Mapping paddy rice fields by applying machine learning algorithms to multi-temporal Sentinel-1A and Landsat data

Sentinel-1A synthetic aperture radar (SAR) data present an opportunity for acquiring crop information without restrictions caused by weather and illumination conditions, at a spatial resolution appropriate for individual rice fields and a temporal resolution sufficient to capture the growth profiles of different crop species. This study investigated the use of multi-temporal Sentinel-1A SAR data and Landsat-derived normalized difference vegetation index (NDVI) data to map the spatial distribution of paddy rice fields across parts of the Sanjiang plain, in northeast China. The satellite sensor data were acquired throughout the rice crop-growing season (May–October). A co-registered set of 10 dual polarization (VH/VV) SAR and NDVI images depicting crop phenological development were used as inputs to Support Vector Machine (SVM) and Random Forest (RF) machine learning classification algorithms in order to map paddy rice fields. The results showed a significant increase in overall classification when the NDVI time-series data were integrated with the various combinations of multi-temporal polarization channels (i.e. VH, VV, and VH/VV). The highest classification accuracies overall (95.2%) and for paddy rice (96.7%) were generated using the RF algorithm applied to combined multi-temporal VH polarization and NDVI data. The SVM classifier was most effective when applied to the dual polarization (i.e. VH and VV) SAR data alone and this generated overall and paddy rice classification accuracies of 91.6% and 82.5%, respectively. The results demonstrate the practicality of implementing RF or SVM machine learning algorithms to produce 10 m spatial resolution maps of paddy rice fields with limited ground data using a combination of multi-temporal SAR and NDVI data, where available, or SAR data alone. The methodological framework developed in this study is apposite for large-scale implementation across China and other major rice-growing regions of the world.     

基于Sentinel-2的UNVI植被指数及性能对比研究

作物精准识别和分类是农业遥感检测的重要内容,对作物长势监测以及估产十分重要。以美国混合农业带为研究区,基于Sentinel-2时间序列影像,根据其传感器响应函数计算了针对Sentinel-2的通用归一化植被指数（Universal Normalized Vegetation Index,UNVI）,并通过两个对比实验,分析UNVI等6个指数在作物精准分类中的性能。实验一以JM（Jeffries-Matusita）距离为指标对不同作物类别之间的可分性进行分析,结果表明UNVI优于NDVI、EVI、WDRVI、NDre1和NDWI指数,在玉米和棉花、玉米和水稻、玉米和水稻的区分上,UNVI优于其他指数区分能力相当,但在其余的作物组合上如棉花和水稻,NDVI等指数则无法将其很好的区分,此时UNVI指数依然可以表现出较好的区分能力;实验二对6种时间序列指数特征分别使用随机森林和支持向量机进行作物分类,结果表明UNVI指数的总体精度和Kappa系数最高,其次是NDre1指数和WDRVI指数,EVI的总体精度和Kappa系数最低,这表明UNVI比其他6个指数更好地区分了研究区大豆、玉米、棉花和水稻等4种主要作物。综上,基于Sentinel-2时间序列的UNVI指数在进行作物分类时与其他5种遥感植被指数相比,具有较大的优势,UNVI可为农作物长势分析和作物估产研究等农业研究和应用的可选植被指数。     

Large-area crop mapping using time-series MODIS 250 m NDVI data: An assessment for the U.S. Central Great Plains

Improved and up-to-date land use/land cover (LULC) data sets that classify specific crop types and associated land use practices are needed over intensively cropped regions such as the U.S. Central Great Plains, to support science and policy applications focused on understanding the role and response of the agricultural sector to environmental change issues. The Moderate Resolution Imaging Spectroradiometer (MODIS) holds considerable promise for detailed, large-area crop-related LULC mapping in this region given its global coverage, unique combination of spatial, spectral, and temporal resolutions, and the cost-free status of its data. The objective of this research was to evaluate the applicability of time-series MODIS 250 m normalized difference vegetation index (NDVI) data for large-area crop-related LULC mapping over the U.S. Central Great Plains. A hierarchical crop mapping protocol, which applied a decision tree classifier to multi-temporal NDVI data collected over the growing season, was tested for the state of Kansas. The hierarchical classification approach produced a series of four crop-related LULC maps that progressively classified: 1) crop/non-crop, 2) general crop types (alfalfa, summer crops, winter wheat, and fallow), 3) specific summer crop types (corn, sorghum, and soybeans), and 4) irrigated/non-irrigated crops. A series of quantitative and qualitative assessments were made at the state and sub-state levels to evaluate the overall map quality and highlight areas of misclassification for each map.The series of MODIS NDVI-derived crop maps generally had classification accuracies greater than 80%. Overall accuracies ranged from 94% for the general crop map to 84% for the summer crop map. The state-level crop patterns classified in the maps were consistent with the general cropping patterns across Kansas. The classified crop areas were usually within 1-5% of the USDA reported crop area for most classes. Sub-state comparisons found the areal discrepancies for most classes to be relatively minor throughout the state. In eastern Kansas, some small cropland areas could not be resolved at MODIS' 250 m resolution and led to an underclassification of cropland in the crop/non-crop map, which was propagated to the subsequent crop classifications. Notable regional areal differences in crop area were also found for a few selected crop classes and locations that were related to climate factors (i.e., omission of marginal, dryland cropped areas and the underclassification of irrigated crops in western Kansas), localized precipitation patterns (overclassification of irrigated crops in northeast Kansas), and specific cropping practices (double cropping in southeast Kansas).     

Investigating collection 4 versus collection 5 MODIS 250 m NDVI time-series data for crop separability in Kansas,USA

The primary objective of this research was to analyse collection 5 versus collection 4 time-series normalized difference vegetation index (NDVI) data from the Moderate Resolution Imaging Spectroradiometer (MODIS) 250 m for the purpose of separating crop types. Using extensive ground reference data from the state of Kansas in the central USA, NDVI value profiles were extracted from different collection versions for 2001 (collections 4 and 5) and 2005 (collection 5 only). Phenological curves for all crops and all data sets were created and visually inspected. Jeffries–Matusita (J-M) distance statistical analysis was performed to assess crop separability. Contrary to expectations, collection 5 time-series MODIS 250 m NDVI data were found to be inferior to collection 4 with respect to crop separability. Specifically, collection 4 data exhibited a greater dynamic range across the growing seasons of the various crop types, and this discriminatory advantage was supported by J-M distance analysis. Though the analysis did not suggest reasons for the outcome, it corroborates the conclusion of the only other similar study in the literature comparing data from collections 4 and 5. Considering the pervasive use of these data for land-cover mapping, it is recommended that MODIS NDVI data from collection 4 should be used where possible for crop type mapping in agricultural regions with climate, geography, and crops similar to Kansas.     

Analysis of multi-temporal SPOT NDVI images for small-scale land-use mapping

Land-use information is required for a number of purposes such as to address food security issues, to ensure the sustainable use of natural resources and to support decisions regarding food trade and crop insurance. Suitable land-use maps often either do not exist or are not readily available. This article presents a novel method to compile spatial and temporal land-use data sets using multi-temporal remote sensing in combination with existing data sources. Satellite Pour l'Observation de la Terre (SPOT)-Vegetation 10-day composite normalized difference vegetation index (NDVI) images (1998–2002) at 1km² resolution for a part of the Nizamabad district, Andhra Pradesh, India, were linked with available crop calendars and information about cropping patterns. The NDVI images were used to stratify the study area into map units represented by 11 distinct NDVI classes. These were then related to an existing land-cover map compiled from high resolution Indian Remote Sensing (IRS)-images (Liss-III on IRS-1C), reported crop areas by sub-district and practised crop calendar information. This resulted in an improved map containing baseline information on both land cover and land use. It is concluded that each defined NDVI class represents a varying but distinct mix of land-cover classes and that the existing land-cover map consists of too many detailed ‘year-specific’ features. Four groups of the NDVI classes present in agricultural areas match well with four categories of practised crop calendars. Differences within a group of NDVI classes reveal area specific variations in cropping intensities. The remaining groups of NDVI classes represent other land-cover complexes. The method illustrated in this article has the potential to be incorporated into remote sensing and Geographical Information System (GIS)-based drought monitoring systems.     

Assessing crop residue cover using shortwave infrared reflectance

Management of crop residues is an important consideration for reducing soil erosion and increasing soil organic carbon. Current methods of measuring residue cover are inadequate for characterizing the spatial variability of residue cover over large fields. The objectives of this research were to determine the spectral reflectance of crop residues and soils and to assess the limits of discrimination that can be expected in mixed scenes. Spectral reflectances of dry and wet crop residues plus three diverse soils were measured over the 400-2400 nm wavelength region. Reflectance values for scenes with varying proportions of crop residues and soils were simulated. Additional spectra of scenes with mixtures of crop residues, green vegetation, and soil were also acquired in corn, soybean, and wheat fields with different tillage treatments. The spectra of dry crop residues displayed a broad absorption feature near 2100 nm, associated with cellulose-lignin, that was absent in spectra of soils. Crop residue cover was linearly related (r²=0.89) to the Cellulose Absorption Index (CAI), which was defined as the relative depth of this absorption feature. Green vegetation cover in the scene attenuated CAI, but was linearly related to the Normalized Difference Vegetation Index (NDVI, r²=0.93). A novel method is proposed to assess soil tillage intensity classes using CAI and NDVI. Regional surveys of soil conservation practices that affect soil carbon dynamics may be feasible using advanced multispectral or hyperspectral imaging systems.     

A comparison of MODIS 250-m EVI and NDVI data for crop mapping: a case study for southwest Kansas

Multi-temporal vegetation index (VI) data from the Moderate Resolution Imaging Spectroradiometer (MODIS) are becoming widely used for large-area crop classification. Most crop-mapping studies have applied enhanced vegetation index (EVI) data from MODIS instead of the more traditional normalized difference vegetation index (NDVI) data because of atmospheric and background corrections incorporated into EVI's calculation and the index's sensitivity over high biomass areas. However, the actual differences in the classification results using EVI versus NDVI have not been thoroughly explored. This study evaluated time-series MODIS 250-m EVI and NDVI for crop-related land use/land cover (LULC) classification in the US Central Great Plains. EVI- and NDVI-derived maps classifying general crop types, summer crop types and irrigated/non-irrigated crops were produced for southwest Kansas. Qualitative and quantitative assessments were conducted to determine the thematic accuracy of the maps and summarize their classification differences. For the three crop maps, MODIS EVI and NDVI data produced equivalent classification results. High thematic accuracies were achieved with both indices (generally ranging from 85% to 90%) and classified cropping patterns were consistent with those reported for the study area (> 0.95 correlation between the classified and USDA-reported crop areas). Differences in thematic accuracy (< 3% difference), spatially depicted patterns (> 90% pixel-level thematic agreement) and classified crop areas between the series of EVI- and NDVI-derived maps were negligible. Most thematic disagreements were restricted to single pixels or small clumps of pixels in transitional areas between cover types. Analysis of MODIS composite period usage in the classification models also revealed that both VIs performed equally well when periods from a specific growing season phase (green, peak or senescence) were heavily utilized to generate a specific crop map.     

Classification of rice cropping systems by empirical mode decomposition and linear mixture model for time-series MODIS 250 m NDVI data in the Mekong Delta,Vietnam

Estimating the area of rice planting is vital for production prediction. This study utilizes time-series MODIS NDVI data from 2002 to 2007 to discriminate rice cropping systems in the Mekong Delta (MD), Vietnam. Data are processed using Empirical Mode Decomposition (EMD) and the Linear Mixture Model (LMM). Various spatial and non-spatial data are also collected for accuracy validation. The results indicate that EMD acts as a well-fitted filter for noise reduction of the time-series NDVI data. The classification results derived from the LMM for 2002 showed an overall classification accuracy of 71.6% and a Kappa coefficient of 0.6. The provincial level area estimates were strongly correlated with the rice statistics. An examination of the change in cropping patterns between 2002 and 2007 showed that 29.0% of the triple irrigated-rice cropping systems had been changed to double irrigated-rice cropping systems and that 12.0% and 9.0% of the double irrigated and rainfed-rice cropping systems, respectively, had been changed to triple rice cropping systems. These changes were verified by visual comparisons with Landsat images.     

NOAA-AVHRR NDVI decomposition and subpixel classification using linear mixing in the Argentinean Pampa

Crop Normalized Difference Vegetation Index (NDVI) time profiles and crop acreage estimates were derived from the application of linear mixture modelling to Advanced Very High Resolution Radiometer (AVHRR) data over a test area in the southern part of the Pampa region, Argentina. Bands 1 and 2 from seven AVHRR scenes (June to January 1991) were combined to produce fraction images of winter crops, summer crops and pastures. A Landsat Thematic Mapper (TM) scene of the region was classified and superimposed to the AVHRR Local Area Coverage (LAC) data by means of a correlation technique. Each class signature was extracted by regressing the AVHRR response on the cover types proportions, estimated from Landsat-TM data, over sets of calibration windows. The crop NDVI profiles were hence derived from the class signatures in bands 1 and 2. These profiles appeared consistent with the cover types, but variability depending on the set of windows was noted. The assessment of the class signatures was indirectly accomplished through the subpixel classifications of the AVHRR data, performed using the different sets of class spectra. Although some discrepancies between AVHRR and Landsat–TM estimates were observed at the individual window level, the classification results compared quite well on a regional scale with Landsat–TM estimates: crop acreage was estimated to an overall accuracy ranging from 89 to 95 per cent according to the spectra used in the classification. Definitely, the proposed methodology should permit a better exploitation of the temporal resolution of AVHRR data in both the areas of yield prediction and vegetation classification. Furthermore, the perational application of such a methodology for crop monitoring will undoubtedlybe facilitated with the coming sensor systems such as the ModerateResolution Imaging Spectroradiometer (MODIS), the SPOT Vegetation Monitoring Instrument or the ‘Satelite Argentino Cientifico’ (SAC–C).     

长期三熟制下谷物产量和土壤氮磷钾及有机质性状的演化

自1973年10月起在麦稻稻三熟制下连续进行了26年定位试验。结果表明,长期配施N肥和猪厩肥可以实现谷物的持续增产,保持土壤有机质的持续增长,并培肥土壤,有机质达到60g/kg的高水平。稻田土壤的有机质数量演化可以幂指数方程表示。土壤N素平衡有余,全N大幅度增加,而有效N的增长差距巨大。全K和速效K显著下降,缓效K变化不大。全P在平衡或盈余时有效P大幅提高。     

Land cover mapping of large areas using chain classification of neighboring Landsat satellite images

Satellite imagery is the major data source for regional to global land cover maps. However, land cover mapping of large areas with medium-resolution imagery is costly and often constrained by the lack of good training and validation data. Our goal was to overcome these limitations, and to test chain classifications, i.e., the classification of Landsat images based on the information in the overlapping areas of neighboring scenes. The basic idea was to classify one Landsat scene first where good ground truth data is available, and then to classify the neighboring Landsat scene using the land cover classification of the first scene in the overlap area as training data. We tested chain classification for a forest/non-forest classification in the Carpathian Mountains on one horizontal chain of six Landsat scenes, and two vertical chains of two Landsat scenes each. We collected extensive training data from Quickbird imagery for classifying radiometrically uncorrected data with Support Vector Machines (SVMs). The SVMs classified 8 scenes with overall accuracies between 92.1% and 98.9% (average of 96.3%). Accuracy loss when automatically classifying neighboring scenes with chain classification was 1.9% on average. Even a chain of six images resulted only in an accuracy loss of 5.1% for the last image compared to a reference classification from independent training data for the last image. Chain classification thus performed well, but we note that chain classification can only be applied when land cover classes are well represented in the overlap area of neighboring Landsat scenes. As long as this constraint is met though, chain classification is a powerful approach for large area land cover classifications, especially in areas of varying training data availability.     

Scene selection and the use of NASA's global orthorectified Landsat dataset for land cover and land use change monitoring

This study examines the utility of NASA's circa 1990 and circa 2000 global orthorectified Landsat dataset for land cover and land use change mapping and monitoring across Africa. This is achieved by comparing the temporal and spatial variation of NDVI, measured independently by the NOAA-AVHRR at the time of Landsat scene acquisition, against the seasonal mean for each Landsat scene extent. Decadal sequences of drift-corrected NOAA-AVHRR imagery were used to calculate NDVI means and standard deviations for the periods covered by the scenes composing the c.1990 and c.2000 Landsat datasets. The specific NOAA-AVHRR NDVI values at the acquisition date of each individual Landsat scene were also calculated and the differences, both from the mean and scaled by standard deviation, were mapped for the Landsat scene footprints in the c.1990 and c.2000 datasets. The resulting maps show the temporal position of each Landsat scene within the seasonal NDVI cycle, and provide a valuable guide to assist in quantifying uncertainty and interpreting land cover and land use changes inferred from these Landsat data.     

A spatial processing algorithm to reduce the effects of mixed pixels and increase the separability between classes

Mixed pixels are an important problem in the identification and proportion estimation of agriculturally important crops in Landsat satellite scenes. The spectral response of mixed pixels is influenced by more than one ground cover type which decreases the separability of component crop classes and hence degrades the performance of classification procedures. An algorithm called CASCADE is described which is based on spatial information and consideration of a linear mixing model. The CASCADE procedure provides a means for allocating a pixel to one of the surrounding, more “homogeneous” regions which it most closely resembles. Processing all pixels in an image with CASCADE before classification significantly increases the separability between crop classes as well as the precision of the crop proportion estimates.     

Scene selection and the use of NASA's global orthorectified Landsat dataset for land cover and land use change monitoring

This study examines the utility of NASA's circa 1990 and circa 2000 global orthorectified Landsat dataset for land cover and land use change mapping and monitoring across Africa. This is achieved by comparing the temporal and spatial variation of NDVI, measured independently by the NOAA‐AVHRR at the time of Landsat scene acquisition, against the seasonal mean for each Landsat scene extent. Decadal sequences of drift‐corrected NOAA‐AVHRR imagery were used to calculate NDVI means and standard deviations for the periods covered by the scenes composing the c.1990 and c.2000 Landsat datasets. The specific NOAA‐AVHRR NDVI values at the acquisition date of each individual Landsat scene were also calculated and the differences, both from the mean and scaled by standard deviation, were mapped for the Landsat scene footprints in the c.1990 and c.2000 datasets. The resulting maps show the temporal position of each Landsat scene within the seasonal NDVI cycle, and provide a valuable guide to assist in quantifying uncertainty and interpreting land cover and land use changes inferred from these Landsat data.     

Discrimination of irrigated and rainfed rice in a tropical agricultural system using SPOT VEGETATION NDVI and rainfall data

The classification of irrigated crops by remote sensing requires the use of time series data, since the timing, cropping intensity and duration of cropping is quite variable over the course of a year. Rice is the dominant irrigated crop in tropical and sub‐tropical Asia, where rainfall is high, but is seasonal and often uni‐modal. Existing crop classification methods for rice are not able to distinguish between rainfed and irrigated crops, leading to errors in classification and estimated irrigated area. This paper describes a technique, a ‘peak detector algorithm’, to successfully discriminate between rainfed and irrigated rice crops in Suphanburi province, Thailand. The methodology uses a three‐year time series of Satellite pour l'Observation de la Terre (SPOT) VEGETATION S10 Normalized Difference Vegetation Index (NDVI) data (10 day composites) to identify cropping intensity (number, timing and peak values). Peak NDVI is then lag‐correlated with long term average rainfall data. There is a high correlation at a 40–50 day lag, between a peak rainfall and a ‘single’ peak NDVI of rainfed rice. In irrigated areas, there are multiple peaks, and multiple correlations with low values for at least 90 days after peak rainfall. The methodology currently uses a mask to remove un‐cropped and non‐rice areas, which is derived from existing Geographical Information Systems (GIS). The method achieves a classification accuracy of 89% or better against independent groundtruth data. The procedure is designed as a second level of analysis to refine classifications using other techniques of mapping irrigated area at global and regional scales.     

Spectral analysis and classification accuracy of coffee crops using Landsat and a topographic‐environmental model

Coffee is an extremely important cash crop, yet previous work indicates that satellite mapping of coffee has produced low classification accuracy. This research examines spectral band combinations and ancillary data for evaluating the classification accuracy and the nature of spectral confusion between coffee and other cover types in a Costa Rican study area. Supervised classification using Landsat Enhanced Thematic Mapper (ETM+) with only red, near‐infrared, and mid‐infrared bands had significantly lower classification accuracy compared to datasets that included more spectral bands and ancillary data. The highest overall accuracy achieved was 65%, including a coffee environmental stratification model (CESM). Producer's and user's accuracy was highest for shade coffee plantations (91.8 and 61.1%) and sun coffee (86.2 and 68.4%) with band combination ETM+ 34567, NDVI, cos (i), and including the use of the CESM. Post‐classification stratification of the optimal coffee growing zone based on elevation and precipitation data did not show significant improvement in land cover classification accuracy when band combinations included both the thermal band and NDVI. A forward stepwise discriminant analysis indicated that ETM+ 5 (mid‐infrared band) had the highest discriminatory power. The best discriminatory subset for all woody cover types including coffee excluded ETM+ 3 and 7; however, the land cover accuracy assessment indicated that overall accuracy, as well as producer's and user's accuracy of shade and sun coffee, were slightly improved with the inclusion of these bands. Although spectral separation between coffee crops and woodland areas was only moderately successful in the Costa Rica study, the overall accuracy, as well as the sun and shade coffee producer's and user's accuracy, were higher than reported in previous research.     

Mapping rice field anopheline breeding habitats in Mali,West Africa,using Landsat ETM+ sensor data

The aim of this study was to determine whether remotely sensed data could be used to identify rice-related malaria vector breeding habitats in an irrigated rice growing area near Niono, Mali. Early stages of rice growth show peak larval production, but Landsat sensor data are often obstructed by clouds during the early part of the cropping cycle (rainy season). In this study, we examined whether a classification based on two Landsat Enhanced Thematic Mapper (ETM)+ scenes acquired in the middle of the season and at harvesting times could be used to map different land uses and rice planted at different times (cohorts), and to infer which rice growth stages were present earlier in the season. We performed a maximum likelihood supervised classification and evaluated the robustness of the classifications with the transformed divergence separability index, the kappa coefficient and confusion matrices. Rice was distinguished from other land uses with 98% accuracy and rice cohorts were discriminated with 84% accuracy (three classes) or 94% (two classes). Our study showed that optical remote sensing can reliably identify potential malaria mosquito breeding habitats from space. In the future, these ‘crop landscape maps' could be used to investigate the relationship between cultivation practices and malaria transmission.     

Hierarchical Land Cover Classification based on MODIS NDVI Time-series

In this paper,we mainly used MODIS NDVI time-series dataset at 16-days temporal resolution and 250-meters spatial resolution to analyze land cover mapping of northeastern China.We used two different filter methods to fit NDVI time-series dataset,and compared their average classes’ separability based on Jeffries-Matusita distance index.In addition,we made use of hierarchical classification method to complete classification,combined with short-wave infrared spectral reflectance data and DEM.We conformed to the principle that separate area hierarchically into several parts first and then classify each part further,and use a single characteristic band first and then multiple feature bands.In the process of classification,we adopted threshold value method,support vector machine,artificial net neural and C5.0 decision tree classification to distinguish each land-cover type hierarchically.Finally,we evaluated the accuracy of the final classification of study area using known land-cover classification data and high-resolution remote sensing imagery,overall accuracy is 84.61%,Kappa coefficient is 0.8262.     

Boosted decision tree classifications of land cover over Turkey integrating MODIS,climate and topographic data

This study investigates the impact of using different combinations of Moderate Resolution Imaging Spectroradiometer (MODIS) and ancillary datasets on overall and per-class classification accuracies for nine land cover types modified from the classification system of the International Geosphere Biosphere Programme (IGBP). Twelve land cover maps were generated for Turkey using boosted decision trees (BDTs) based on the stepwise addition of 14 explanatory variables derived from a time series of 16-day MODIS composites between 2000 and 2006 (Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI) and four spectral bands) and ancillary climate and topographic data (minimum and maximum air temperature, precipitation, potential evapotranspiration, aspect, elevation, distance to sea and slope) at 500-m resolution. Evaluation of the 12 BDTs indicated that the BDT built as a function of all the MODIS and climate variables, aspect and elevation produced the highest degree of overall classification accuracy (79.8%) and kappa statistic (0.76) followed by the BDTs that additionally included distance to sea (DtS), and both DtS and slope. Based on an independent validation dataset derived from a pre-existing national forest map and Landsat images of Turkey, the highest overall accuracy (64.7%) and kappa coefficient (0.58) among the 12 land cover maps was achieved by using MODIS-derived NDVI time series only, followed by NDVI and EVI time series combined; NDVI, EVI and four MODIS spectral bands; and the combination of all MODIS and climate data, aspect, elevation and distance to sea, respectively. The largest improvements in producer's accuracies were observed for grasslands (+50%), barrenlands (+46%) and mixed forests (+39%) and in user's accuracies for grasslands (+53%), shrublands (+30%) and mixed forests (+28%), in relation to the lowest producer's accuracy. The results of this study indicate that BDTs can increase the accuracy of land cover classifications at the national scale.     

Evaluating different NDVI composite techniques using NOAA-14 AVHRR data

Normalized Difference Vegetation Index (NDVI) data derived from Advanced Very High Resolution Radiometer (AVHRR) data are influenced by cloud contamination, which is common in individual AVHRR scenes. Maximum value compositing (MVC) of NDVI data has been employed to minimize cloud contamination. Two types of weekly NDVI composites were built for crop seasons in summer: one from all available AVHRR data (named the traditional NDVI composite) and the other from solely cloud-free AVHRR data (named the conditional NDVI composite). The MVC method was applied to both composites. The main objective of this study was to compare the two types of NDVI composites using Texas data. The NDVI seasonal profiles produced from the conditional NDVI composites agreed with the field measured leaf area index (LAI) data, reaching maximum values at similar times. However, the traditional NDVI composites showed irregular patterns, primarily due to cloud contamination. These study results suggest that cloud detection for individual AVHRR scenes should be strongly recommended before producing weekly NDVI composites. Appropriate AVHRR data pre-processing is important for composite products to be used for short-term vegetation condition and biomass studies, where the traditional NDVI composite data do not eliminate cloud-contaminated pixels. In addition, this study showed that atmosphere composition affected near-infrared reflectance more than visible reflectance. The near-infrared reflectance was increasingly adjusted through atmospheric correction.