Using aboveground vegetation attributes as proxies for mapping peatland belowground carbon stocks

Lopatin, Javier; Kattenborn, Teja; Galleguillos, Mauricio; Perez‐Quezada, Jorge F.; Schmidtlein, Sebastian

doi:10.1016/j.rse.2019.111217

Cited by 30 publications

(30 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Specifically, variation and heterogeneity in spectral reflectance increased with increasing Shannon diversity among plots. Increasing spectral heterogeneity has been previously used as a proxy for diversity in studies that have mapped peatland ecological diversity using remote sensing [91][92][93]. This type of study is based on the species-area relationship in which species richness increases with scale across a heterogeneous landscape [94].…”

Section: Remote Sensing Of Boreal Peatland Species Diversitymentioning

confidence: 99%

Characterizing Boreal Peatland Plant Composition and Species Diversity with Hyperspectral Remote Sensing

et al. 2019

View full text Add to dashboard Cite

Peatlands, which account for approximately 15% of land surface across the arctic and boreal regions of the globe, are experiencing a range of ecological impacts as a result of climate change. Factors that include altered hydrology resulting from drought and permafrost thaw, rising temperatures, and elevated levels of atmospheric carbon dioxide have been shown to cause plant community compositional changes. Shifts in plant composition affect the productivity, species diversity, and carbon cycling of peatlands. We used hyperspectral remote sensing to characterize the response of boreal peatland plant composition and species diversity to warming, hydrologic change, and elevated CO2. Hyperspectral remote sensing techniques offer the ability to complete landscape-scale analyses of ecological responses to climate disturbance when paired with plot-level measurements that link ecosystem biophysical properties with spectral reflectance signatures. Working within two large ecosystem manipulation experiments, we examined climate controls on composition and diversity in two types of common boreal peatlands: a nutrient rich fen located at the Alaska Peatland Experiment (APEX) in central Alaska, and an ombrotrophic bog located in northern Minnesota at the Spruce and Peatland Responses Under Changing Environments (SPRUCE) experiment. We found a strong effect of plant functional cover on spectral reflectance characteristics. We also found a positive relationship between species diversity and spectral variation at the APEX field site, which is consistent with other recently published findings. Based on the results of our field study, we performed a supervised land cover classification analysis on an aerial hyperspectral dataset to map peatland plant functional types (PFTs) across an area encompassing a range of different plant communities. Our results underscore recent advances in the application of remote sensing measurements to ecological research, particularly in far northern ecosystems.

show abstract

Section: Remote Sensing Of Boreal Peatland Species Diversitymentioning

confidence: 99%

Characterizing Boreal Peatland Plant Composition and Species Diversity with Hyperspectral Remote Sensing

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Michez et al 2016;Lu and He 2017). Another option is to derive metrics describing the 3D structure of the canopy derived from the UAV-based photogrammetric point clouds (Brodu and Lague 2012;Getzin et al 2012;Kattenborn et al 2014Lopatin et al 2019). Yet, these 2D texture and 3D structure approaches cannot automatically and sufficiently exploit the available information to detect differences in vegetation characteristics that are readily perceived by human eyes (Lopatin et al 2018;.…”

Section: Introductionmentioning

confidence: 99%

Convolutional Neural Networks accurately predict cover fractions of plant species and communities in Unmanned Aerial Vehicle imagery

Kattenborn

Eichel

Wiser

et al. 2020

Remote Sens Ecol Conserv

Self Cite

View full text Add to dashboard Cite

Unmanned Aerial Vehicles (UAV) greatly extended our possibilities to acquire high resolution remote sensing data for assessing the spatial distribution of species composition and vegetation characteristics. Yet, current pixel-or texturebased mapping approaches do not fully exploit the information content provided by the high spatial resolution. Here, to fully harness this spatial detail, we apply deep learning techniques, that is, Convolutional Neural Networks (CNNs), on regular tiles of UAV-orthoimagery (here 2-5 m) to identify the cover of target plant species and plant communities. The approach was tested with UAV-based orthomosaics and photogrammetric 3D information in three case studies, that is, (1) mapping tree species cover in primary forests, (2) mapping plant invasions by woody species into forests and open land and (3) mapping vegetation succession in a glacier foreland. All three case studies resulted in high predictive accuracies. The accuracy increased with increasing tile size (2-5 m) reflecting the increased spatial context captured by a tile. The inclusion of 3D information derived from the photogrammetric workflow did not significantly improve the models. We conclude that CNN are powerful in harnessing high resolution data acquired from UAV to map vegetation patterns. The study was based on low cost red, green, blue (RGB) sensors making the method accessible to a wide range of users. Combining UAV and CNN will provide tremendous opportunities for ecological applications.

show abstract

“…The change of vegetation cover is commonly detected by NDVI, which is a comprehensive indicator to reflect the growth of plants and is directly linked to aboveground biomass. The aboveground is closely related to the belowground biomass (Oñatibia et al, 2017;Lopatin et al, 2019), which is determined the root's ability in reducing soil erosion (Hu and Huang, 2019;Ozdemir and Yilmaz, 2020).…”

Section: Introductionmentioning

confidence: 99%

Quantifying the Effects of Vegetation Restorations on the Soil Erosion Export and Nutrient Loss on the Loess Plateau

et al. 2020

View full text Add to dashboard Cite

The transport of eroded soil to rivers changes the nutrient cycles of river ecosystems and has significant impacts on the regional eco-environment and human health. The Loess Plateau, a leading vegetation restoration region in China and the world, has experienced severe soil erosion and nutrient loss, however, the extent to which vegetation restoration prevents soil erosion export (to rivers) and it caused nutrient loss is unknown. To evaluate the effects of the first stage of the Grain for Green Project (GFGP) on the Loess Plateau (started in 1999 and ended in 2013), we analyzed the vegetation change trends and quantified the effects of GFGP on soil erosion export (to rivers) and it caused nutrient loss by considering soil erosion processes. The results were as follows: (1) in the first half of study period (from 1982 to 1998), the vegetation cover changed little, but after the implementation of the first stage of the GFGP (from 1999 to 2013), the vegetation cover of 75.0% of the study area showed a significant increase; (2) The proportion of eroded areas decreased from 41.8 to 26.7% as a result of the GFGP, and the erosion intensity lessened in most regions; the implementation significantly reduce the soil nutrient loss; (3) at the county level, soil erosion export could be avoided significantly by the increasing of vegetation greenness in the study area (R = −0.49). These results illustrate the relationships among changes in vegetation cover, soil erosion and nutrient export, which could provide a reference for local government for making ecology-relative policies.

show abstract

Using aboveground vegetation attributes as proxies for mapping peatland belowground carbon stocks

Cited by 30 publications

References 56 publications

Characterizing Boreal Peatland Plant Composition and Species Diversity with Hyperspectral Remote Sensing

Characterizing Boreal Peatland Plant Composition and Species Diversity with Hyperspectral Remote Sensing

Convolutional Neural Networks accurately predict cover fractions of plant species and communities in Unmanned Aerial Vehicle imagery

Quantifying the Effects of Vegetation Restorations on the Soil Erosion Export and Nutrient Loss on the Loess Plateau

Contact Info

Product

Resources

About