Stephen Boles

Mapping H5N1 highly pathogenic avian influenza risk in Southeast Asia

The highly pathogenic avian influenza (HPAI) H5N1 virus that emerged in southern China in the mid-1990s has in recent years evolved into the first HPAI panzootic. In many countries where the virus was detected, the virus was successfully controlled, whereas other countries face periodic reoccurrence despite significant control efforts. A central question is to understand the factors favoring the continuing reoccurrence of the virus. The abundance of domestic ducks, in particular free-grazing ducks feeding in intensive rice cropping areas, has been identified as one such risk factor based on separate studies carried out in Thailand and Vietnam. In addition, recent extensive progress was made in the spatial prediction of rice cropping intensity obtained through satellite imagery processing. This article analyses the statistical association between the recorded HPAI H5N1 virus presence and a set of five key environmental variables comprising elevation, human population, chicken numbers, duck numbers, and rice cropping intensity for three synchronous epidemic waves in Thailand and Vietnam. A consistent pattern emerges suggesting risk to be associated with duck abundance, human population, and rice cropping intensity in contrast to a relatively low association with chicken numbers. A statistical risk model based on the second epidemic wave data in Thailand is found to maintain its predictive power when extrapolated to Vietnam, which supports its application to other countries with similar agro-ecological conditions such as Laos or Cambodia. The models potential application to mapping HPAI H5N1 disease risk in Indonesia is discussed.

Combining remote sensing and ground census data to develop new maps of the distribution of rice agriculture in China

[1] Large-scale assessments of the potential for food production and its impact on biogeochemical cycling require the best possible information on the distribution of cropland. This information can come from ground-based agricultural census data sets and/ or spaceborne remote sensing products, both with strengths and weaknesses. Official cropland statistics for China contain much information on the distribution of crop types, but are known to significantly underestimate total cropland areas and are generally at coarse spatial resolution. Remote sensing products can provide moderate to fine spatial resolution estimates of cropland location and extent, but supply little information on crop type or management. We combined county-scale agricultural census statistics on total cropland area and sown area of 17 major crops in 1990 with a fine-resolution land-cover map derived from 1995–1996 optical remote sensing (Landsat) data to generate 0.5� resolution maps of the distribution of rice agriculture in mainland China. Agricultural census data were used to determine the fraction of crop area in each 0.5� grid cell that was in single rice and each of 10 different multicrop paddy rice rotations (e.g., winter wheat/ rice), while the remote sensing land-cover product was used to determine the spatial distribution and extent of total cropland in China. We estimate that there were 0.30 million km 2 of paddy rice cropland; 75% of this paddy land was multicropped, and 56% had two rice plantings per year. Total sown area for paddy rice was 0.47 million km 2 . Paddy rice agriculture occurred on 23% of all cultivated land in China. INDEX TERMS: 0315 Atmospheric Composition and Structure: Biosphere/atmosphere interactions; 1615 Global Change: Biogeochemical processes (4805); KEYWORDS: paddy rice, maps, China, multicropping rotation, Landsat

Characterization of forest types in Northeastern China, using multi-temporal SPOT-4 VEGETATION sensor data

Abstract In this study, we explored the potential of multi-temporal SPOT-4 VEGETATION (VGT) sensor data for characterization of temperate and boreal forests in Northeastern China. As the VGT sensor has a short-wave infrared (SWIR) band that is sensitive to vegetation, soil moisture and leaf water content, the Normalized Difference Water Index (NDWI) was calculated in addition to the Normalized Difference Vegetation Index (NDVI). A forest map of Northeast China was generated from an unsupervised classification of 25 10-day VGT composite data (NDVI and NDWI) over the period of March 11–20, 1999 to November 11–20, 1999. Seven different forest categories were distinguished from the 1-km spatial resolution VGT data. The VGT forest map was compared to estimates of forest area derived from Landsat 7 Enhanced Thematic Mapper (ETM+) images. There was a good agreement on spatial distribution and area of forest between the VGT product and the TM product, however, the VGT product provided additional information on forest type. Analysis of NDVI and NDWI over the plant growing season allows for the identification of distinct growth patterns between the different forest types. It is evident that VGT data can be used to provide timely and detailed forest maps with limited ancillary data needed. The VGT-derived forest maps could be very useful as input to biogeochemical models (particularly carbon cycle models) that require timely estimates of forest area and type.

Sensitivity of vegetation indices to atmospheric aerosols: continental-scale observations in Northern Asia

Abstract Satellite observations play an important role in characterization of the interannual variation of vegetation. Here, we report anomalies of two vegetation indices for Northern Asia (40°N–75°N, and 45°E–179°E), using images from the SPOT-4 VEGETATION (VGT) sensor over the period of April 1, 1998 to November 20, 2001. The Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI), which are correlated to a number of vegetation properties (e.g., net primary production, leaf area index), were compared. The results show that there is a large disagreement between NDVI and EVI anomalies in 1998 and 1999 for Northern Asia. The NDVI anomaly in 1998 was largely affected by atmospheric contamination, predominantly aerosols from extensive forest fires in that year. The EVI anomaly in 1998 was less sensitive to residual atmospheric contamination, as it is designed to be, and thus EVI is a useful alternative vegetation index for the large-scale study of vegetation. The EVI anomaly also suggests that potential vegetation productivity in Northern Asia was highest in 1998 but declined substantially in 2001, consistent with precipitation data from 1998–2001.

Bias in land cover change estimates due to misregistration

Land cover change may be overestimated due to positional error in multi-temporal images. To assess the potential magnitude of this bias, we introduced random positional error to identical classified images and then subtracted them. False land cover change ranged from less than 5% for a 5-class AVHRR classification, to more than 33% for a 20-class Landsat TM classification. The potential for false change was higher with more classes. However, false change could not be reliably estimated simply by number of classes, since false change varied significantly by simulation trial when class size remained constant. Registration model root mean squared (rms) error may underestimate the actual image co-registration asccuracy. In simulations with 5 to 50 ground control locations, the mean model rms error was always less than the actual population rms error. The model rms error was especially unreliable when small sample sizes were used to develop second order rectification models. We introduce a bootstrap resampling method to estimate false land cover change due to positional error. Although the bootstrap estimates were unbiased, the precision of the estimates may be too low to be of practical value in some land cover change applications.

Landscape-scale characterization of cropland in China using Vegetation and Landsat TM images

In this landscape-scale study we explored the potential for multitemporal 10-day composite data from the Vegetation sensor to characterize land cover types, in combination with Landsat TM image and agricultural census data. The study area (175 km by 165 km) is located in eastern Jiangsu Province, China. The Normalized Difference Vegetation Index (NDVI ) and the Normalized Difference Water Index (NDWI ) were calculated for seven 10-day composite (VGT-S10) data from 11 March to 20 May 1999. Multi-temporal NDVI and NDWI were visually examined and used for unsupervised classification. The resultant VGT classification map at 1 km resolution was compared to the TM classification map derived from unsupervised classification of a Landsat 5 TM image acquired on 26 April 1996 at 30 m resolution to quantify percent fraction of cropland within a 1 km VGT pixel; resulting in a mean of 60% for pixels classified as cropland, and 47% for pixels classified as cropland/natural vegetation mosaic. The estimates of cropland area from VGT data and TM image were also aggregated to county-level, using an administrative county map, and then compared to the 1995 county-level agricultural census data. This landscape-scale analysis incorporated image classification (e.g. coarse-resolution VGT data, fineresolution TM data), statistical census data (e.g. county-level agricultural census data) and a geographical information system (e.g. an administrative county map), and demonstrated the potential of multi-temporal VGT data for mapping of croplands across various spatial scales from landscape to region. This analysis also illustrated some of the limitations of per-pixel classification at the 1 km resolution for a heterogeneous landscape.

Uncertainties in estimates of cropland area in China: a comparison between an AVHRR-derived dataset and a Landsat TM-derived dataset

The large uncertainties in estimates of cropland area in China may have significant implications for major cross-cutting themes of global environmental change-food production and trade, water resources, and the carbon and nitrogen cycles. Many earlier studies have indicated significant under-reporting of cropland area in China from official agricultural census statistics datasets. Space-borne remote sensing analyses provide an alternative and independent approach for estimating cropland area in China. In this study, we report estimates of cropland area from the National Land Cover Dataset (NLCD-96) at the 1:100,000 scale, which was generated by a multi-year National Land Cover Project in China through visual interpretation and digitization of Landsat TM images acquired mostly in 1995 and 1996. We compared the NLCD-96 dataset to another land cover dataset at I-km spatial resolution (the IGBP DIScover dataset version 2.0), which was generated from monthly Advanced Very High Resolution Radiometer (AVHRR)-derived Normalized Difference Vegetation Index (NDVI) from April, 1992 to March, 1993. The data comparison highlighted the limitation and uncertainty of cropland area estimates from the DIScover dataset

Quantitative relationships between field-measured leaf area index and vegetation index derived from VEGETATION images for paddy rice fields

In an effort to develop the quantitative relationships between fieldmeasured leaf area index (LAI ) and VEGETATION-derived vegetation indices for paddy rice fields, we have measured LAI of paddy rice fields at 10-day intervals at five sampling sites in Jiangning County, Jiangsu Province of China during the rice growing season (July to October) of 1999, using a LI-COR LAI-2000 plant canopy analyser. Twenty-seven 10-day VEGETATION (VGT) synthesis products (VGT-S10) from 1-10 March to 21-30 November 1999 were acquired. Normalized difference vegetation index (NDVI) values were calculated for the VGT-S10 products, using ground surface reflectance values of VGT spectral bands (B3-near-infrared; B2-red). After rice transplanting in mid to late June, LAI increased rapidly and reached its plateau by early to mid August. There were similar temporal dynamics of NDVI and LAI among the five sampling sites over the growing season of paddy rice in 1999. Simple linear regression analyses indicate that there are statistically significant linear relationships between NDVI and LAI data over the growing season of paddy rice in 1999.

Mapping Single‐, Double‐, and Triple‐crop Agriculture in China at 0.5° × 0.5° by Combining County‐scale Census Data with a Remote Sensing‐derived Land Cover Map

Abstract Assessments of agriculture at the national scale require the best possible information on the distribution of cropland and the management of that cropland. Official cropland statistics for China contain much information on the distribution of crop types, but are known to underestimate total cropland area and are generally at coarse spatial resolution. Remote sensing products can provide moderate to fine spatial resolution estimates of cropland location and extent, but usually supply little information on crop type or management. We combined 1990 county‐scale agricultural census statistics on total cropland area and sown area of 17 major crops with a fine‐resolution land‐cover map derived from 1995‐96 optical remote sensing (Landsat Thematic Mapper) data to generate 0.5° resolution maps of the distribution of 47 different single‐ and multi‐crop rotations in mainland China. We estimated that, of 1.3‐million km2 of cropland, approximately 60% was single‐cropped, 30% was double‐cropped, and 10% was triple‐cropped. Total sown or planted area was 2.0‐million km2.

Large-scale observations of alpine snow and ice cover in Asia: Using multi-temporal VEGETATION sensor data

Abstract In this study we used twenty-five 10-day synthetic products (composites) of the VEGETATION (VGT) sensor on SPOT 4 satellite from 11-20 March 1999 to 11-20 November 1999 to map snow and ice cover at 1-km spatial resolution in an alpine environment of Asia. The study area is within 73°-103° E and 25°-40° N (a total land area of 4.7 million km 2 ), and includes the highest alpine region in the world (e.g. Himalaya Mountains, Qinghai-Tibet Plateau, Pamirs Plateau, Karakoram Range). In the VGT-based mapping approach, the Normalized Difference Snow/Ice Index (NDSII) is calculated as NDSII=(B2-MIR)/(B2+MIR), where the ground reflectance values of the red band (VGT band B2) and the mid-infrared band (VGT band MIR) are used. When NDSII and the reflectance value of the near-infrared band (VGT band B3) in a pixel (1 km 2 1 km) meet the thresholds of NDSII S 0.40 and band B3>0.11, the pixel is assigned to be snow/ice cover. Total areas of snow/ice cover in 2 the alpine region had large seasonal variations, ranging from 0.46 million km in 11-20 March 1999 to 0.05 million km 2 in 1-10 August 1999 to 0.60 million km 2 in 11-20 November 1999. Spatial distributions of snow/ice cover also varied substantially across the alpine region. There were large areas of snow/ice cover in the north-western part of the alpine region in March and April, but large areas of snow/ice cover in the eastern part of the alpine region in October and November. Qualitatively, spatial patterns and temporal dynamics of snow/ice cover in the alpine region are closely correlated to the plateau monsoon climate and its precipitation patterns.