Grassland habitats provide many ecosystem services but are threatened by agricultural intensification and urbanization. While the lack of accurate and comprehensive inventories at the national scale makes them difficult to manage, advances in spatial modeling using open remote sensing data and open-source software, as well as the increasing use of ecological archives, provide new perspectives for mapping natural habitats. In this context, this study evaluated the contribution of spectral and environmental variables to discriminate and then map grassland habitats throughout France. To this end, 92 spectral variables derived from moderate-resolution imaging spectroradiometer data, 19 bioclimatic variables derived from WorldClim data, 4 topographic variables derived from the European Union Digital Elevation Model (DEM), and 8 soil variables derived from SoilGrids data were combined at a spatial resolution of 250 m. Reference plots that characterized 6 and 21 grassland ecosystems at European Nature Information System (EUNIS) levels 2 (broad habitats) and 3 (habitats), respectively, were collected from vegetation archives. We first performed descriptive analysis that included habitat description, ordination, and pairwise separability. We then performed predictive analysis of grassland habitats using a cross-validated random forest model that included a spatial constraint. While environmental and spectral variables characterized most grassland habitats well and consistently, some confusion occurred between habitats with similar abiotic conditions. The main grassland habitat types were correctly mapped at EUNIS level 2 (F1 score = 0.68), but not at EUNIS level 3 (F1 score = 0.52). In addition, the two variables that contributed most to the model were the near-infrared spectral band in spring and the minimum temperature of the coldest month. The model’s prediction at EUNIS level 2 for mainland France provides the map of grassland habitats at a new spatial scale.
AdamoMTarantinoCLucasRM, et al. (2015) Combined use of expert knowledge and earth observation data for the land cover mapping of an Italian grassland area: An EODHaM system application. In: 2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), July 2015, pp. 3065–3068.
2.
AdamoMTomaselliVTarantinoC, et al. (2020) Knowledge-based classification of grassland ecosystem based on multi-temporal WorldView-2 Data and FAO-LCCS taxonomy. Remote Sensing12(9): 1447.
AliICawkwellFDwyerE, et al. (2016) Satellite remote sensing of grasslands: From observation to management. Journal of Plant Ecology9(6): 649–671.
5.
AllenVGBatelloCBerrettaEJ, et al. (2011) An international terminology for grazing lands and grazing animals. Grass and Forage Science66(1): 2–28.
6.
Álvarez-MartínezJMJiménez-AlfaroBBarquínJ, et al. (2018) Modelling the area of occupancy of habitat types with remote sensing. Methods in Ecology and Evolution9(3): 580–593.
7.
AmiciVFilibeckGRocchiniD, et al. (2017) Are CORINE land cover classes reliable proxies of plant species assemblages? A test in Mediterranean forest landscapes. Plant Biosystems - An International Journal Dealing with all Aspects of Plant Biology. DOI: 10.1080/11263504.2017.1407372.
8.
BalakrishnamaSGanapathirajuA (1998) Linear Discriminant Analysis - A Brief Tutorial. Institute for Signal and Information Processing, 1998, pp. 1–8.
9.
BarthaGKocsisS (2011) Standardization of geographic data: The European Inspire Directive. European Journal of Geography2(2): 79–89.
10.
BelgiuMDrăguţL (2016) Random forest in remote sensing: A review of applications and future directions. ISPRS Journal of Photogrammetry and Remote Sensing114: 24–31.
11.
BensettitiFBoulletVChavaudret-LaborieC, et al. (eds) (2005) Cahiers d’habitats. In: Natura 2000. Connaissance et Gestion Des Habitats et Des Espèces d’intérêt Communautaire. Paris: La Documentation française.
12.
BreimanL (2001) Random Forests. Machine Learning45(1): 5–32.
13.
BuckOMillánVEGKlinkA, et al. (2015) Using information layers for mapping grassland habitat distribution at local to regional scales. International Journal of Applied Earth Observation and Geoinformation37: 83–89.
14.
BüttnerGMauchaGKosztraB (2016) High-resolution layers. European Landscape Dynamics: CORINE Land Cover Data: 61.
15.
CarliEMassimiMAngeliniP, et al. (2020) How to improve the distribution maps of habitat types at national scale. Rendiconti Lincei. Scienze Fisiche e Naturali31(3): 881–888.
16.
CharmanPEMurphyBW (2007) Soils: Their Properties and Management. New York: Oxford University Press.
17.
CheţanMADornikAUrdeaP (2018) Analysis of recent changes in natural habitat types in the Apuseni Mountains (Romania), using multi-temporal Landsat satellite imagery (1986–2015). Applied Geography97: 161–175.
18.
ChignellSMLuizzaMWSkachS, et al. (2018) An integrative modeling approach to mapping wetlands and riparian areas in a heterogeneous Rocky Mountain watershed. Remote Sensing in Ecology and Conservation4(2): 150–165.
19.
ChristinSHervetÉLecomteN (2019) Applications for deep learning in ecology. Methods in Ecology and Evolution10(10): 1632–1644.
20.
ChytrýMTichýLHennekensSM, et al. (2020) EUNIS Habitat Classification: Expert system, characteristic species combinations and distribution maps of European habitats. Applied Vegetation Science. 23(4): 648–675
21.
ClairMGaudillatVMichezN, et al. (2017) HABREF v3. 1, Référentiel Des Typologies d’habitats et de Végétation Pour La France. Guide Méthodologique. Service du patrimoine naturel, Muséum national d’histoire naturelle, Paris.
22.
CorbaneCLangSPipkinsK, et al. (2015) Remote sensing for mapping natural habitats and their conservation status – New opportunities and challenges. International Journal of Applied Earth Observation and Geoinformation37: 7–16.
23.
CornyJVoundyE-MMonchicourtC, et al. (2019) Web applications for interoperability of biodiversity data in France. Biodiversity Information Science and Standards3: e37947.
24.
Dalle FratteMBrusaGCeraboliniBEL (2019) A low-cost and repeatable procedure for modelling the regional distribution of Natura 2000 terrestrial habitats. Journal of Maps15(2): 79–88.
25.
DaviesCEMossDHillMO (2004) EUNIS Habitat Classification Revised 2004. European Environment Agency - European Topic Centre on Nature Protection and Biodiversity, pp. 127–143.
26.
DeákBRádaiZLukácsK, et al. (2020) Fragmented dry grasslands preserve unique components of plant species and phylogenetic diversity in agricultural landscapes. Biodiversity and Conservation29(14): 4091–4110.
27.
DidanK (2015) MOD13Q1 MODIS/Terra Vegetation Indices 16-Day L3 Global 250m SIN Grid V006. NASA EOSDIS LP DAAC.
28.
DixonAPFaber-LangendoenDJosseC, et al. (2014) Distribution mapping of world grassland types. Journal of Biogeography41(11): 2003–2019.
Faber-LangendoenDKeeler-WolfTMeidingerD, et al. (2014) EcoVeg: A new approach to vegetation description and classification. Ecological Monographs84(4): 533–561.
31.
FickSEHijmansRJ (2017) WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas. International Journal of Climatology37(12): 4302–4315.
32.
FletcherRJDidhamRKBanks-LeiteC, et al. (2018) Is habitat fragmentation good for biodiversity?Biological Conservation226: 9–15.
33.
FontanaFRixenCJonasT, et al. (2008) Alpine grassland phenology as seen in AVHRR, VEGETATION, and MODIS NDVI time series - a comparison with in situ measurements. Sensors8(4): 2833–2853.
34.
FrankeJKeuckVSiegertF (2012) Assessment of grassland use intensity by remote sensing to support conservation schemes. Journal for Nature Conservation20(3): 125–134.
GautamRHsuNCLauWK-M, et al. (2013) Satellite observations of desert dust-induced Himalayan snow darkening. Geophysical Research Letters40(5): 988–993.
37.
GauthierPPonsVLetourneauA, et al. (2017) Combining population monitoring with habitat vulnerability to assess conservation status in populations of rare and endangered plants. Journal for Nature Conservation37: 83–95.
38.
GavishYO’ConnellJMarshCJ, et al. (2018) Comparing the performance of flat and hierarchical habitat/land-Cover classification models in a NATURA 2000 site. ISPRS Journal of Photogrammetry and Remote Sensing136: 1–12.
39.
GiganteDAcostaATRAgrilloE, et al. (2018) Habitat conservation in Italy: The state of the art in the light of the first European Red List of Terrestrial and Freshwater Habitats. Rendiconti Lincei. Scienze Fisiche e Naturali29(2): 251–265.
40.
GriffithsPNendelCPickertJ, et al. (2020) Towards national-scale characterization of grassland use intensity from integrated Sentinel-2 and Landsat time series. Remote Sensing of Environment238: 111124.
41.
GuisanAWeissSBWeissAD (1999) GLM versus CCA spatial modeling of plant species distribution. Plant Ecology143(1): 107–122.
42.
Halladin-DąbrowskaAKaniaAKopećD (2020) The t-SNE algorithm as a tool to improve the quality of reference data used in accurate mapping of heterogeneous non-forest vegetation. Remote Sensing12(1): 39.
43.
HeKSBradleyBACordAF, et al. (2015) Will remote sensing shape the next generation of species distribution models?Remote Sensing in Ecology and Conservation1(1): 4–18.
44.
HenglTde JesusJMHeuvelinkGBM, et al. (2017) SoilGrids250m: Global gridded soil information based on machine learning. PLoS ONE12(2): e0169748.
45.
Hubert-MoyLThibaultJFabreE, et al. (2019) Mapping grassland frequency using decadal MODIS 250 m time-series: Towards a national inventory of semi-natural grasslands. Remote Sensing11(24): 3041.
46.
HussonFLêSPagèsJ (2017) Exploratory Multivariate Analysis by Example Using R. Boca Raton, FL: CRC Press.
47.
IngladaJVincentAAriasM, et al. (2017) Operational high resolution land cover map production at the country scale using satellite image time series. Remote Sensing9(1): 95.
48.
JanssenJAMRodwellJSCriadoMG, et al. (2016) European Red List of Habitats: Part 2. Terrestrial and Freshwater Habitats. European Union.
49.
JarićIQuétierFMeinardY (2019) Procrustean beds and empty boxes: On the magic of creating environmental data. Biological Conservation237: 248–252.
50.
JomierRPoncetRMichezN (2019) How about habitats and Darwin Core?Biodiversity Information Science and Standards3: e39271.
51.
JongmanRHMücherCABunceRG, et al. (2019) A review of approaches for automated habitat mapping and their potential added value for biodiversity monitoring projects. Journal of Landscape Ecology12(3): 53–69.
52.
KallimanisASTouloumisKTzanopoulosJ, et al. (2015) Vegetation coverage change in the EU: patterns inside and outside Natura 2000 protected areas. Biodiversity and Conservation24(3): 579–591.
53.
KerneisEChevallierCPonsY (2007) Pasture production, water management and utilization conflicts in the fen lands of Western France: Is summer a key period?Fourrages191: 323–335.
54.
KhatamiRMountrakisGStehmanSV (2016) A meta-analysis of remote sensing research on supervised pixel-based land-cover image classification processes: General guidelines for practitioners and future research. Remote Sensing of Environment177: 89–100.
55.
KopeckýMČížkovአ(2010) Using topographic wetness index in vegetation ecology: Does the algorithm matter?Applied Vegetation Science13(4): 450–459.
56.
KosmidouVPetrouZBunceRGH, et al. (2014) Harmonization of the Land Cover Classification System (LCCS) with the General Habitat Categories (GHC) classification system. Ecological Indicators36: 290–300.
57.
KuhnM (2008) Caret package. Journal of Statistical Software28(5): 1–26.
58.
KuhnMJohnsonK (2013) Applied Predictive Modeling. New York: Springer-Verlag.
59.
LiuCWhiteMNewellG (2018) Detecting outliers in species distribution data. Journal of Biogeography45(1): 164–176.
60.
LombaAGuerraCAlonsoJ, et al. (2014) Mapping and monitoring High Nature Value farmlands: Challenges in European landscapes. Journal of Environmental Management143: 140–150.
61.
LouvelJGaudillatVPoncetL (2013) EUNIS, European Nature Information System, Système d’information européen sur la nature: classification des habitats: traduction française: habitats terrestres et d’eau douce. MEDDE. Paris: MNHN-DIREV-SPN. Available at: https://inpn.mnhn.fr/docs/ref_habitats/EUNIS_trad_francais.pdf. (accessed 15 April 2021)
62.
MarconciniMEschTBachoferF, et al. (2020) Digital Earth in Europe. In: GuoHGoodchildMFAnnoniA (eds) Manual of Digital Earth. Singapore: Springer, pp. 647–681.
63.
MaxwellAEWarnerTAFangF (2018) Implementation of machine-learning classification in remote sensing: An applied review. International Journal of Remote Sensing39(9): 2784–2817.
64.
MoranNNielandSKleinschmitB (2017) Combining machine learning and ontological data handling for multi-source classification of nature conservation areas. International Journal of Applied Earth Observation and Geoinformation54: 124–133.
65.
MoudrýVLecoursVGdulováK, et al. (2018) On the use of global DEMs in ecological modelling and the accuracy of new bare-earth DEMs. Ecological Modelling383: 3–9.
66.
MücherCAHennekensSMBunceRGH, et al. (2009) Modelling the spatial distribution of Natura 2000 habitats across Europe. Landscape and Urban Planning92(2): 148–159.
67.
MuukkonenPHeiskanenJ (2007) Biomass estimation over a large area based on standwise forest inventory data and ASTER and MODIS satellite data: A possibility to verify carbon inventories. Remote Sensing of Environment107(4): 617–624.
68.
PesaresiSManciniAQuattriniG, et al. (2020) Mapping Mediterranean forest plant associations and habitats with functional principal component analysis using Landsat 8 NDVI time series. Remote Sensing12(7): 1132.
69.
PlantureuxSPottierECarrèreP (2012) Permanent grassland: New challenges, new definitions?Fourrages211: 181–193.
70.
QGIS Development Team (2018) QGIS Version 2.18. 22. Geographic Information System.
71.
R. Core Team (2019) A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing, 2012. https://www.R-project.org.
72.
RaabCStrohHGTonnB, et al. (2018) Mapping semi-natural grassland communities using multi-temporal RapidEye remote sensing data. International Journal of Remote Sensing39(17): 5638–5659.
73.
RandinCFAshcroftMBBolligerJ, et al. (2020) Monitoring biodiversity in the Anthropocene using remote sensing in species distribution models. Remote Sensing of Environment239: 111626.
74.
RapinelSDusseuxPBouzilléJ-B, et al. (2018) Structural and functional mapping of geosigmeta in Atlantic coastal marshes (France) using a satellite time series. Plant Biosystems3504: 1–8.
75.
RapinelSMonyCLecoqL, et al. (2019) Evaluation of Sentinel-2 time-series for mapping floodplain grassland plant communities. Remote Sensing of Environment223: 115–129.
76.
RapinelSRozoCDelboscP, et al. (2020) Contribution of free satellite time-series images to mapping plant communities in the Mediterranean Natura 2000 site: The example of Biguglia Pond in Corse (France). Mediterranean Botany41(2): 181–191.
77.
RefaeilzadehPTangLLiuH (2009) Cross-validation. In: Encyclopedia of Database Systems. New York: Springer, pp. 532–538.
78.
RichardsJA (2013) Remote Sensing Digital Image Analysis: An Introduction. 5th ed. Berlin: Springer-Verlag.
79.
RocchiniDPetrasVPetrasovaA, et al. (2017) Open data and open source for remote sensing training in ecology. Ecological Informatics40: 57–61.
80.
RodwellJSEvansDSchaminéeJHJ (2018) Phytosociological relationships in European Union policy-related habitat classifications. Rendiconti Lincei. Scienze Fisiche e Naturali29(2): 237–249.
81.
SchusterCSchmidtTConradC, et al. (2015) Grassland habitat mapping by intra-annual time series analysis – comparison of RapidEye and TerraSAR-X satellite data. International Journal of Applied Earth Observation and Geoinformation34: 25–34.
82.
ShiferawHBewketWEckertS (2019) Performances of machine learning algorithms for mapping fractional cover of an invasive plant species in a dryland ecosystem. Ecology and Evolution9(5): 2562–2574.
83.
StroblCBoulesteixA-LZeileisA, et al. (2007) Bias in random forest variable importance measures: Illustrations, sources and a solution. BMC Bioinformatics8(1): 25.
84.
Suarez-SeoaneSJimenez-AlfaroBObesoJR (2020) Habitat-partitioning improves regional distribution models in multi-habitat species: A case study with the European bilberry. Biodiversity and Conservation29(3): 987–1008.
85.
SunDYuYGoldbergMD (2011) Deriving water fraction and flood maps from MODIS images using a decision tree approach. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing4(4): 814–825.
86.
TrullNBöhmMCarrJ (2018) Patterns and biases of climate change threats in the IUCN Red List. Conservation Biology32(1): 135–147.
87.
UllerudHABrynAHalvorsenR, et al. (2018) Consistency in land-cover mapping: Influence of field workers, spatial scale and classification system. Applied Vegetation Science21(2): 278–288.
88.
ValaviRElithJLahoz-MonfortJJ, et al. (2019) blockCV: An R package for generating spatially or environmentally separated folds for k-fold cross-validation of species distribution models. Methods in Ecology and Evolution10(2): 225–232.
89.
Van OijenMBellocchiGHöglindM (2018) Effects of climate change on grassland biodiversity and productivity: The need for a diversity of models. Agronomy8(2): 14.
90.
WeberDSchaepman-StrubGEckerK (2018) Predicting habitat quality of protected dry grasslands using Landsat NDVI phenology. Ecological Indicators91: 447–460.
91.
XieYShaZYuM (2008) Remote sensing imagery in vegetation mapping: A review. Journal of Plant Ecology1(1): 9–23.
92.
XuDChenBShenB, et al. (2019) The classification of grassland types based on object-based image analysis with multisource data. Rangeland Ecology and Management72(2): 318–326.
93.
YatesKLBouchetPJCaleyMJ, et al. (2018) Outstanding challenges in the transferability of ecological models. Trends in Ecology and Evolution33(10): 790–802.
94.
ZhangYMigliavaccaMPenuelasJ, et al. (2021) Advances in hyperspectral remote sensing of vegetation traits and functions. Remote Sensing of Environment252: 112121.
95.
ZhaoYLiuZWuJ (2020) Grassland ecosystem services: A systematic review of research advances and future directions. Landscape Ecology35(4): 793–814.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
