Leveraging clustering and geostatistics to improve the modeling of sub-grid land-atmosphere interactions in Earth system models

2021 ◽  
Author(s):  
Nathaniel Chaney ◽  
Laura Torres-Rojas ◽  
Jason Simon
Keyword(s):  
Correlation Length ◽  
Land Surface ◽  
Spatial Organization ◽  
Sensible Heat Flux ◽  
Surface Fluxes ◽  
Surface Model ◽  
Earth System ◽  
Time Step ◽  
System Models ◽  
Earth System Models

<p>Multi-scale spatial heterogeneity over the land surface (meter to km scales) can play a pivotal role in the development of clouds and precipitation. To model this process within Earth system models (ESMs; ~100 km spatial resolution), sub-grid reduced-order modeling approaches are used. More specifically, state-of-the-art ESMs sub-divide the land surface of each grid cell into representative clusters (e.g., forest, lakes, and grasslands) that are learned a-priori from available high-resolution satellite remote sensing data (e.g., STRM, Landsat and Sentinel-2) via clustering. However, until recently, these clusters have remained spatially agnostic making it infeasible to infer spatial statistics of the modeled sub-grid heterogeneity over land that are required by the atmospheric model to ensure proper development of simulated convection (e.g., spatial correlation length of surface evaporation). This presentation will introduce an approach that leverages the precomputed cluster positions in space to construct an effective and efficient approach to assemble the experimental semivariogram from the sub-grid clusters within ESMs. As a proof of concept, we will show results by applying the novel method on sub-grid model output from the HydroBlocks land surface model over a 100 km domain centered at the Southern Great Plains site in Oklahoma, United States. Furthermore, to illustrate the added-value that the experimental semivariograms will have towards improving the modeling of land-atmosphere interactions, we will illustrate the results from large-eddy simulations over the domain that show how differences in correlation length of surface fluxes can have, at times, a dramatic impact on the development of clouds and convection in the atmosphere. When implemented in ESMs, this new approach will make it possible to infer the modeled sub-grid spatial organization of the surface fluxes (e.g., sensible heat flux) per time step with negligible increases in computation expense.</p>

scholarly journals Use of various remote sensing land cover products for PFT mapping over Siberia

2013 ◽  
Vol 6 (1) ◽  
pp. 255-296
Author(s):  
C. Ottlé ◽  
J. Lescure ◽  
F. Maignan ◽  
B. Poulter ◽  
T. Wang ◽  
...  
Keyword(s):  
High Resolution ◽  
Land Cover ◽  
Land Surface ◽  
Land Cover Classification ◽  
Classification Systems ◽  
Tree Cover ◽  
Surface Model ◽  
Earth System ◽  
System Models ◽  
Earth System Models

Abstract. High-latitude ecosystems play an important role in the global carbon cycle and in regulating the climate system and are presently undergoing rapid environmental change. Accurate land cover datasets are required to both document these changes as well as to provide land-surface information for benchmarking and initializing earth system models. Earth system models also require specific land cover classification systems based on plant functional types, rather than species or ecosystems, and so post-processing of existing land cover data is often required. This study compares over Siberia, multiple land cover datasets against one another and with auxiliary data to identify key uncertainties that contribute to variability in Plant Functional Type (PFT) classifications that would introduce errors in earth system modeling. Land cover classification systems from GLC 2000, GlobCover 2005 and 2009, and MODIS collections 5 and 5.1 are first aggregated to a common legend, and then compared to high-resolution land cover classification systems, continuous vegetation fields (MODIS-VCF) and satellite-derived tree heights (to discriminate against sparse, shrub, and forest vegetation). The GlobCover dataset, with a lower threshold for tree cover and taller tree heights and a better spatial resolution, tends to have better distributions of tree cover compared to high-resolution data. It has therefore been chosen to build new PFTs maps for the ORCHIDEE land surface model at 1 km scale. Compared to the original PFT dataset, the new PFT maps based on GlobCover 2005 and an updated cross-walking approach mainly differ in the characterization of forests and degree of tree cover. The partition of grasslands and bare soils now appears more realistic compared with ground-truth data. This new vegetation map provides a framework for further development of new PFTs in the ORCHIDEE model like shrubs, lichens and mosses, to better represent the water and carbon cycles in northern latitudes. Updated land cover datasets are critical for improving and maintaining the relevance of earth system models for assessing climate and human impacts on biogeochemistry and biophysics. The new PFT map at 5 km scale is available for download from the PANGAEA website, at: doi:10.1594/PANGAEA.810709.

Where are the peatlands? Comparing two global peatland maps through L-band brightness temperature simulations

2021 ◽  
Author(s):  
Michel Bechtold ◽  
Sarith P. Mahanama ◽  
Rolf H. Reichle ◽  
Randal D. Koster ◽  
Gabrielle J. M. De Lannoy
Keyword(s):  
Soil Moisture ◽  
Brightness Temperature ◽  
Land Surface ◽  
Land Surface Model ◽  
Open Loop ◽  
Surface Model ◽  
Earth System ◽  
System Models ◽  
L Band ◽  
Earth System Models

<p>Mapping the global peatland distribution is important for embedding peatland processes into Earth System Models. Peatland maps are typically compiled from nation-specific soil or ecosystem maps or based on machine learning tools trained on such data. Here, we evaluate the performance of a land surface model with two different peatland map inputs in providing critical land surface estimates (soil moisture, temperature) to a Radiative Transfer Model (RTM) for L-band brightness temperature (Tb). We hypothesize that an improved performance of the land surface model in Tb space indicates a better spatial peatland distribution input within the footprint of Tb observations (~40 km).</p><p>We employ the NASA Catchment Land Surface Model (CLSM) with a recently added module for peatland hydrology (PEATCLSM modules). We run this model at a 9-km EASEv2 resolution over the Northern Hemisphere for two soil maps that differ in their peatland distributions. The applied soil distributions are: (MAP1) a combination of the Harmonized World Soil Database and the State Soil Geographic Database, also used to generate the Soil Moisture Active Passive (SMAP) Level-4 soil moisture product, and (MAP2) a hybrid of HWSD-STATSGO and the ‘PEATMAP’ product, which is mainly compiled from national peatland maps. MAP2 indicates ~30 % more peatland area over the Northern Hemisphere. For both peat distributions, CLSM is run and parameters of the RTM are calibrated with 10 years of multi-angular L-band Tb observations from the Soil Moisture and Ocean Salinity SMOS mission. Afterwards, CLSM is run together with the calibrated RTM within a data assimilation system, with and without (open-loop) assimilating SMAP Tb observations, for the period 2015-2020. Our results demonstrate that Tb misfits (in both the open-loop and assimilation runs) are reduced in the areas with the largest differences in peat distribution, thus indicating a basic validity of assuming a peatland-like hydrological dynamics for the larger peat extent of MAP2. Results will be discussed in the context of how peatlands are defined in global peatland maps and the question of what is typically modeled as a peatland in Earth System Models. We propose the evaluation of future releases of peatland maps in Tb space as a tool to evaluate their suitability for implementation into Earth System Models.</p>

New Vegetation Radiative Transfer Schemes for Land Surface Models

2020 ◽  
Author(s):  
Tristan Quaife
Keyword(s):  
Radiative Transfer ◽  
Computational Efficiency ◽  
Land Surface ◽  
Multiple Scales ◽  
Transfer Problem ◽  
Building Blocks ◽  
Surface Model ◽  
Earth System ◽  
System Models ◽  
Earth System Models

<p>The land surface components of climate and Earth system models tend to utilise relatively simple representations of vegetation radiative transfer processes to determine key land surface properties such as albedo, land surface temperature and the absorption of sunlight for photosynthesis. This simplicity is driven, in large part, by a need for computational efficiency. However, a growing number of studies have pointed to the need for more complex radiative transfer in these models.</p><p>An almost ubiquitous assumption in such radiative transfer schemes is that a vegetation canopy can be represented by a plane-parallel, turbid medium – a perfectly flat box in which scattering elements (i.e. leaves, branches, trunks, etc.) are randomly distributed. Real canopies typically exhibit quite complex, non-random structures often involving the clumping of leaves and branches at multiple scales. Furthermore, the optical properties of canopies are typically assumed to be vertically and horizontally homogeneous which does not allow for realistic representation of, for example, forest stands with mixed species or understory vegetation.</p><p>This presentation examines recent developments that have the potential to overcome these and other deficiencies in land surface model radiative transfer schemes, whilst maintaining sufficient computational efficiency to make them viable for inclusion in climate and Earth system models. This is achieved by using the same solutions to the transfer problem as currently employed in climate models as the building blocks to construct canopies that can vary both vertically and horizontally. </p><p> </p>

scholarly journals A stand-alone tree demography and landscape structure module for Earth system models: integration with global forest data

2014 ◽  
Vol 11 (2) ◽  
pp. 2343-2382 ◽  
Author(s):  
V. Haverd ◽  
B. Smith ◽  
L. P. Nieradzik ◽  
P. R. Briggs
Keyword(s):  
Land Surface ◽  
Second Generation ◽  
Land Surface Model ◽  
Woody Biomass ◽  
Surface Model ◽  
Earth System ◽  
Large Area ◽  
System Models ◽  
Biomass Turnover ◽  
Earth System Models

Abstract. Poorly constrained rates of biomass turnover are a key limitation of Earth system models (ESM). In light of this, we recently proposed a new approach encoded in a model called Populations-Order-Physiology (POP), for the simulation of woody ecosystem stand dynamics, demography and disturbance-mediated heterogeneity. POP is suitable for continental to global applications and designed for coupling to the terrestrial ecosystem component of any ESM. POP bridges the gap between first generation Dynamic Vegetation Models (DVMs) with simple large-area parameterisations of woody biomass (typically used in current ESMs) and complex second generation DVMs, that explicitly simulate demographic processes and landscape heterogeneity of forests. The key simplification in the POP approach, compared with second-generation DVMs, is to compute physiological processes such as assimilation at grid-scale (with CABLE or a similar land surface model), but to partition the grid-scale biomass increment among age classes defined at sub grid-scale, each subject to its own dynamics. POP was successfully demonstrated along a savanna transect in northern Australia, replicating the effects of strong rainfall and fire disturbance gradients on observed stand productivity and structure. Here, we extend the application of POP to a range of forest types around the globe, employing paired observations of stem biomass and density from forest inventory data to calibrate model parameters governing stand demography and biomass evolution. The calibrated POP model is then coupled to the CABLE land surface model and the combined model (CABLE-POP) is evaluated against leaf-stem allometry observations from forest stands ranging in age from 3 to 200 yr. Results indicate that simulated biomass pools conform well with observed allometry. We conclude that POP represents a preferable alternative to large-area parameterisations of woody biomass turnover, typically used in current ESMs.

scholarly journals A New Global Storage‐Area‐Depth Data Set for Modeling Reservoirs in Land Surface and Earth System Models

2018 ◽  
Vol 54 (12) ◽  
Author(s):  
Wondmagegn Yigzaw ◽  
Hong‐Yi Li ◽  
Yonas Demissie ◽  
Mohamad I. Hejazi ◽  
L. Ruby Leung ◽  
...  
Keyword(s):  
Land Surface ◽  
Earth System ◽  
Data Set ◽  
System Models ◽  
Depth Data ◽  
Storage Area ◽  
Earth System Models

scholarly journals A Budyko framework for estimating how spatial heterogeneity and lateral moisture redistribution affect average evapotranspiration rates as seen from the atmosphere

2017 ◽  
Vol 21 (1) ◽  
pp. 217-233 ◽  
Author(s):  
Elham Rouholahnejad Freund ◽  
James W. Kirchner
Keyword(s):  
Spatial Heterogeneity ◽  
Land Surface ◽  
Potential Evapotranspiration ◽  
Aridity Index ◽  
Grid Cell ◽  
Earth System ◽  
Grid Cells ◽  
Closure Relation ◽  
System Models ◽  
Earth System Models

Abstract. Most Earth system models are based on grid-averaged soil columns that do not communicate with one another, and that average over considerable sub-grid heterogeneity in land surface properties, precipitation (P), and potential evapotranspiration (PET). These models also typically ignore topographically driven lateral redistribution of water (either as groundwater or surface flows), both within and between model grid cells. Here, we present a first attempt to quantify the effects of spatial heterogeneity and lateral redistribution on grid-cell-averaged evapotranspiration (ET) as seen from the atmosphere over heterogeneous landscapes. Our approach uses Budyko curves, as a simple model of ET as a function of atmospheric forcing by P and PET. From these Budyko curves, we derive a simple sub-grid closure relation that quantifies how spatial heterogeneity affects average ET as seen from the atmosphere. We show that averaging over sub-grid heterogeneity in P and PET, as typical Earth system models do, leads to overestimations of average ET. For a sample high-relief grid cell in the Himalayas, this overestimation bias is shown to be roughly 12 %; for adjacent lower-relief grid cells, it is substantially smaller. We use a similar approach to derive sub-grid closure relations that quantify how lateral redistribution of water could alter average ET as seen from the atmosphere. We derive expressions for the maximum possible effect of lateral redistribution on average ET, and the amount of lateral redistribution required to achieve this effect, using only estimates of P and PET in possible source and recipient locations as inputs. We show that where the aridity index P/PET increases with altitude, gravitationally driven lateral redistribution will increase average ET (and models that overlook lateral redistribution will underestimate average ET). Conversely, where the aridity index P/PET decreases with altitude, gravitationally driven lateral redistribution will decrease average ET. The effects of both sub-grid heterogeneity and lateral redistribution will be most pronounced where P is inversely correlated with PET across the landscape. Our analysis provides first-order estimates of the magnitudes of these sub-grid effects, as a guide for more detailed modeling and analysis.

Impact of improved land model depth and hydrology on climate change projections

2020 ◽  
Author(s):  
Norman Steinert ◽  
Fidel González-Rouco ◽  
Stefan Hagemann ◽  
Philipp de Vrese ◽  
Elena García-Bustamante ◽  
...  
Keyword(s):  
Climate Change ◽  
Land Surface ◽  
Earth System ◽  
Energy Propagation ◽  
Near Surface ◽  
Bottom Boundary ◽  
System Models ◽  
Model Component ◽  
And Storage ◽  
Earth System Models

<p>The representation of the thermal and hydrological state in the land model component of Earth System Models is crucial to have a realistic simulation of subsurface processes and the coupling between the atmo-, lito- and biosphere. There is evidence suggesting an inaccurate simulation of subsurface thermodynamics in current-generation Earth System Models, which have Land Surface Models that are too shallow. In simulations with a bottom boundary too close to the surface, the energy propagation and spatio-temporal variability of subsurface temperatures are affected. This potentially restrains the simulation of land-air interactions and subsurface phenomena, e.g. energy/moisture balance and storage capacity, freeze/thaw cycles and permafrost evolution. We introduce modifications for a deeper soil into the JSBACH soil model component of the MPI-ESM for climate projections of the 21st century. Subsurface layers are added progressively to increase the bottom boundary depth from 10m to 1400m. This leads to near-surface cooling of the soil and encourages regional terrestrial energy uptake by one order of magnitude and more. <br>The depth-changes in the soil also have implications for the hydrological regime, in which the moisture between the surface and the bedrock is sensitive to variations in the thermal regime. Additionally, we compare two different global soil parameter datasets that have major implications for the vertical distribution and availability of soil moisture and its exchange with the land surface. The implementation of supercool water and water phase changes in the soil creates a coupling between the soil thermal and hydrological regimes. In both cases of bottom boundary and water depth changes, we explore the sensitivity of JSBACH from the perspective of changes in the soil thermodynamics, energy balance and storage, as well as the effect of including freezing and thawing processes and their influence on the simulation of permafrost areas in the Northern Hemisphere high latitudes. The latter is of particular interest due to their vulnerability to long-term climate change.</p>

scholarly journals Two-way coupling between the sub-grid land surface and river networks in Earth system models

2020 ◽  
Author(s):  
Nathaniel W. Chaney ◽  
Laura Torres-Rojas ◽  
Noemi Vergopolan ◽  
Colby K. Fisher
Keyword(s):  
High Resolution ◽  
Land Surface ◽  
Environmental Data ◽  
Added Value ◽  
Earth System ◽  
Field Scale ◽  
System Models ◽  
Routing Scheme ◽  
Earth System Models ◽  
River Routing

Abstract. Over the past decade, there has been appreciable progress towards modeling the water, energy, and carbon cycles at field-scales (10–100 m) over continental to global extents. One such approach, named HydroBlocks, accomplishes this task while maintaining computational efficiency via sub-grid tiles, or Hydrologic Response Units (HRUs), learned via a hierarchical clustering approach from available global high-resolution environmental data. However, until now, there has yet to be a macroscale river routing approach that is able to leverage HydroBlocks' approach to sub-grid heterogeneity, thus limiting the added value of field-scale land surface modeling in Earth System Models (e.g., riparian zone dynamics, irrigation from surface water, and interactive floodplains). This paper introduces a novel dynamic river routing scheme in HydroBlocks that is intertwined with the modeled field-scale land surface heterogeneity. The primary features of the routing scheme include: 1) the fine-scale river network of each macroscale grid cell's is derived from very high resolution (

Sign in / Sign up

Export Citation Format

Share Document