Christos Domenikiotis

Early cotton yield assessment by the use of the NOAA/AVHRR derived Vegetation Condition Index (VCI) in Greece

Satellite data can significantly contribute to agricultural monitoring. The reflected radiation, as recorded by satellite sensors, provides an indication of the type, density and condition of canopy. A widely used index for vegetation monitoring is the Normalized Difference Vegetation Index (NDVI) derived from the National Oceanic and Atmospheric Administration/Advanced Very High Resolution Radiometer (NOAA/AVHRR) data provided in high temporal resolution. An extension of the NDVI is the Vegetation Condition Index (VCI). VCI is a tool for monitoring agrometeorological conditions, providing a quantitative estimation of weather impact to vegetation. The primary objective of this paper is the quantitative assessment of the cotton yield before the end of the growing season by examining the weather effects as they are depicted by the VCI. The study area comprises several cotton producing areas in Greece. Ten-day NDVI maximum value composites (MVC) are initially utilized for the period 1982–1999. The correlation between VCI images as extracted from NDVI and the 10-day intervals during the growing season is examined to identify the critical periods associated mostly with the yield. Empirical relationships between VCI and yield are developed. The models are tested on an independent dataset. The results show that an early estimation of the cotton yield trend is feasible by the use of the VCI.

Agreement assessment of NOAA/AVHRR NDVI with Landsat TM NDVI for mapping burned forested areas

Meteorological satellites are appropriate for operational applications related to early warning, monitoring and damage assessment of forest fires. Environmental or resources satellites, with better spatial resolution than meteorological satellites, enable the delineation of the affected areas with a higher degree of accuracy. In this study, the agreement of two datasets, coming from National Oceanic and Atmospheric Administration/Advanced Very High Resolution Radiometer (NOAA/AVHRR) and Landsat TM, for the assessment of the burned area, was investigated. The study area comprises a forested area, burned during the forest fire of 21-24 July 1995 in Penteli, Attiki, Greece. Based on a colour composite image of Landsat TM a reference map of the burned area was produced. The scatterplot of the multitemporal Normalized Difference Vegetation Index (NDVI) images, from both Landsat TM and NOAA/AVHRR sensors, was used to detect the spectral changes due to the removal of vegetation. The extracted burned area was compared to the digitized reference map. The synthesis of the maps was carried out using overlay techniques in a Geographic Information System (GIS). It is illustrated that the NOAA/AVHRR NDVI accuracy is comparable to that from Landsat TM data. As a result NOAA/AVHRR data can, operationally, be used for mapping the extent of the burned areas.

Cotton yield estimation based on NOAA/AVHRR produced NDVI

Abstract The high temporal resolution of the NOAA/AVHRR satellite data can provide an efficient and consistent way for biomass and yield monitoring and assessment. The reflected radiation provides an indication of the type and density of canopy. A combination of spectral bands results in the vegetation indices, which are based on the spectral signatures of vegetation and account for the difference in response of the vegetation to the visible and near infrared channels. A widely used index for vegetation monitoring is the Normalized Difference Vegetation Index (NDVI), which is the normalized difference of visible (red) and near infrared channels. The condition, distribution, structure and the development of the vegetation through the phenological stages can affect the relation between yield and NDVI. The primary objective of this study consists of examining the feasibility of applying a regression model for cotton yield monitoring by using the NDVI. NOAA/AVHRR images of six areas covered with cotton have been examined for a period of three years. The results show that the temporal variability of cotton yield from year to year can be monitored by the use of NDVI.

Rainfall-frequency mapping for Greece

Abstract For the design of hydrotechnical projects in ungauged watersheds, the flood flow is estimated by various methods, which demand the estimation of rainfall of particular critical duration and return period. For medium-sized and large basins, the storms causing flood flows have usually duration larger than 24 hours. In this study daily rainfall data from 24 meteorological stations for the period 1950 to 1981 were used. These stations are evenly distributed over Greece. From these data, the rainfall depths for various durations were computed, i.e. 1 to 7 days. The Extreme Value I (Gumbel) theoretical distribution had the best fit to the data from other theoretical distributions and it was fitted to the maximum annual rainfall depths for various durations. As a result, the Depth-Duration_Frequency relationships for each station were estimated and mapped for Greece. Moreover, the analysis showed that the rainfall of various durations and return periods represents a certain percentage of the mean annual precipitation for hydrologicaly homogeneous areas of Greece. The homogeneous areas were identified through factor analysis of monthly precipitation data from 37 meteorological stations. The results of this study can be used for the estimation of rainfall at ungauged sites in medium and large watersheds but they should be applied with caution in mountainous areas. For the estimation of rainfall in these areas the climatic conditions of the region and the orographic enhancement of rainfall should be considered.

Early cotton production assessment in Greece based on a combination of the drought Vegetation Condition Index (VCI) and the Bhalme and Mooley Drought Index (BMDI)

A new methodological approach is presented for quantifying the meteorological effects on cotton production during the growing season in Greece. The proposed Bhalme and Mooley Vegetation Condition Index (BMVCI) is based on the Vegetation Condition Index (VCI) extracted by National Oceanic and Atmospheric Administration (NOAA) Advanced Very High Resolution Radiometer (AVHRR) data. In this approach the VCI data is processed with the Bhalme and Mooley methodology to assess the accumulated meteorological effects on cotton from April to August. The resulting index is at the same scale as the Z-Index, which is the classification of the Palmer Drought Severity Index (PDSI) extensively used for drought monitoring. For this study 16 years of data are examined to illustrate that the weather development as identified from satellite data with the use of BMVCI confirm unfavourable conditions for cotton production. For the validation of BMVCI an empirical relationship between the cotton production and the BMVCI values is derived. The resultant high correlation coefficient refers to very encouraging results and confirms the usefulness of the proposed integrated methodological approach as an effective tool for early assessment of the cotton production in Greece.

Remotely-Sensed innovative approach for the cumulative meteorological effects on cotton production

In this study an innovative approach for investigating the accumulated meteorological effects on cotton production during the growing season is presented. The quantification of the meteorological effects is based on the incorporation of the Bhalme and Mooley Drought Index (BMDI) methodology into the Vegetation Condition Index (VCI) extracted by NOAA/AVHRR data. The resulted Bhalme and Mooley Vegetation Condition Index (BMVCI) uses the same scale as the Z-Index of the Palmer Drought Severity Index (PDSI) for drought monitoring. The study area consists of the country of Greece. Eighteen years of NOAA/AVHRR data are examined and processed with the BMVCI to examine the unfavourable conditions for cotton production. For the validation of BMVCI an empirical relationship between the cotton production and the BMVCI values is derived. The method is developed based on the first sixteen years time series data and validated utilizing the following two years. The resultant high correlation coefficient and the approximation of the production for the validated years refer to very favourable results and confirms the usefulness of this integrated methodological approach as an effective tool to assess cotton production in Greece.

Water balance of forested mountainous watersheds using satellite-derived actual evapotranspiration

The use of actual evapotranspiration derived by satellite data at watershed scale in water balance modelling of forested mountainous watersheds is studied. Mean monthly maximum composites of the Normalized Difference Vegetation Index (NDVI), derived from the National Oceanic and Atmospheric Administration’s (NOAA) / Advanced Very High Resolution Radiometer (AVHRR) were correlated with monthly actual evapotranspiration rates estimated by a water balance model. The water balance model was applied to three mountainous and forested watersheds of Central Thessaly in Greece and the actual basin-wide evapotranspiration was estimated using two methods for the estimation of basin wide precipitation and two methods of potential evapotranspiration. The derived values of actual evapotranspiration were then correlated to NDVI data, and the developed equations were validated temporally and spatially. The actual evapotranspiration estimates, derived from NDVI and used in the water balance model, resulted in equally accurate simulations of monthly runoff when compared with the simulations acquired from the classical application of water balance model.