A suite of global, cross-scale topographic variables for environmental and biodiversity modeling.

Amatulli G, Domisch S, Tuanmu MN, Parmentier B, Ranipeta A, Malczyk J, Jetz W. · 2018

Read A suite of global, cross-scale topographic variables for environmental and biodiversity modeling. by Amatulli G, Domisch S, Tuanmu MN, Parmentier B, Ranipeta A, Malczyk J, Jetz W. (2018) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc5859920.txt.

Introduction

Amatulli and co-authors present a global dataset of topographic variables. Their article appeared in Scientific Data. The dataset serves research in ecology, climate, and land science. It was built from two digital elevation models of the Earth. One has 250-metre detail, and the other has 90-metre detail. From them, the team derived slope, aspect, roughness, and landform classes.

Each variable was also averaged to five coarser grid sizes. The coarsest grids reach 100 kilometres per cell. Tests in four mountain regions show local and scale differences. Many variables track each other closely, the team notes. The full set is free to download and view online. It gives modellers one standard terrain source across scales.

Catalog record

CREATORAmatulli G, Domisch S, Tuanmu MN, Parmentier B, Ranipeta A, Malczyk J, Jetz W.
YEAR2018
COLLECTIONPAPERS
FORMATTXT
FILE SIZE53 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC5859920

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A suite of global, cross-scale topographic variables for environmental and biodiversity modeling By Giuseppe Amatulli, Sami Domisch, Mao-Ning Tuanmu, Benoit Parmentier, Ajay Ranipeta, Jeremy Malczyk, Walter Jetz Published in: Scientific Data DOI: 10.1038/sdata.2018.40 ABSTRACT Abstract Topographic variation underpins a myriad of patterns and processes in hydrology, climatology, geography and ecology and is key to understanding the variation of life on the planet. A fully standardized and global multivariate product of different terrain features has the potential to support many large-scale research applications, however to date, such datasets are unavailable. Here we used the digital elevation model products of global 250 m GMTED2010 and near-global 90 m SRTM4.1dev to derive a suite of topographic variables: elevation, slope, aspect, eastness, northness, roughness, terrain roughness index, topographic position index, vector ruggedness measure, profile/tangential curvature, first/second order partial derivative, and 10 geomorphological landform classes. We aggregated each variable to 1, 5, 10, 50 and 100 km spatial grains using several aggregation approaches. While a cross-correlation underlines the high similarity of many variables, a more detailed view in four mountain regions reveals local differences, as well as scale variations in the aggregated variables at different spatial grains. All newly-developed variables are available for download at Data Citation 1 and for download and visualization at http://www.earthenv.org/topography. Subject terms: Ecological modelling, Geomorphology, Hydrogeology BACKGROUND & SUMMARY Spatial heterogeneity is an important driver of environmental complexity in a region and influences the sub-regional variation of (i) abiotic factors such as micro/macro climates, soil composition, dynamic processes of the hydrological systems and (ii) biotic factors such as species richness and structure, population complexity, animal movement1. Topography as measured by elevation and its derived variables (e.g., slope and aspect), is key for characterizing spatial heterogeneity and the abiotic environment in a given area, subsequently driving hydrological, geomorphological, and biological processes1,2. For instance, elevation has numerous dependencies regarding topographic complexity, micro/macro climates or land cover, and can be used to assess biodiversity patterns across the globe by relating species occurrences to environmental factors3,4. Besides using the ‘raw’ elevation from a digital elevation model (DEM), a multitude of topographic metrics can be extracted from the DEM to better understand the physical geographic context and landscape properties of a study region3,5. For instance, slope and terrain curvatures (defined as a measure of the concavity and convexity, or convergence and divergence) play an important role in catchment-related hydrological responses driving the flow direction, water runoff velocity, water accumulation, soil erosion and soil moisture6. Similarly, topographic variation strongly influences the accumulation and heterogeneity of mountain/alpine snow cover7 and landslide formation8. All these factors regulate the water availability in soil and thus directly influence vegetation moisture, and this constitutes an important element for wildfire risk modeling9. Specific terrain attributes may also provide important conditions facilitating the movements and migration of mobile species such as birds10,11. Moreover, terrain characteristics may represent vital refugia for species under climate change12, a function that likely varies with spatial grain. Traditionally, DEMs have been derived from aerial photography with stereoscopy or ground surveys. In the last decades, advances in remote sensing technology have contributed to the availability of DEM products with different spatial grains. These products provide gridded information about land surface elevation, where fine spatial grain DEMs (<10 m) derived from laser sensors (e.g., Light Detection and Ranging—LIDAR), and DEMs on medium (10-100 m) to coarse spatial grains (>100 m) are obtained from optical sensors (e.g., Advanced Spaceborne Thermal Emission and Reflection Radiometer—ASTER) and radar sensors (Shuttle Radar Topography Mission—SRTM)13. Beside a global cover age at coarse spatial grains, also national and regional DEMs at fine grains are available, such as ArcticDEM (2 m), TanDEM-X (12 m), 3DEP (1 m), CDEM (25 m), AW3D (5 m). Nonetheless, these DEMs are not fully calibrated and their local extension does not allow a global implementation. Environmental and biological processes depend strongly on the spatial grain of the input variables14,15. For instance, to successfully predict species occurrences using species distribution models, the spatial grain of the environmental variables has to match and agree with the sample size and spatial data accuracy, extent of the species data, movement and dispersal range, etc. Coarsening the environmental predictors to, e.g., 50-100 km, may be needed to match coarse atlas data, expert range information, or point records with a high spatial uncertainty16,17. A variety of DEMs elaborated under different projects are currently available and grouped/listed at http://worldgrids.org/doku.php/wiki:dem_sources. However, the calculation and aggregation of elevation-derived variables across spatial grains are typically done on a case-by-case basis. Hence, a coherent and standardized framework for obtaining range-wide, spatially aligned, and ready-to-use topographic variables at different spatial grains is missing. Currently available DEMs may lack the quality required for many applications. For instance, ASTER GDEM contains artifacts and outliers (i.e., abrupt rise—‘humps/bumps’; fall—‘pits’) that require post-processing to remove errors or offsets using additional sources (i.e., SRTM or ICESat laser altimetry)18. These anomalies can produce large elevation errors on local scales and impede the use of ASTER GDEM for specific applications. Hydrological applications, for example, require high quality standardized inputs to derive watershed properties that are, in turn, used to track water flows and quality. Similarly, watershed delineation, drainage divide and channel network increasingly rely on digital data from DEMs and derived topography metrics rather than manual or visual interpretation from photography or ground data. In addition, as topography is one of the major driver of local temperature and precipitation patterns, topographic layers (elevation and aspect in particular) are important inputs in many climate layers (e.g., WorldClim http://www.worldclim.org/). Errors or inaccuracies in the inputs can propagate throughout the analyses and derived datasets. Assembling readily available data of high quality is therefore useful in many fields. Of particular relevance for many applications is a global coverage and comparability among different regions. This is emphasized by assessment processes such as those of the Intergovernmental Platform for Biodiversity and Ecosystem Services (IPBES)19, where global availability and comparability of data is crucial. The modeling of species distributions worldwide, as targeted in projects such as Map of Life (https://www.mol.org20), requires a rich stack of global environmental information, which to date, has only been partially developed in regards to terrain. To address the lack of readily available topographic variables at a global extent, we calculated a suite of elevation-derived topographic variables based on the 250 m Global Multi-resolution Terrain Elevation Data 2010 dataset21 (hereafter ‘GMTED’). As a comparison, we also calculated the most relevant variables using the widely used 90 m Shuttle Radar Topographic Mission22,23 void filled dataset SRTM4.1dev24,25 (hereafter ‘SRTM’). We resampled (i.e., aggregated) each metric to coarser spatial grains (1, 5, 10, 50, 100 km) using several spatial aggregation approaches to characterize the global topographic heterogeneity (see Aggregated topographic variables session). In total, we computed 15 continuous variables that characterize the shape of the relief (elevation, slope, aspect sine/cosine, eastness, northness, roughness, terrain ruggedness index, topographic position index, vector ruggedness measure, profile/tangential curvature, first/second order partial derivative) and 1 categorical variable, describing ten major landform classes. The variables (Data Citation 1) are ready to use as input data in various environmental models and analyses, and we provide the code to calculate user-specified variables (e.g., coefficient of variation, range) on custom spatial grains. The newly developed layers

Read the complete text