Variation in photosynthetic and nonphotosynthetic vegetation along edaphic and compositional gradients in northwestern Amazonia

M. A. Higgins, G. P. Asner, E. Perez, N. Elespuru, A. Alonso

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Tropical forests vary substantially in aboveground properties such as canopy height, canopy structure, and plant species composition, corresponding to underlying variations in soils and geology. Forest properties are often difficult to detect and map in the field, however, due to the remoteness and inaccessibility of these forests. Spectral mixture analysis of Landsat imagery allows mapping of photosynthetic and nonphotosynthetic vegetation quantities (PV and NPV), corresponding to biophysical properties such as canopy openness, forest productivity, and disturbance. Spectral unmixing has been used for applications ranging from deforestation monitoring to identifying burn scars from past fires, but little is known about variations in PV and NPV in intact rainforests. Here we use spectral unmixing of Landsat imagery to map PV and NPV in northern Amazonia, and to test their relationship to soils and plant species composition. To do this we sampled 117 sites crossing a geological boundary in northwestern Amazonia for soil cation concentrations and plant species composition. We then used the Carnegie Landsat Analysis System to map PV and NPV for these sites from multiple dates of Landsat imagery. We found that soil cation concentrations and plant species composition consistently explain a majority of the variation in remotely sensed PV and NPV values. After combining PV and NPV into a single variable (PV-NPV), we determined that the influence of soil properties on canopy properties was inseparable from the influence of plant species composition. In all cases, patterns in PV and NPV corresponded to underlying geological patterns. Our findings suggest that geology and soils regulate canopy PV and NPV values in intact tropical forests, possibly through changes in plant species composition.

Original languageEnglish (US)
Pages (from-to)3505-3513
Number of pages9
JournalBiogeosciences
Volume11
Issue number13
DOIs
StatePublished - Jul 3 2014
Externally publishedYes

Fingerprint

Amazonia
Landsat
species diversity
vegetation
canopy
imagery
soil
forest canopy
geology
tropical forests
tropical forest
cations
cation
deforestation
systems analysis
rainforest
rain forests
plant species
soil properties
soil property

ASJC Scopus subject areas

  • Ecology, Evolution, Behavior and Systematics
  • Earth-Surface Processes

Cite this

Variation in photosynthetic and nonphotosynthetic vegetation along edaphic and compositional gradients in northwestern Amazonia. / Higgins, M. A.; Asner, G. P.; Perez, E.; Elespuru, N.; Alonso, A.

In: Biogeosciences, Vol. 11, No. 13, 03.07.2014, p. 3505-3513.

Research output: Contribution to journalArticle

@article{59de9b26f8604d4e83b8c03a99ebba5d,
title = "Variation in photosynthetic and nonphotosynthetic vegetation along edaphic and compositional gradients in northwestern Amazonia",
abstract = "Tropical forests vary substantially in aboveground properties such as canopy height, canopy structure, and plant species composition, corresponding to underlying variations in soils and geology. Forest properties are often difficult to detect and map in the field, however, due to the remoteness and inaccessibility of these forests. Spectral mixture analysis of Landsat imagery allows mapping of photosynthetic and nonphotosynthetic vegetation quantities (PV and NPV), corresponding to biophysical properties such as canopy openness, forest productivity, and disturbance. Spectral unmixing has been used for applications ranging from deforestation monitoring to identifying burn scars from past fires, but little is known about variations in PV and NPV in intact rainforests. Here we use spectral unmixing of Landsat imagery to map PV and NPV in northern Amazonia, and to test their relationship to soils and plant species composition. To do this we sampled 117 sites crossing a geological boundary in northwestern Amazonia for soil cation concentrations and plant species composition. We then used the Carnegie Landsat Analysis System to map PV and NPV for these sites from multiple dates of Landsat imagery. We found that soil cation concentrations and plant species composition consistently explain a majority of the variation in remotely sensed PV and NPV values. After combining PV and NPV into a single variable (PV-NPV), we determined that the influence of soil properties on canopy properties was inseparable from the influence of plant species composition. In all cases, patterns in PV and NPV corresponded to underlying geological patterns. Our findings suggest that geology and soils regulate canopy PV and NPV values in intact tropical forests, possibly through changes in plant species composition.",
author = "Higgins, {M. A.} and Asner, {G. P.} and E. Perez and N. Elespuru and A. Alonso",
year = "2014",
month = "7",
day = "3",
doi = "10.5194/bg-11-3505-2014",
language = "English (US)",
volume = "11",
pages = "3505--3513",
journal = "Biogeosciences",
issn = "1726-4170",
publisher = "European Geosciences Union",
number = "13",

}

TY - JOUR

T1 - Variation in photosynthetic and nonphotosynthetic vegetation along edaphic and compositional gradients in northwestern Amazonia

AU - Higgins, M. A.

AU - Asner, G. P.

AU - Perez, E.

AU - Elespuru, N.

AU - Alonso, A.

PY - 2014/7/3

Y1 - 2014/7/3

N2 - Tropical forests vary substantially in aboveground properties such as canopy height, canopy structure, and plant species composition, corresponding to underlying variations in soils and geology. Forest properties are often difficult to detect and map in the field, however, due to the remoteness and inaccessibility of these forests. Spectral mixture analysis of Landsat imagery allows mapping of photosynthetic and nonphotosynthetic vegetation quantities (PV and NPV), corresponding to biophysical properties such as canopy openness, forest productivity, and disturbance. Spectral unmixing has been used for applications ranging from deforestation monitoring to identifying burn scars from past fires, but little is known about variations in PV and NPV in intact rainforests. Here we use spectral unmixing of Landsat imagery to map PV and NPV in northern Amazonia, and to test their relationship to soils and plant species composition. To do this we sampled 117 sites crossing a geological boundary in northwestern Amazonia for soil cation concentrations and plant species composition. We then used the Carnegie Landsat Analysis System to map PV and NPV for these sites from multiple dates of Landsat imagery. We found that soil cation concentrations and plant species composition consistently explain a majority of the variation in remotely sensed PV and NPV values. After combining PV and NPV into a single variable (PV-NPV), we determined that the influence of soil properties on canopy properties was inseparable from the influence of plant species composition. In all cases, patterns in PV and NPV corresponded to underlying geological patterns. Our findings suggest that geology and soils regulate canopy PV and NPV values in intact tropical forests, possibly through changes in plant species composition.

AB - Tropical forests vary substantially in aboveground properties such as canopy height, canopy structure, and plant species composition, corresponding to underlying variations in soils and geology. Forest properties are often difficult to detect and map in the field, however, due to the remoteness and inaccessibility of these forests. Spectral mixture analysis of Landsat imagery allows mapping of photosynthetic and nonphotosynthetic vegetation quantities (PV and NPV), corresponding to biophysical properties such as canopy openness, forest productivity, and disturbance. Spectral unmixing has been used for applications ranging from deforestation monitoring to identifying burn scars from past fires, but little is known about variations in PV and NPV in intact rainforests. Here we use spectral unmixing of Landsat imagery to map PV and NPV in northern Amazonia, and to test their relationship to soils and plant species composition. To do this we sampled 117 sites crossing a geological boundary in northwestern Amazonia for soil cation concentrations and plant species composition. We then used the Carnegie Landsat Analysis System to map PV and NPV for these sites from multiple dates of Landsat imagery. We found that soil cation concentrations and plant species composition consistently explain a majority of the variation in remotely sensed PV and NPV values. After combining PV and NPV into a single variable (PV-NPV), we determined that the influence of soil properties on canopy properties was inseparable from the influence of plant species composition. In all cases, patterns in PV and NPV corresponded to underlying geological patterns. Our findings suggest that geology and soils regulate canopy PV and NPV values in intact tropical forests, possibly through changes in plant species composition.

UR - http://www.scopus.com/inward/record.url?scp=84903760766&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84903760766&partnerID=8YFLogxK

U2 - 10.5194/bg-11-3505-2014

DO - 10.5194/bg-11-3505-2014

M3 - Article

AN - SCOPUS:84903760766

VL - 11

SP - 3505

EP - 3513

JO - Biogeosciences

JF - Biogeosciences

SN - 1726-4170

IS - 13

ER -