Ecosystem properties of semiarid savanna grassland in West Africa and its relationship with environmental variability

Torbern Tagesson, Rasmus Fensholt, Idrissa Guiro, Mads Olander Rasmussen, Silvia Huber, Cheikh Mbow, Monica Garcia, Stéphanie Marie Anne F Horion, Inge Sandholt, Bo Holm-Rasmussen, Frank M Göttsche, Marc-Etienne Ridler, Niklas Olén, Jørgen Lundegaard Olsen, Andrea Ehammer, Mathias Madsen, Folke S Olesen, Jonas Ardö

    Research output: Contribution to journalJournal articleResearchpeer-review

    68 Citations (Scopus)

    Abstract

    The Dahra field site in Senegal, West Africa, was established in 2002 to monitor ecosystem properties of semi-arid savanna grassland and their responses to climatic and environmental change. This paper describes the environment and the ecosystem properties of the site using a unique set of in situ data. The studied variables include hydroclimatic variables, species composition, albedo, normalized difference vegetation index (NDVI), hyperspectral characteristics (350-1800 nm), surface reflectance anisotropy, brightness temperature, fraction of absorbed photosynthetic active radiation (FAPAR), biomass, vegetation water content, and land-atmosphere exchanges of carbon (NEE) and energy. The Dahra field site experiences a typical Sahelian climate and is covered by coexisting trees (~3% canopy cover) and grass species, characterizing large parts of the Sahel. This makes the site suitable for investigating relationship between ecosystem properties and hydroclimatic variables for semi-arid savanna ecosystems of the region. There were strong inter-annual, seasonal and diurnal dynamics in NEE, with high values of ~-7.5 g C m(-2) day(-1) during the peak of the growing season. We found neither browning nor greening NDVI trends from 2002 to 2012. Inter-annual variation in species composition was strongly related to rainfall distribution. NDVI and FAPAR were strongly related to species composition, especially for years dominated by the species Zornia glochidiata. This influence was not observed in inter-annual variation in biomass and vegetation productivity, thus challenging dryland productivity models based on remote sensing. Surface reflectance anisotropy (350-1800 nm) at the peak of the growing season varied strongly depending on wavelength and viewing angle thereby having implications for the design of remotely sensed spectral vegetation indices covering different wavelength regions. The presented time series of in situ data has great potential for dryland dynamics studies, global climate change related research and evaluation and parameterization of remote sensing products and dynamic vegetation models. This article is protected by copyright. All rights reserved.

    Original languageEnglish
    JournalGlobal Change Biology
    Volume21
    Issue number1
    Pages (from-to)250–264
    ISSN1354-1013
    DOIs
    Publication statusPublished - 2015

    Cite this