Strong contribution of autumn phenology to changes in satellite-derived growing season length estimates across Europe (1982–2011)

I. Garonna, R. de Jong, A.J.W. de Wit, C.A. Mücher, B. Schmid, M.E. Schaepman

Research output: Contribution to journalArticleAcademicpeer-review

95 Citations (Scopus)

Abstract

Land Surface Phenology (LSP) is the most direct representation of intra-annual dynamics of vegetated land surfaces as observed from satellite imagery. LSP plays a key role in characterizing land-surface fluxes, and is central to accurately parameterizing terrestrial biosphere–atmosphere interactions, as well as climate models. In this article, we present an evaluation of Pan-European LSP and its changes over the past 30 years, using the longest continuous record of Normalized Difference Vegetation Index (NDVI) available to date in combination with a landscape-based aggregation scheme. We used indicators of Start-Of-Season, End-Of-Season and Growing Season Length (SOS, EOS and GSL, respectively) for the period 1982–2011 to test for temporal trends in activity of terrestrial vegetation and their spatial distribution. We aggregated pixels into ecologically representative spatial units using the European Landscape Classification (LANMAP) and assessed the relative contribution of spring and autumn phenology. GSL increased significantly by 18–24 days decade-1 over 18–30% of the land area of Europe, depending on methodology. This trend varied extensively within and between climatic zones and landscape classes. The areas of greatest growing-season lengthening were the Continental and Boreal zones, with hotspots concentrated in southern Fennoscandia, Western Russia and pockets of continental Europe. For the Atlantic and Steppic zones, we found an average shortening of the growing season with hotspots in Western France, the Po valley, and around the Caspian Sea. In many zones, changes in the NDVI-derived end-of-season contributed more to the GSL trend than changes in spring green-up, resulting in asymmetric trends. This underlines the importance of investigating senescence and its underlying processes more closely as a driver of LSP and global change.
Original languageEnglish
Pages (from-to)3457-3470
JournalGlobal Change Biology
Volume20
Issue number11
DOIs
Publication statusPublished - 2014

Fingerprint

phenology
land surface
growing season
autumn
Satellites
NDVI
Climate models
Satellite imagery
EOS
surface flux
senescence
global change
satellite imagery
Spatial distribution
Europe
climate modeling
pixel
Agglomeration
Pixels
Fluxes

Keywords

  • land-surface phenology
  • high-resolution radiometer
  • vegetation index ndvi
  • spring phenology
  • climate-change
  • time-series
  • monitoring vegetation
  • trends
  • avhrr
  • models

Cite this

@article{cede3fd5b73349e39d33bfe076963c2c,
title = "Strong contribution of autumn phenology to changes in satellite-derived growing season length estimates across Europe (1982–2011)",
abstract = "Land Surface Phenology (LSP) is the most direct representation of intra-annual dynamics of vegetated land surfaces as observed from satellite imagery. LSP plays a key role in characterizing land-surface fluxes, and is central to accurately parameterizing terrestrial biosphere–atmosphere interactions, as well as climate models. In this article, we present an evaluation of Pan-European LSP and its changes over the past 30 years, using the longest continuous record of Normalized Difference Vegetation Index (NDVI) available to date in combination with a landscape-based aggregation scheme. We used indicators of Start-Of-Season, End-Of-Season and Growing Season Length (SOS, EOS and GSL, respectively) for the period 1982–2011 to test for temporal trends in activity of terrestrial vegetation and their spatial distribution. We aggregated pixels into ecologically representative spatial units using the European Landscape Classification (LANMAP) and assessed the relative contribution of spring and autumn phenology. GSL increased significantly by 18–24 days decade-1 over 18–30{\%} of the land area of Europe, depending on methodology. This trend varied extensively within and between climatic zones and landscape classes. The areas of greatest growing-season lengthening were the Continental and Boreal zones, with hotspots concentrated in southern Fennoscandia, Western Russia and pockets of continental Europe. For the Atlantic and Steppic zones, we found an average shortening of the growing season with hotspots in Western France, the Po valley, and around the Caspian Sea. In many zones, changes in the NDVI-derived end-of-season contributed more to the GSL trend than changes in spring green-up, resulting in asymmetric trends. This underlines the importance of investigating senescence and its underlying processes more closely as a driver of LSP and global change.",
keywords = "land-surface phenology, high-resolution radiometer, vegetation index ndvi, spring phenology, climate-change, time-series, monitoring vegetation, trends, avhrr, models",
author = "I. Garonna and {de Jong}, R. and {de Wit}, A.J.W. and C.A. M{\"u}cher and B. Schmid and M.E. Schaepman",
year = "2014",
doi = "10.1111/gcb.12625",
language = "English",
volume = "20",
pages = "3457--3470",
journal = "Global Change Biology",
issn = "1354-1013",
publisher = "Wiley",
number = "11",

}

Strong contribution of autumn phenology to changes in satellite-derived growing season length estimates across Europe (1982–2011). / Garonna, I.; de Jong, R.; de Wit, A.J.W.; Mücher, C.A.; Schmid, B.; Schaepman, M.E.

In: Global Change Biology, Vol. 20, No. 11, 2014, p. 3457-3470.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Strong contribution of autumn phenology to changes in satellite-derived growing season length estimates across Europe (1982–2011)

AU - Garonna, I.

AU - de Jong, R.

AU - de Wit, A.J.W.

AU - Mücher, C.A.

AU - Schmid, B.

AU - Schaepman, M.E.

PY - 2014

Y1 - 2014

N2 - Land Surface Phenology (LSP) is the most direct representation of intra-annual dynamics of vegetated land surfaces as observed from satellite imagery. LSP plays a key role in characterizing land-surface fluxes, and is central to accurately parameterizing terrestrial biosphere–atmosphere interactions, as well as climate models. In this article, we present an evaluation of Pan-European LSP and its changes over the past 30 years, using the longest continuous record of Normalized Difference Vegetation Index (NDVI) available to date in combination with a landscape-based aggregation scheme. We used indicators of Start-Of-Season, End-Of-Season and Growing Season Length (SOS, EOS and GSL, respectively) for the period 1982–2011 to test for temporal trends in activity of terrestrial vegetation and their spatial distribution. We aggregated pixels into ecologically representative spatial units using the European Landscape Classification (LANMAP) and assessed the relative contribution of spring and autumn phenology. GSL increased significantly by 18–24 days decade-1 over 18–30% of the land area of Europe, depending on methodology. This trend varied extensively within and between climatic zones and landscape classes. The areas of greatest growing-season lengthening were the Continental and Boreal zones, with hotspots concentrated in southern Fennoscandia, Western Russia and pockets of continental Europe. For the Atlantic and Steppic zones, we found an average shortening of the growing season with hotspots in Western France, the Po valley, and around the Caspian Sea. In many zones, changes in the NDVI-derived end-of-season contributed more to the GSL trend than changes in spring green-up, resulting in asymmetric trends. This underlines the importance of investigating senescence and its underlying processes more closely as a driver of LSP and global change.

AB - Land Surface Phenology (LSP) is the most direct representation of intra-annual dynamics of vegetated land surfaces as observed from satellite imagery. LSP plays a key role in characterizing land-surface fluxes, and is central to accurately parameterizing terrestrial biosphere–atmosphere interactions, as well as climate models. In this article, we present an evaluation of Pan-European LSP and its changes over the past 30 years, using the longest continuous record of Normalized Difference Vegetation Index (NDVI) available to date in combination with a landscape-based aggregation scheme. We used indicators of Start-Of-Season, End-Of-Season and Growing Season Length (SOS, EOS and GSL, respectively) for the period 1982–2011 to test for temporal trends in activity of terrestrial vegetation and their spatial distribution. We aggregated pixels into ecologically representative spatial units using the European Landscape Classification (LANMAP) and assessed the relative contribution of spring and autumn phenology. GSL increased significantly by 18–24 days decade-1 over 18–30% of the land area of Europe, depending on methodology. This trend varied extensively within and between climatic zones and landscape classes. The areas of greatest growing-season lengthening were the Continental and Boreal zones, with hotspots concentrated in southern Fennoscandia, Western Russia and pockets of continental Europe. For the Atlantic and Steppic zones, we found an average shortening of the growing season with hotspots in Western France, the Po valley, and around the Caspian Sea. In many zones, changes in the NDVI-derived end-of-season contributed more to the GSL trend than changes in spring green-up, resulting in asymmetric trends. This underlines the importance of investigating senescence and its underlying processes more closely as a driver of LSP and global change.

KW - land-surface phenology

KW - high-resolution radiometer

KW - vegetation index ndvi

KW - spring phenology

KW - climate-change

KW - time-series

KW - monitoring vegetation

KW - trends

KW - avhrr

KW - models

U2 - 10.1111/gcb.12625

DO - 10.1111/gcb.12625

M3 - Article

VL - 20

SP - 3457

EP - 3470

JO - Global Change Biology

JF - Global Change Biology

SN - 1354-1013

IS - 11

ER -