Estimating crop area using seasonal time series of enhanced vegetation index from MODIS satellite imagery

Cereal grain is one of the main export commodities of Australian agriculture. Over the past decade, crop yield forecasts for wheat and sorghum have shown appreciable utility for industry planning at shire, state, and national scales. There is now an increasing drive from industry for more accurate and cost-effective crop production forecasts. In order to generate production estimates, accurate crop area estimates are needed by the end of the cropping season. Multivariate methods for analysing remotely sensed Enhanced Vegetation Index (EVI) from 16-day Moderate Resolution Imaging Spectroradiometer (MODIS) satellite imagery within the cropping period (i.e. April-November) were investigated to estimate crop area for wheat, barley, chickpea, and total winter cropped area for a case study region in NE Australia. Each pixel classification method was trained on ground truth data collected from the study region. Three approaches to pixel classification were examined: (i) cluster analysis of trajectories of EVI values from consecutive multi-date imagery during the crop growth period; (ii) harmonic analysis of the time series (HANTS) of the EVI values; and (iii) principal component analysis (PCA) of the time series of EVI values. Images classified using these three approaches were compared with each other, and with a classification based on the single MODIS image taken at peak EVI. Imagery for the 2003 and 2004 seasons was used to assess the ability of the methods to determine wheat, barley, chickpea, and total cropped area estimates. The accuracy at pixel scale was determined by the percent correct classification metric by contrasting all pixel scale samples with independent pixel observations. At a shire level, aggregated total crop area estimates were compared with surveyed estimates. All multi-temporal methods showed significant overall capability to estimate total winter crop area. There was high accuracy at pixel scale (>98% correct classification) for identifying overall winter cropping. However, discrimination among crops was less accurate. Although the use of single-date EVI data produced high accuracy for estimates of wheat area at shire scale, the result contradicted the poor pixel-scale accuracy associated with this approach, due to fortuitous compensating errors. Further studies are needed to extrapolate the multi-temporal approaches to other geographical areas and to improve the lead time for deriving cropped-area estimates before harvest.

Remote detection and distinction of ants using nest-site specific LISS-derived Normalised Difference Vegetation Index

This study in Western Ghats, India, investigates the relation between nesting sites of ants and a single remotely sensed variable: the Normalised Difference Vegetation Index (NDVI). We carried out sampling in 60 plots each measuring 30 x 30 m and recorded nest sites of 13 ant species. We found that NDVI values at the nesting sites varied considerably between individual species and also between the six functional groups the ants belong to. The functional groups Cryptic Species, Tropical Climate Specialists and Specialist Predators were present in regions with high NDVI whereas Hot Climate Specialists and Opportunists were found in sites with low NDVI. As expected we found that low NDVI values were associated with scrub jungles and high NDVI values with evergreen forests. Interestingly, we found that Pachycondyla rufipes, an ant species found only in deciduous and evergreen forests, established nests only in sites with low NDVI (range = 0.015 - 0.1779). Our results show that these low NDVI values in deciduous and evergreen forests correspond to canopy gaps in otherwise closed deciduous and evergreen forests. Subsequent fieldwork confirmed the observed high prevalence of P. rufipes in these NDVI-constrained areas. We discuss the value of using NDVI for the remote detection and distinction of ant nest sites.

Real-time climate monitoring using an AVHRR-based vegetation index

A normalized difference vegetation index (NDVI) has been produced and archived on a 1° latitude by 1° longitude grid between 55°S and 75°N. The many sources of data errors in the NDVI include cloud contamination, scan angle biases, changes in solar zenith angle, and sensor degradation. Week-to-week variability, primarily caused by cloud contamination and scan angle biases, can be minimized by temporally filtering the data. Orbital drift and sensor degradation introduces interannual variability into the dataset. These trends make the usefulness of a long-term climatology uncertain and limit the usefulness of the NDVI. Elimination of these problems should produce an index that can be used for climate monitoring.

Monitoring the vegetation start of season (SOS) across the island of Ireland using the MERIS global vegetation index

The aim of this study was to develop a methodology, based on satellite remote sensing, to estimate the vegetation Start of Season (SOS) across the whole island of Ireland on an annual basis. This growing body of research is known as Land Surface Phenology (LSP) monitoring. The SOS was estimated for each year from a 7-year time series of 10-day composited, 1.2 km reduced resolution MERIS Global Vegetation Index (MGVI) data from 2003 to 2009, using the time series analysis software, TIMESAT. The selection of a 10-day composite period was guided by in-situ observations of leaf unfolding and cloud cover at representative point locations on the island. The MGVI time series was smoothed and the SOS metric extracted at a point corresponding to 20% of the seasonal MGVI amplitude. The SOS metric was extracted on a per pixel basis and gridded for national scale coverage. There were consistent spatial patterns in the SOS grids which were replicated on an annual basis and were qualitatively linked to variation in landcover. Analysis revealed that three statistically separable groups of CORINE Land Cover (CLC) classes could be derived from differences in the SOS, namely agricultural and forest land cover types, peat bogs, and natural and semi-natural vegetation types. These groups demonstrated that managed vegetation, e.g. pastures has a significantly earlier SOS than in unmanaged vegetation e.g. natural grasslands. There was also interannual spatio-temporal variability in the SOS. Such variability was highlighted in a series of anomaly grids showing variation from the 7-year mean SOS. An initial climate analysis indicated that an anomalously cold winter and spring in 2005/2006, linked to a negative North Atlantic Oscillation index value, delayed the 2006 SOS countrywide, while in other years the SOS anomalies showed more complex variation. A correlation study using air temperature as a climate variable revealed the spatial complexity of the air temperature-SOS relationship across the Republic of Ireland as the timing of maximum correlation varied from November to April depending on location. The SOS was found to occur earlier due to warmer winters in the Southeast while it was later with warmer winters in the Northwest. The inverse pattern emerged in the spatial patterns of the spring correlates. This contrasting pattern would appear to be linked to vegetation management as arable cropping is typically practiced in the southeast while there is mixed agriculture and mostly pastures to the west. Therefore, land use as well as air temperature appears to be an important determinant of national scale patterns in the SOS. The TIMESAT tool formed a crucial component of the estimation of SOS across the country in all seven years as it minimised the negative impact of noise and data dropouts in the MGVI time series by applying a smoothing algorithm. The extracted SOS metric was sensitive to temporal and spatial variation in land surface vegetation seasonality while the spatial patterns in the gridded SOS estimates aligned with those in landcover type. The methodology can be extended for a longer time series of FAPAR as MERIS will be replaced by the ESA Sentinel mission in 2013, while the availability of full resolution (300m) MERIS FAPAR and equivalent sensor products holds the possibility of monitoring finer scale seasonality variation. This study has shown the utility of the SOS metric as an indicator of spatiotemporal variability in vegetation phenology, as well as a correlate of other environmental variables such as air temperature. However, the satellite-based method is not seen as a replacement of ground-based observations, but rather as a complementary approach to studying vegetation phenology at the national scale. In future, the method can be extended to extract other metrics of the seasonal cycle in order to gain a more comprehensive view of seasonal vegetation development.

MODIS-based vegetation index has sufficient sensitivity to indicate stand-level intra-seasonal climatic stress in oak and beech forests

Context: Variation in photosynthetic activity of trees induced by climatic stress can be effectively evaluated using remote sensing data. Although adverse effects of climate on temperate forests have been subjected to increased scrutiny, the suitability of remote sensing imagery for identification of drought stress in such forests has not been explored fully. Aim: To evaluate the sensitivity of MODIS-based vegetation index to heat and drought stress in temperate forests, and explore the differences in stress response of oaks and beech. Methods: We identified 8 oak and 13 beech pure and mature stands, each covering between 4 and 13 MODIS pixels. For each pixel, we extracted a time series of MODIS NDVI from 2000 to 2010. We identified all sequences of continuous unseasonal NDVI decline to be used as the response variable indicative of environmental stress. Neural Networks-based regression modelling was then applied to identify the climatic variables that best explain observed NDVI declines. Results: Tested variables explained 84–97% of the variation in NDVI, whilst air temperature-related climate extremes were found to be the most influential. Beech showed a linear response to the most influential climatic predictors, while oak responded in a unimodal pattern suggesting a better coping mechanism. Conclusions: MODIS NDVI has proved sufficiently sensitive as a stand-level indicator of climatic stress acting upon temperate broadleaf forests, leading to its potential use in predicting drought stress from meteorological observations and improving parameterisation of forest stress indices.

Calibration of a Species-Specific Spectral Vegetation Index for Leaf Area Index (LAI) Monitoring: Example with MODIS Reflectance Time-Series on Eucalyptus Plantations

The leaf area index (LAI) is a key characteristic of forest ecosystems. Estimations of LAI from satellite images generally rely on spectral vegetation indices (SVIs) or radiative transfer model (RTM) inversions. We have developed a new and precise method suitable for practical application, consisting of building a species-specific SVI that is best-suited to both sensor and vegetation characteristics. Such an SVI requires calibration on a large number of representative vegetation conditions. We developed a two-step approach: (1) estimation of LAI on a subset of satellite data through RTM inversion; and (2) the calibration of a vegetation index on these estimated LAI. We applied this methodology to Eucalyptus plantations which have highly variable LAI in time and space. Previous results showed that an RTM inversion of Moderate Resolution Imaging Spectroradiometer (MODIS) near-infrared and red reflectance allowed good retrieval performance (R-2 = 0.80, RMSE = 0.41), but was computationally difficult. Here, the RTM results were used to calibrate a dedicated vegetation index (called "EucVI") which gave similar LAI retrieval results but in a simpler way. The R-2 of the regression between measured and EucVI-simulated LAI values on a validation dataset was 0.68, and the RMSE was 0.49. The additional use of stand age and day of year in the SVI equation slightly increased the performance of the index (R-2 = 0.77 and RMSE = 0.41). This simple index opens the way to an easily applicable retrieval of Eucalyptus LAI from MODIS data, which could be used in an operational way.

Spatial and temporal patterns of greenness on the Yamal Peninsula, Russia: interactions of ecological and social factors affecting the Arctic normalized difference vegetation index

The causes of a greening trend detected in the Arctic using the normalized difference vegetation index (NDVI) are still poorly understood. Changes in NDVI are a result of multiple ecological and social factors that affect tundra net primary productivity. Here we use a 25 year time series of AVHRR-derived NDVI data (AVHRR: advanced very high resolution radiometer), climate analysis, a global geographic information database and ground-based studies to examine the spatial and temporal patterns of vegetation greenness on the Yamal Peninsula, Russia. We assess the effects of climate change, gas-field development, reindeer grazing and permafrost degradation. In contrast to the case for Arctic North America, there has not been a significant trend in summer temperature or NDVI, and much of the pattern of NDVI in this region is due to disturbances. There has been a 37% change in early-summer coastal sea-ice concentration, a 4% increase in summer land temperatures and a 7% change in the average time-integrated NDVI over the length of the satellite observations. Gas-field infrastructure is not currently extensive enough to affect regional NDVI patterns. The effect of reindeer is difficult to quantitatively assess because of the lack of control areas where reindeer are excluded. Many of the greenest landscapes on the Yamal are associated with landslides and drainage networks that have resulted from ongoing rapid permafrost degradation. A warming climate and enhanced winter snow are likely to exacerbate positive feedbacks between climate and permafrost thawing. We present a diagram that summarizes the social and ecological factors that influence Arctic NDVI. The NDVI should be viewed as a powerful monitoring tool that integrates the cumulative effect of a multitude of factors affecting Arctic land-cover change.

Assessing landscape changes in Laos using MODIS Enhanced Vegetation Index

The ongoing rapid and vast land cover and land use transformations in Laos are only documented by punctual local case studies; information on national level is barely available. We explore ways to address this by using MODIS vegetation index times series data to detect medium to large scale transformation on the national level.

(Table 1) Normalized difference vegetation index in control and warming plots, Abisco Research Station

Extreme winter warming events in the sub-Arctic have caused considerable vegetation damage due to rapid changes in temperature and loss of snow cover. The frequency of extreme weather is expected to increase due to climate change thereby increasing the potential for recurring vegetation damage in Arctic regions. Here we present data on vegetation recovery from one such natural event and multiple experimental simulations in the sub-Arctic using remote sensing, handheld passive proximal sensors and ground surveys. Normalized difference vegetation index (NDVI) recovered fast (2 years), from the 26% decline following one natural extreme winter warming event. Recovery was associated with declines in dead Empetrum nigrum (dominant dwarf shrub) from ground surveys. However, E. nigrum healthy leaf NDVI was also reduced (16%) following this winter warming event in experimental plots (both control and treatments), suggesting that non-obvious plant damage (i.e., physiological stress) had occurred in addition to the dead E. nigrum shoots that was considered responsible for the regional 26% NDVI decline. Plot and leaf level NDVI provided useful additional information that could not be obtained from vegetation surveys and regional remote sensing (MODIS) alone. The major damage of an extreme winter warming event appears to be relatively transitory. However, potential knock-on effects on higher trophic levels (e.g., rodents, reindeer, and bear) could be unpredictable and large. Repeated warming events year after year, which can be expected under winter climate warming, could result in damage that may take much longer to recover.

Temperature, precipitation and Normalized Difference Vegetation Index (NDVI) from 1982 to 2009 in the Koshi River Basin of the Middle Himalayas

The aim of this paper is to find out if there is a significant difference in using NDVI dataset processed by harmonic analysis method to evaluate its dynamic and response to climate change, compared with the original data.

Normalised Difference Vegetation Index

Resources created at the University of Southampton for the module Remote Sensing for Earth Observation

Enhanced Vegetation Index

Resources created at the University of Southampton for the module Remote Sensing for Earth Observation

Sugarcane leaf area estimate obtained from the corrected normalized difference vegetation index (NDVI)

v. 46, n. 2, p. 140-148, apr./jun. 2016.

Evaluation of spectral and texture features for object-based vegetation species classification using Support Vector Machines

The use of appropriate features to characterize an output class or object is critical for all classification problems. This paper evaluates the capability of several spectral and texture features for object-based vegetation classification at the species level using airborne high resolution multispectral imagery. Image-objects as the basic classification unit were generated through image segmentation. Statistical moments extracted from original spectral bands and vegetation index image are used as feature descriptors for image objects (i.e. tree crowns). Several state-of-art texture descriptors such as Gray-Level Co-Occurrence Matrix (GLCM), Local Binary Patterns (LBP) and its extensions are also extracted for comparison purpose. Support Vector Machine (SVM) is employed for classification in the object-feature space. The experimental results showed that incorporating spectral vegetation indices can improve the classification accuracy and obtained better results than in original spectral bands, and using moments of Ratio Vegetation Index obtained the highest average classification accuracy in our experiment. The experiments also indicate that the spectral moment features also outperform or can at least compare with the state-of-art texture descriptors in terms of classification accuracy.