Land cover classification using multi‐temporal MERIS vegetation indices

The spectral, spatial, and temporal resolutions of Envisat's Medium Resolution Imaging Spectrometer (MERIS) data are attractive for regional‐ to global‐scale land cover mapping. Moreover, two novel and operational vegetation indices derived from MERIS data have considerable potential as discriminating variables in land cover classification. Here, the potential of these two vegetation indices (the MERIS global vegetation index (MGVI), MERIS terrestrial chlorophyll index (MTCI)) was evaluated for mapping eleven broad land cover classes in Wisconsin. Data acquired in the high and low chlorophyll seasons were used to increase inter‐class separability. The two vegetation indices provided a higher degree of inter‐class separability than data acquired in many of the individual MERIS spectral wavebands. The most accurate landcover map (73.2%) was derived from a classification of vegetation index‐derived data with a support vector machine (SVM), and was more accurate than the corresponding map derived from a classification using the data acquired in the original spectral wavebands.     

Evaluation of MODIS and NOAA AVHRR vegetation indices with in situ measurements in a semi‐arid environment

Much effort has been made in recent years to improve the spectral and spatial resolution of satellite sensors to develop improved vegetation indices reflecting surface conditions. In this study satellite vegetation indices from the Moderate Resolution Imaging Spectroradiometer (MODIS) and the Advanced Very High Resolution Radiometer (AVHRR) are evaluated against two years of in situ measurements of vegetation indices in Senegal. The in situ measurements are obtained using four masts equipped with self‐registrating multispectral radiometers designed for the same wavelengths as the satellite sensor channels. In situ measurements of the MODIS Normalized Difference Vegetation Index (NDVI) and AVHRR NDVI are equally sensitive to vegetation; however, the MODIS NDVI is consistently higher than the AVHRR NDVI. The MODIS Enhanced Vegetation Index (EVI) proved more sensitive to dense vegetation than both AVHRR NDVI and MODIS NDVI. EVI and NDVI based on the MODIS 16‐day constrained view angle maximum value composite (CV‐MVC) product captured the seasonal dynamics of the field observations satisfactorily but a standard 16‐day MVC product estimated from the daily MODIS surface reflectance data without view angle constraints yielded higher correlations between the satellite indices and field measurements (R ² values ranging from 0.74 to 0.98). The standard MVC regressions furthermore approach a 1?:?1 line with in situ measured values compared to the CV‐MVC regressions. The 16‐day MVC AVHRR data did not satisfactorily reflect the variation in the in situ data. Seasonal variation in the in situ measurements is captured reasonably with R ² values of 0.75 in 2001 and 0.64 in 2002, but the dynamic range of the AVHRR satellite data is very low—about a third to a half of the values from in situ measurements. Consequently the in situ vegetation indices were emulated much better by the MODIS indices than by the AVHRR NDVI.     

Land‐cover mixing and spectral vegetation indices

Vegetation indices have been widely used as indicators of seasonal and inter‐annual variations in vegetation caused by either human activities or climate, with the overall goal of observing and documenting changes in the Earth system. While existing satellite remote sensing systems, such as NASA's Multi‐angle Imaging SpectroRadiometer (MISR) and Moderate Resolution Imaging Spectroradiometer (MODIS), are providing improved vegetation index data products through correcting for the distortions in surface reflectance caused by atmospheric particles as well as ground covers below vegetation canopy, the impact of land‐cover mixing on vegetation indices has not been fully addressed. In this study, based on real image spectral samples for two‐component mixtures of forest and common nonforest land‐cover types directly extracted from a 1.1?km MISR image by referencing a 30?m land‐cover classification, the effect of land‐cover mixing on the Normalized Difference Vegetation Index (NDVI) and the Enhanced Vegetation Index (EVI) has been quantitatively evaluated. When the areal fraction of forest was lower than 80%, both NDVI and EVI varied greatly with mixed land‐cover types, although EVI varied less than NDVI. Such a phenomenon can cause errors in applications based on use of these vegetation indices. This study suggests that methods that reduce land‐cover mixing effects should be introduced when developing new spectral vegetation indices.     

Investigating the relationship between ground‐measured LAI and vegetation indices in an alpine meadow,north‐west China

A field campaign was carried out in the alpine meadow of Heihe River Basin, north‐west China on 11–15 July 2002. Several bio‐geophysical parameters such as leaf area index (LAI) were measured according to VALERI sampling procedures within 38 elementary sampling units (ESUs) in the 3 km×3 km ‘VALERI’ site. A quarter scene of Landsat 7 ETM+ with acquisition times close to the field campaign time was atmospherically and geographically corrected. Three kinds of spectral vegetation index maps including NDVI, SR and MSAVI in the sampling area were derived from the corrected ETM+ image. The two sets of LAI data measured with LAI‐2000 and TRAC instrument at the same site were inter‐compared. This is particularly meaningful for assessing the accuracy of LAI measurements. The relationships between the measured LAI and the three kinds of vegetation indices were also investigated. These comparisons found good relationships between the measured LAI and the different vegetation indices in most cases. Among them, NDVI seems the most promising estimator for extraction of LAI for the alpine meadow. In addition, the LAI‐2000 seems to perform better for LAI measurement in the alpine meadow than the TRAC instrument.     

Correlation between potato yield and MODIS‐derived vegetation indices

Monitoring of crop growth and forecasting its yield well before harvest is very important for crop and food management. Remote sensing images are capable of identifying crop health, as well as predicting its yield. Vegetation indices (VIs), such as the normalized difference vegetation index (NDVI), leaf area index (LAI) and fraction of photosynthetically active radiation (fPAR) calculated from remotely sensed data have been widely used to monitor crop growth and to predict crop yield. This study used 8 day TERRA MODIS reflectance data of 500 m resolution for the years 2005 to 2006 to estimate the yield of potato in the Munshiganj area of Bangladesh. The satellite data has been validated using ground truth data from fields of 50 farmers. Regression models are developed between VIs and field level potato yield for six administrative units of Munshiganj District. The yield prediction equations have high coefficients of correlation (R ²) and are 0.84, 0.72 and 0.80 for the NDVI, LAI and fPAR, respectively. These equations were validated by using data from 2006 to 2007 seasons and found that an average error of estimation is about 15% for the study region. It can be concluded that VIs derived from remote sensing can be an effective tool for early estimation of potato yield.     

Identification of drought‐vulnerable areas using NOAA AVHRR data

Drought is a recurring phenomenon in many parts of India, bringing significant water shortages, economic losses and adverse social consequences. The western regions of India (Rajasthan and Gujarat provinces) have suffered with severe droughts several times in the past. In this study meteorological and satellite data were used for monitoring drought in the southern part of Rajasthan. Monthly rainfall data from six stations were used to derive the Standardized Precipitation Index (SPI). The Advanced Very High Resolution Radiometer (AVHRR) onboard the National Oceanic and Atmospheric Administration (NOAA) series of satellite was used for calculating brightness temperature (BT), the Normalized Difference Vegetative Index (NDVI) and the Water Supplying Vegetation Index (WSVI). BT was converted to the Vegetation Condition Index (VCI) and the Temperature Condition Index (TCI), which are useful indices for the estimation of vegetation health and drought monitoring. The analysis was carried out for a period of four years (2002–2005) and from the SPI analysis it was observed that 2002 was a drought year. On the basis of the satellite‐based indices, the study area was divided into categories of extreme, severe, moderate and slight drought and normal condition. We found that in 2002 all of the area under study was affected by drought with greater intensity, mostly classed as extreme and severe drought conditions. An analysis was carried out of the study area divided into four zones on the basis of rainfall distribution, soil characteristics, cropping patterns and other physical characteristics. This analysis revealed that zone 1 was the most drought‐prone area in all four years; zone 4 was the next area most affected by severe drought, followed by zones 2 and 3, which were less affected by drought conditions.     

Evaluation of AVHRR PAL and GIMMS 10‐day composite NDVI time series products using SPOT‐4 vegetation data for the African continent

Global 8 km resolution AVHRR (advanced very high resolution radiometer) NDVI (normalized difference vegetation index) 10‐day composite data sets have been used for numerous local to global scale vegetation time series studies during recent years. AVHRR Pathfinder (PAL) NDVI was available from 1981 until 2001, and the new AVHRR GIMMS NDVI was available from 1981 to the present time. A number of aspects potentially introduce noise in the NDVI data set due to the AVHRR sensor design and data processing. NDVI from SPOT‐4 VGT data is considered an improvement over AVHRR, and for this reason it is important to examine how and if the differences in sensor design and processing influence continental scale NDVI composite products. In this study, the quality of these AVHRR NDVI time series are evaluated by the continental scale 1 km resolution SPOT‐4 vegetation (VGT) 10‐day composite (S10) NDVI data. Three years of AVHRR PAL (1998–2000) and seven years of GIMMS (1998–2004) have been compared to 8 km resampled SPOT‐4 VGT (1998–2004) data. The dynamic range of SPOT‐4 VGT NDVI tends to be higher than the AVHRR PAL NDVI, whereas there is an exact match between AVHRR GIMMS NDVI and SPOT‐4 VGT NDVI. Ortho‐regression analysis on annually integrated values of AVHRR PAL/GIMMS and SPOT‐4 VGT on a continental scale reveals high correlations amongst the AVHRR and the SPOT data set, with lowest RMSE (root mean square error) on the GIMMS/SPOT‐4 VGT compared to the PAL/SPOT‐4 VGT.

Analyses on decade data likewise show that a linear relation exists between Spot‐4 VGT NDVI and the two AVHRR composite products; GIMMS explaining most of the Spot‐4 VGT NDVI variance compared to PAL. These results show that the AVHRR GIMMS NDVI is more consistent with Spot‐4 VGT NDVI compared to AVHRR PAL versus Spot‐4 VGT NDVI (in terms of RMSE and dynamic range) and can therefore be considered the more accurate long time AVHRR data record. Analyses performed on monthly maximum composites and decade composite data, however, reveal intra‐annual variations in the correlation between SPOT‐4 VGT and the two AVHRR data sets, which are attributed to different cloud masking algorithms. The SPOT‐4 VGT cloud‐screening algorithm is insufficient, thereby suppressing the rainy season NDVI.     

Modelling land‐cover types using multiple endmember spectral mixture analysis in a desert city

Spectral mixture analysis is probably the most commonly used approach among sub‐pixel analysis techniques. This method models pixel spectra as a linear combination of spectral signatures from two or more ground components. However, spectral mixture analysis does not account for the absence of one of the surface features or spectral variation within pure materials since it utilizes an invariable set of surface features. Multiple endmember spectral mixture analysis (MESMA), which addresses these issues by allowing endmembers to vary on a per pixel basis, was employed in this study to model Landsat ETM+ reflectance in the Phoenix metropolitan area. Image endmember spectra of vegetation, soils, and impervious surfaces were collected with the use of a fine resolution Quickbird image and the pixel purity index. This study employed 204 (3×17×4) total four‐endmember models for the urban subset and 96 (6×6×2×4) total five‐endmember models for the non‐urban subset to identify fractions of soil, impervious surface, vegetation and shade. The Pearson correlation between the fraction outputs from MESMA and reference data from Quickbird 60 cm resolution data for soil, impervious, and vegetation were 0.8030, 0.8632, and 0.8496 respectively. Results from this study suggest that the MESMA approach is effective in mapping urban land covers in desert cities at sub‐pixel level.     

Modelling the population density of China at the pixel level based on DMSP/OLS non‐radiance‐calibrated night‐time light images

The spatial distribution of population density is crucial for analysing the relationships among economic growth, environmental protection and resource use. In this study we simulated China's population density in 1998 at 1 km×1 km resolution by integrating DMSP/OLS non‐radiance‐calibrated night‐time images, SPOT/VGT 10‐day maximum NDVI composite, population census data and vector county boundaries. Population density, both inside and outside of light patches, was estimated for four types of counties, which were classified according to their light characteristics. The model for estimating population density inside the light patches was developed based on a significant correlation between light intensity and population, while the model for estimating population density outside of light patches was constructed by combining Coulomb's law with electric field superposition principle. Our method was simpler and less expensive than existing methods for spatializing population density. The results were consistent with other estimates but exhibited more spatial heterogeneity and richer information.     

Implementation of GOES‐based land surface temperature diurnal cycle to AVHRR

To enable frequent estimates of land surface temperature (LST) from satellite measurements, and to characterize the land surface temperature diurnal (LSTD) cycle, two new LST retrieval algorithms are applied to observations from the Geostationary Operational Environmental Satellite (GOES). Evaluation against the atmospheric radiation measurement (ARM) observations indicates that LST can be determined from the real‐time GOES‐8 observations within r.m.s. accuracy of about 2 K. In order to combine the advantages of geostationary and polar orbiting instruments, the LSTD estimated from GOES can be incorporated into LST retrievals from polar orbiting imager National Oceanic and Atmospheric Administration/Advanced Very High Resolution Radiometer (NOAA/AVHRR) using a newly proposed fitting algorithm, with r.m.s. errors close to those obtained directly from GOES‐8.     

Spectral vegetation indices for estimating shrub cover,green phytomass and leaf turnover in a sedge‐shrub tundra

Using field observations, we determined the relationships between spectral indices and the shrub ratio, green phytomass and leaf turnover of a sedge-shrub tundra community in the Arctic National Wildlife Refuge, Alaska, USA. We established a 50‐m × 50‐m plot (69.73°N 143.62°W) located on a floodplain of the refuge. The willow shrub (Salix lanata) and sedge (Carex bigelowii) dominated the plot vegetation. In July to August 2007, we established ten 0.5‐m × 0.5‐m quadrats on both shrub‐covered ground (shrub quadrats) and on ground with no shrubs (sedge quadrats). The shrub ratio was more strongly correlated with the normalized difference vegetation index (NDVI, R2 of 0.57) than the normalized difference infrared index (NDII), the soil-adjusted vegetation index (SAVI) or the enhanced vegetation index (EVI). On the other hand, for both green phytomass and leaf turnover, the strongest correlation was with NDII (R ² of 0.63 and 0.79, respectively).     

Estimating tree‐cover change in Australia: challenges of using the MODIS vegetation index product

Time series of the vegetation index product MOD13Q1 from the Moderate Resolution Imagery Spectroradiometer (MODIS) were assessed for estimating tree foliage projective cover (FPC) and cover change from 2000 to 2006. The MOD13Q1 product consists of the enhanced vegetation index (EVI) and the normalized difference vegetation index (NDVI). There were four challenges in using the MOD13Q1 product to derive tree FPC: assessing the impact of the sensor's varying view geometry on the vegetation index values; decoupling tree and grass cover contributions to the vegetation index signal; devising a method to relate the temporally composited vegetation index pixels to Lidar estimates of tree FPC for calibration; and estimating the accuracy of the FPC and FPC change measurements using independently derived Lidar, Landsat and MODIS cover estimates. The results show that, for complex canopies, the varying view geometry influenced the vegetation indices. The EVI was more sensitive to the view angle than the NDVI, indicating that it is sensitive to vegetation structure. An existing time series method successfully extracted the evergreen vegetation index signal while simultaneously minimizing the impact of varying view geometry. The vegetation indices were better suited to monitoring tree cover change than deriving accurate single‐date estimates of cover at regional to continental scales. The EVI was more suited to monitoring change in high‐biomass regions (cover >50%) where the NDVI begins to saturate.     

Sub‐pixel reflectance unmixing in estimating vegetation water content and dry biomass of corn and soybeans cropland using normalized difference water index (NDWI) from satellites

Estimating vegetation cover, water content, and dry biomass from space plays a significant role in a variety of scientific fields including drought monitoring, climate modelling, and agricultural prediction. However, getting accurate and consistent measurements of vegetation is complicated very often by the contamination of the remote sensing signal by the atmosphere and soil reflectance variations at the surface. This study used Landsat TM/ETM+ and MODIS data to investigate how sub‐pixel atmospheric and soil reflectance contamination can be removed from the remotely sensed vegetation growth signals. The sensitivity of spectral bands and vegetation indices to such contamination was evaluated. Combining the strengths of atmospheric models and empirical approaches, a hybrid atmospheric correction scheme was proposed. With simplicity, it can achieve reasonable accuracy in comparison with the 6S model. Insufficient vegetation coverage information and poor evaluation of fractional sub‐pixel bare soil reflectance are major difficulties in sub‐pixel soil reflectance unmixing. Vegetation coverage was estimated by the Normalized Difference Water Index (NDWI). Sub‐pixel soil reflectance was approximated from the nearest bare soil pixel. A linear reflectance mixture model was employed to unmix sub‐pixel soil reflectance from vegetation reflectance. Without sub‐pixel reflectance contamination, results demonstrate the true linkage between the growth of sub‐pixel vegetation and the corresponding change in satellite spectral signals. Results suggest that the sub‐pixel soil reflectance contamination is particularly high when vegetation coverage is low. After unmixing, the visible and shortwave infrared reflectances decrease and the near‐infrared reflectances increase. Vegetation water content and dry biomass were estimated using the unmixed vegetation indices. Superior to the NDVI and the other NDWIs, the SWIR (1650 nm) band‐based NDWI showed the best overall performance. The use of the NIR (1240 nm), which is a unique band of MODIS, was also discussed.     

An extended AVHRR 8‐km NDVI dataset compatible with MODIS and SPOT vegetation NDVI data

Daily daytime Advanced Very High Resolution Radiometer (AVHRR) 4‐km global area coverage data have been processed to produce a Normalized Difference Vegetation Index (NDVI) 8‐km equal‐area dataset from July 1981 through December 2004 for all continents except Antarctica. New features of this dataset include bimonthly composites, NOAA‐9 descending node data from August 1994 to January 1995, volcanic stratospheric aerosol correction for 1982–1984 and 1991–1993, NDVI normalization using empirical mode decomposition/reconstruction to minimize varying solar zenith angle effects introduced by orbital drift, inclusion of data from NOAA‐16 for 2000–2003 and NOAA‐17 for 2003–2004, and a similar dynamic range with the MODIS NDVI. Two NDVI compositing intervals have been produced: a bimonthly global dataset and a 10‐day Africa‐only dataset. Post‐processing review corrected the majority of dropped scan lines, navigation errors, data drop outs, edge‐of‐orbit composite discontinuities, and other artefacts in the composite NDVI data. All data are available from the University of Maryland Global Land Cover Facility (http://glcf.umiacs.umd.edu/data/gimms/).     

Inter‐comparison of AVHRR‐based fire susceptibility indicators for the Mediterranean ecosystems of southern Italy

The existing parameters based on Advanced Very High Resolution Radiometer (AVHRR) data and devised for fire susceptibility estimation (FSE) were applied in different regions of southern Italy. Their performances were evaluated by using a wide data sample of National Oceanic and Atmospheric Administration (NOAA)‐12 and ‐14 summer imagery acquired from 1996 to 1999. In order to test their effectiveness, each different parameter was tested by applying the same thresholding procedure on every individual parameter independent from its pre‐established classification by the authors. The evaluation was performed by comparing fire archives (provided by the Italian National Forestry Service) to the results obtained. The most satisfactory results were obtained by using a combination of Normalized Difference Vegetation Index (NDVI) and thermal channels. These experimental analyses confirmed that improvements were achieved from methods that combine NDVI with thermal channels, in particular when the two indicators are first classified separately and then combined in a single index. This allows a valid reduction of the number of pixels classified as fire vulnerable compared with methods that apply a joined classification of NDVI and surface temperature (T _s). Finally, the use of the AVHRR channel 3 (thermal data) proved to be more effective than T _s. Such evaluations are a valuable support for the assessment of how satellite‐based parameters can be profitably used to improve the estimation of fire susceptibility in operational applications. Our findings can be directly extended to other Mediterranean‐like ecosystems.     

Comparing farmer‐based and satellite‐derived deforestation estimates in the Amazon basin using a hybrid classifier

The Amazon basin remains a major hotspot of tropical deforestation, presenting a clear need for timely, accurate and consistent data on forest cover change. We assessed the utility of a hybrid classification technique, iterative guided spectral class rejection (IGSCR), for accurately mapping Amazonian deforestation using annual imagery from the Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper Plus (ETM+) from 1992 to 2002. The mean overall accuracy of the 11 annual classifications was 95% with a standard deviation of 1.4%, and z‐score analysis revealed that all classifications were significant at the 0.05 level. The IGSCR thus seems inherently suitable for monitoring forest cover in the Amazon. The resulting classifications were sufficiently accurate to assess preliminarily the magnitude and causes of discrepancies between farmer‐reported and satellite‐based estimates of deforestation at the household level using a sample of 220 farms in Rôndonia mapped in the field in 1992 and 2002. The field‐ and satellite‐derived estimates were significantly different only at the 0.10 level for the 220 farms studied, with the satellite‐derived deforestation estimates 8.9% higher than estimates derived from in situ survey methods. Some of this difference was due to a tendency of farmers to overestimate the amount of forest within their property in our survey. Given the objectivity and reduced expense of satellite‐based deforestation monitoring, we recommend that it be an integral part of household‐level analysis of the causes, patterns and processes of deforestation.     

Modelling spatial‐temporal change of Poyang Lake using multitemporal Landsat imagery

Poyang Lake is a seasonal lake, exchanging water with the lower branch of the Yangtze River. During the spring and summer flooding season it inundates a large area while in the winter it shrinks considerably, creating a large tract of marshland for wild migratory birds. A better knowledge of the water coverage duration and the beginning and ending dates for the vast range of marshlands surrounding the lake is important for the measurement, modelling and management of marshland ecosystems. In addition, the abundance of a special type of snail (Oncomelania hupensis), the intermediate host of parasite schistosome (Schistosoma japonicum) in this region, is also heavily dependent on the water coverage information. However, there is no accurate digital elevation model (DEM) for the lake bottom and the inundated marshland, nor is there sufficient water level information over this area. In this study, we assess the feasibility of the use of multitemporal Landsat images for mapping the spatial‐temporal change of Poyang Lake water body and the temporal process of water inundation of marshlands. Eight cloud‐free Landsat Thematic Mapper images taken during a period of one year were used in this study. We used the normalized difference water index (NDWI) and the modified normalized difference water index (MNDWI) methods to map water bodies. We then examined the annual spatial‐temporal change of the Poyang Lake water body. Finally we attempted to obtain the duration of water inundation of marshlands based on the temporal sequence of water extent determined from the Landsat images. The results showed that although the images can be used to capture the snapshots of water coverage in this area, they are insufficient to provide accurate estimation of the spatial‐temporal process of water inundation over the marshlands through linear interpolation.     

Estimation of chlorophyll‐a concentration in Lake Tai,China using in situ hyperspectral data

Reflectance spectra of water in Lake Tai of East China were measured at 28 monitoring stations with an ASD FieldSpec spectroradiometer at an interval of 1.58 nm over five days in each month from June to August of 2004. Water samples collected at these stations were analyzed in the laboratory to determine chlorophyll‐a (chl‐a) concentration. Twenty‐eight spectral reflectance curves were standardized and correlated with chl‐a concentration. Examination of these curves reveals a peak reflectance at 719 nm. Chl‐a concentration level in the Lake was most closely correlated with the reflectance near 700 nm. If regressed against the reflectance at the wavelength of 667 nm (R ₆₆₇), chl‐a concentration was not accurately estimated at R ² = 0.494. Accuracy of estimation was improved to R ² = 0.817 using the maximum reflectance. A higher accuracy of 0.837 was achieved using the peak reflectance at 719 nm (R ₇₁₉) because it does not drift with the level of chl‐a concentration. The highest accuracy of estimation was achieved at R ² = 0.868 using R ₇₁₉/R ₆₆₇.     

Differentiation of semi‐arid vegetation types based on multi‐angular observations from MISR and MODIS

Mapping accurately vegetation type is one of the main challenges for monitoring arid and semi‐arid grasslands with remote sensing. The multi‐angle approach has been demonstrated to be useful for mapping vegetation types in deserts. The current paper presents a study on the use of directional reflectance derived from two sensor systems, using two different models to analyse the data and two different classifiers as a means of mapping vegetation types. The multiangle imaging spectroradiometer (MISR) and the moderate resolution imaging spectroradiometer (MODIS) provide multi‐spectral and angular, off‐nadir observations. In this study, we demonstrate that reflectance from MISR observations and reflectance anisotropy patterns derived from MODIS observations are capable of working together to increase classification accuracy. The patterns are described by parameters of the modified Rahman–Pinty–Verstraete and the RossThin‐LiSparseMODIS bidirectional reflectance distribution function (BRDF) models. The anisotropy patterns derived from MODIS observations are highly complementary to reflectance derived from radiances observed by MISR. Support vector machine algorithms exploit the information carried by the same data sets more effectively than the maximum likelihood classifier.     

Updating land cover classification using a rule‐based decision system

This paper proposes a land cover classification methodology in agricultural contexts that provides satisfactory results with a single satellite image per year acquired during the growing period. Our approach incorporates ancillary data such as cropping history (inter‐annual crop rotations), context (altitude, soil type) and structure (parcels size and shape) to compensate for the lack of radiometric data resulting from the use of a single image. The originality of the proposed method resides in the three successive steps used: S1: per‐pixel classification of a single SPOT XS image with a restricted number of land cover classes (RL) chosen to ensure good accuracy; S2: conversion of RLs into a per‐parcel classification system using ancillary parcel boundaries; and S3: parcel allocation using exhaustive land cover classes (EL) and its refinement through the application of decision rules. The method was tested on a 120?km² area (Sousson river basin, Gers, France) where exhaustive knowledge of land cover for two successive years allowed complete validation of our method. It allocated 87% of the parcels with a 75% accuracy rate according to the exhaustive list (EL). This is a satisfactory result obtained with one SPOT XS image in a small agricultural parcel context.