that atmospheric column; and (c) remote atmospheric feedbacks, where a change in land at location 1 drives a large-scale atmospheric response that can in turn impact the land at location 2. Examples of each feedback consider the impact of a change in albedo α on absorbed shortwave energy SWabs, sensible heat flux SH, cloud cover, downward shortwave energy at the surface SWdown, downward longwave energy at the surface LWdown, and surface temperature Ts.
You can read more about this topic here:
Laguë, M. M., G. B. Bonan, and A. L. S. Swann, 2019: Separating the Impact of Individual Land Surface Properties on the Terrestrial Surface Energy Budget in both the Coupled and Uncoupled Land–Atmosphere System. J. Climate, 32, 5725–5744, https://doi.org/10.1175/JCLI-D-18-0812.1.
Radiative Feedbacks on Land Surface Change and Associated Tropical Precipitation Shifts
Changes in the different physical aspects of the land surface alter both the surface and atmospheric energy budget through different pathways. Understanding how the land surface impacts the atmospheric energy budget allows us to determine how land surface properties are modifying large-scale atmospheric circulation.
Changes in albedo lead to a direct increase in absorbed shortwave radiation at the surface, and thus directly alter the net flux of energy into the base of the atmospheric column. This increase in absorbed shortwave has a direct effect on the atmospheric column energy balance. In addition, changes to the atmosphere driven by the initial terrestrial albedo change — like changes in air temperatures, snow and ice extent, atmospheric water vapour, and clouds — go on to further modify the column and top of atmosphere energy budgets.
Not all land surface properties directly modify the total amount of energy absorbed by the land surface. With all else held equal, changes in terrestrial evaporation lead to re-partitioning of sensible heat, latent heat, and emitted longwave radiation at the base of the column, but don’t directly result in a net change in the total amount of energy fluxed between the land surface and the atmosphere. However, changes in terrestrial evaporation can modify atmospheric water vapour, temperatures, and clouds, in turn modifying snow and ice extent. Each of these responses drive their own change in the atmospheric column energy budget.Using atmospheric radiative kernels, we can decompose the effect of a given change in the land surface on both shortwave and longwave radiation in the atmospheric column into the components driven:
- directly by the land surface change
- in response to the land surface change by
- changes in snow and ice
- changes in atmospheric water vapour
- changes in surface and atmospheric air temperatures
- changes in clouds
We can then leverage the relationship between the annual mean atmospheric column energy balance, cross-equatorial atmospheric energy transport, and the zonal mean location of the Intertropical Convergence Zone (ITCZ) to attribute tropical precipitation shifts to individual aspects of the atmospheric response to land surface change (see figure 12 in Laguë et al. 2021b). Making land darker and making it harder for land to evaporate water both lead to northwards shifts in tropical precipitation as a result of a hemispheric energy imbalance, but the drivers of this energy imbalance differ. The direct increase in absorbed shortwave radiation at the surface drives the hemispheric energy imbalance resulting from making land darker, while the shortwave effects of cloud responses drive the response in the case of terrestrial evaporation.
You can read more about this topic here, or free on EarthArXiv:
M. M. Laguë, A. L. S. Swann, William R. Boos. Radiative feedbacks on land surface change and associated tropical precipitation shifts. Journal of Climate, 34, 16 (2021): 6651-6672 (2021). EarthArXiv Preprint. DOI: 10.1175/JCLI-D-20-0883.1 .
SLIM: The Simple Land Interface Model
The Simple Land Interface Model (SLIM) is an idealized land surface model that couples to the Community Earth System Model (CESM) in place of the Community Land Model (CLM).
Changes in vegetation encompass simultaneous changes in many different physical properties of the land surface, both in the real world and in complex land surface models. This is one of the many reasons studying land-atmosphere coupling is interesting and exciting, but it also means that when the atmosphere responds in an interesting way to a change in vegetation, it is often difficult if not impossible to identify which particular aspect of the vegetation change is responsible for the atmospheric response.
Using and idealized model within a complex ESM is useful as it allows the user to isolate changes in specific land surface properties. We can compare the atmospheric response to a change in vegetation using a complex land surface model, then see how that same atmosphere responds to the individual or combined changes in the land surface associated with that vegetation change (e.g. the changes in albedo, aerodynamic roughness, evaporative resistance etc.).
The source code for SLIM is available on github, here:
https://github.com/marysa/SimpleLand
Instructions for setting up a CESM-SLIM simulation are available on the github wiki, here:
https://github.com/marysa/SimpleLand/wiki
A description of the SLIM model physics is available in the Supplemental Information of Laguë et al. (2019), here:
https://journals.ametsoc.org/jcli/article/32/18/5725/344078/Separating-the-Impact-of-Individual-Land-Surface
Northland: Competing warming and cooling effects of suppressed terrestrial evaporation
With all else held equal, reducing evaporation from the land surface leads to warming of the land surface (in regions with non-zero soil moisture), as the energy which used to be allocated to latent heat flux (evaporation) instead must be removed from the land surface as sensible heat or emitted longwave radiation – both of which correspond to an increase in surface temperature. This warming is locally isolated to the part of the land where evaporation is reduced. However, decreasing evaporation from the land surface also reduces atmospheric water vapour. Atmospheric water vapour is a strong greenhouse gas, thus decreasing atmospheric water vapour has a cooling effect on the land surface. This cooling effect can occur over large spatial scales, as the atmosphere can mix water vapour depleted air away from the region of reduced evaporation.
When evaporation is suppressed over sufficiently large land areas – such as a continent covering the entire Northern Hemisphere of a planet – the water vapour cooling effect can dominate any local warming from suppressed evaporation. We have identified a trade-off in the local warming vs. global cooling effect of suppressing terrestrial evaporation using a series of global circulation model simulations with various idealized continental configurations. We find that the water vapour cooling effect is larger when the total land area is larger, but also that, for the same total land area, the water vapour cooling effect is larger for larger contiguous continents (vs. continents broken up with oceans). The suppressing terrestrial evaporation with the modern continental configuration of Earth would lie towards the left of this figure, as there is ample ocean at every latitude.
Schematic showing the possible surface temperature response to suppressed terrestrial evaporation for a variety of NH continental configurations. Land area generally increases from left to right, though for a given total land area, larger continents sit further to the right on the curve than smaller, more numerous continents. Qualitative locations of suppressing terrestrial evaporation on TwoPatchLand, NorthWestLand, ThreePatchLand, ThreeQuarterLand, and Northland are shown by the maps of temperature change for each continental configuration, with the annual mean change in land surface temperature noted on each map.
Read more here:
M. M. Laguë, M. Pietschnig, S. Ragen, T. Smith, D. S. Battisti. Terrestrial evaporation and global climate: lessons from Northland, a planet with a hemispheric continent. In Press. Journal of Climate, 2020. DOI: 10.1175/JCLI-D-20-0452.1. EarthArXiv Preprint.
Mid-latitude Afforestation
Forests modify the surface energy budget by modifying the physical properties of the land surface. Trees tend to be darker than grasslands, have deeper roots and higher leaf areas, and tend to be more aerodynamically rough. In the high latitudes, the albedo effect of forests dominates, such that replacing grasslands with forests generally leads to warming. In the tropics, the evaporative effect of forests tends to dominate, such that increasing forest cover generally leads to cooling by increasing evapotranspiration from the land surface. You can read more about this in a great 2008 review article by Gordon Bonan.
In the mid-latitudes, it isn’t always clear what aspect of increased forest cover is going to dominate the surface energy budget. As part of my MSc work with Dr. Abigail Swann, we explored the effects of increasing forest area in the Northern Hemisphere mid-latitudes. In our modeling study, we found that initially increased forest cover resulted in increased evapotranspiration from the mid-latitude land surface, as the forests had deeper roots, larger leaf areas, and lower albedos (i.e. absorbed more incident shortwave energy) than the grasslands they replaced. However, as total forest area continued to increase, the total latent heat flux from the mid-latitude land surface plateaued, as soil moisture became a limiting factor for evapotranspiration. Once latent heat fluxes plateaued, continuing to increase forest cover resulted in large increases in surface temperature, as latent cooling of the land surface could no longer offset the increased absorbed shortwave radiation resulting from lower land surface albedos. This trade-off between increased latent heat flux and surface temperatures (which impact sensible heat flux and emitted longwave radiation) results in threshold responses in boundary layer moisture and low cloud cover.
The change in the average outgoing terms of the surface energy budget over land area between 30N and 60N for the each simulation with latent heat flux (left solid bar), the net flux of longwave radiation at the surface (middle checkered bar), and sensible heat (right striped bar) (W/m2). Error bars show 95% confidence bounds.
In addition to locally impacting surface energy fluxes and temperatures in the mid-latitudes, incrementally increasing the area of mid-latitude forest cover also impacts global-scale atmospheric circulation. In particular, reduced land surface albedo from increased forest cover — in addition to loss of mid-latitude cloud cover which allows more shortwave radiation to reach the surface — drives a hemispheric energy imbalance, with more energy being absorbed by the Northern Hemisphere as northern mid-latitude forest cover increases. While there are threshold responses in the partitioning of the mid-latitude terrestrial surface energy budget, we found that the large-scale atmospheric circulation responds to the total amount of energy being added to the mid-latitude atmosphere, rather than to the specific partitioning of the mid-latitude surface energy budget.
Read more here:
M. M. Laguë, and A. L. S. Swann. Progressive Mid-latitude Afforestation: Impacts on Clouds, Global Energy Transport, and Precipitation. Journal of Climate, 29(15):5561?-5573, 2016. DOI: 10.1175/JCLI-D-15-0748.1 https://doi.org/10.1175/JCLI-D-15-0748.1.
Collaborations
In addition to my primary areas of research, I am involved in several active interdisciplinary collaborations exploring the role of land in the coupled climate system. Previous collaborations include:
- Exoplanets
- Tele-coupling between far-ranging ecosystems
- Theoretical deforestation
- Temperature changes from the physiological response of CO2
- Arctic cloud responses to increased high-latitude forest cover
~
Exoplanet Albedo
The Trappist-1 system includes several planets orbiting an ultra-cool red dwarf star. Using idealized climate model simulations, we explored the effects of albedo on the climate of land-only exoplanets orbiting this star, which has a very different radiative spectrum than Earth’s Sun.
You can read more about this here:
A. J. Rushby, A. L. Shields, E. T. Wolf, M. M. Laguë, A. Burgasser. The Effect of Land-Albedo Feedback on the Climate of Land-Dominated Planets in the TRAPPIST-1 System. The Astrophysical Journal, 2020. DOI: 10.3847/1538-4357/abbe04. ArXiv Preprint.
~
Teleconnections & Telecoupling
Changes in vegetation modify physical properties of and processes on the land surface. This drives changes in the exchange of energy and water between the land and the atmosphere. Changes in the atmosphere in response to changes in land surface fluxes can occur both locally (thus feeding back on the initial vegetation change) or remotely (by triggering atmospheric teleconnections). These remote atmospheric changes can then feed back on the vegetated land surface in regions far removed from the initial vegetation change – this is termed an “ecoclimate teleconnection”.
Ecosystems can interact across scales not only through atmospheric teleconnections — changes in atmospheric circulation in response to land surface change — but also through various other telecoupling mechanisms. This includes, but is not limited to, the advection of nutrients (e.g. evia dust) by the atmosphere, transport of nutrients and organisms through stream systems, and transport of nutrients and organisms by other organisms (e.g. migratory birds and mammals).
You can read more about this here:
A. L. S. Swann, M. M. Laguë, E. S. Garcia, J. P. Field, D. D. Breshears, D. J. P. Moore, S. R. Saleska, S. C. Stark, J. C. Villegas, D. J. Law, and D. M. Minor. Continental-scale consequences of tree die-offs in North America: identifying where forest loss matters most. Environmental Research Letters, 13(5): 55014, 2018. DOI: 10.1088/1748-9326/aaba0f. http://iopscience.iop.org/article/10.1088/1748-9326/aaba0f/meta
F. Tromboni, J. Liu, E. Ziaco, D. Breshears, K. Thompson, W. Dodds, K. Dahlin, E. La Rue, J. Thorp, Andrés Vinã, M. M. Laguë, A. Maasri, H. Yang, S. Chandra, S. Fei. Macrosystems as metacoupled human and natural systems. In Press. Frontiers in Ecology. 2020.
~
LUMIP: Idealized Deforestation
The Land Use Model Intercomparison Project (LUMIP) includes a set of simulations exploring the effects of idealized deforestation, where forest cover is incrementally removed from all currently forested regions of the globe. This has implications for both the physical water/energy coupling of the land surface to the rest of the Earth System, and for terrestrial carbon storage.
Deforestation is shown to generally warm in the tropics an cool in the higher latitudes, though the latitude at which the temperature response changes sign ranges from 11 to 43 degrees latitude across models. While the biogeophysical effects of large-scale deforestation generate a cooling effect, globally averaged, the warming effect from the resulting loss in carbon dominates the total temperature response.
Read more here:
L. Boysen, V. Brovkin, J. Pongratz, D. Lawrence, P. Lawrence, N. Vuichard, P. Peylin, S. Liddicoat, T. Hajima, Y. Zhang, M. Rocher, C. Delire, R. Stéférian, V. Arora, L. Nieradzik, P. Anthoni, W. Thiery, M. M. Laguë, D. Lawrence, M. Lo. Global climate response to idealized deforestation in CMIP6 models. Biogeosciences, 2020. DOI: 10.5194/bg-2020-229.
~
Physiological Response to CO2: Impact on Surface Temperatures
Increased atmospheric carbon dioxide leads to warming through CO2‘s radiative effects. However, vegetation also responds to increased atmospheric CO2 – in particular, at highCO2 plants can photosynthesize the same amount with more closed stomata (pores on the leaves that regulate gas exchange). Open stomata take up CO2 and release water. Most climate models project that plants will close their stomata due to increased atmospheric CO2 under future climate projections. This reduces transpiration – latent heat flux – from the land surface, and thus leads to warming.
This warming – the warming driven by the closure of plants’ stomata, which is independent of the warming driven by the radiative effects of CO2 – can be quantified in terms of climate sensitivity, and indeed is actually already implicitly included in the calculation of climate sensitivity of CMIP5 and CMIP5 models. The physiological response to CO2 accounts for an average of 6% (with a range of 1.4%–13.9%) of the total transient climate response (TCR), and contributes disproportionately to the inter-model spread in TCR.
Read more here:
C. M. Zarakas, A. L. S. Swann, M. M. Laguë, K. C. Armour, J. T. Randeron. Plant Physiology Increases the Magnitude and Spread of the Transient Climate Response in CMIP6 Earth System Models. Journal of Climate, 2020. DOI: 10.1175/JCLI-D-20-0078.1. EarthArXiv Preprint.
~
Arctic Cloud Response to Forest Expansion
Northward expansion of forests in the Arcitc under climate change will modify the physical properties of the land surface. However, different types of trees modify the land surface in different ways. For example, needleleaf trees are darker (have a lower albedo) than broadleaf trees, but they also tend to have a higher resistance to evaporation.
Using SLIM, we considered the effect of an idealized increase in Arctic forest cover of 4 tree types: a needleaf tree and a broadleaf tree, and two “hybrid” trees – a “dark broadleaf” (with the albedo of a needleleaf tree and the evaporative resistance of a broadleaf tree) and a “bright needleleaf” (with the albedo of a broadleaf tree and the evaporative resistance of a neeldleaf tree).
Needleaf trees led to the most warming, while broadleaf trees led to the least. However, the two hybrid tree types led to comparable amounts of warming, but for very different reasons. The dark broadleaf tree warmed as a result of albedo, while the bright needleleaf tree warmed as a result of changes in cloud cover which, in our simulations, happened to be of the right magnitude to generate comparable magnitude to the albedo change! This study highlights the importance of understanding not only how potential Arctic forest expansion will impact the Arctic land surface, but also the importance of understanding how the Arctic atmosphere will respond to changes in high-latitude vegetation.
Read more here:
J. E. Kim, M. M. Laguë, S. Pennypacker, E. Dawson, and A. L. S. Swann. Evaporative Resistance is of Equal Importance as Surface Albedo in High Latitude Surface Temperatures Due to Cloud Feedbacks. Geophysical Research Letters, 47(4), 2020. e2019GL085663. DOI: 10.1029/2019GL085663. EarthArXiv Preprint.
~
Marysa Lague 