SADC THEMA Drought Service Drought Monitoring based on
SADC THEMA Drought Service Drought Monitoring based on Precipitation Indices M. Masocha (Ph. D), I. Gwitira, A. Murwira (Ph. D), K. S. Murwira (Ph. D), M. D. Shekede Department of Geography and Environmental science University of Zimbabwe
Drought Monitoring: Introduction • Monitoring drought is important for national agricultural and environmental planning • There are several indices that are used to monitor drought. • We differentiate two classes: – Traditional Precipitation based indices such as the Standardized precipitation Index (SPI) – Remote sensing based indices, in particular satellitebased indices, such as Vegetation condition index (VCI), among others.
SADC THEMA Drought Service Precipitation based Drought Monitoring Indices M. Masocha (Ph. D), I. Gwitira, A. Murwira (Ph. D), K. S. Murwira (Ph. D), M. D. Shekede Department of Geography and Environmental science University of Zimbabwe
Standardised Precipitation Index (SPI) • SPI is used to quantify the precipitation deficit, based on the probability of precipitation for multiple time scales (Mc. Kee et al. , 1995) • Calculation of the SPI requires a long-term monthly precipitation database with 30 year or more of data • This long-term record is fitted to a probability distribution, which is then transformed to into a normal distribution so that the mean SPI for the location and desired period is zero
Computing SPI • SPI is calculated by taking the difference of the precipitation from the mean for a particular time scale and then dividing it by the standard deviation i. e. , : where xi the precipitation of the selected period during the ith year; and are the mean and the standard deviation of the selected period, respectively
SPI Interpretation • A drought event occurs any time the SPI is continuously negative and reaches an intensity where the SPI is – 2. 0 or less; a drought event ends when the SPI becomes positive SPI Values Condition >= 2 extremely wet 1. 5 to 2 very wet 1 to 1. 5 moderately wet -1 to 1 near normal -1. 5 to -1 moderately dry -2 to -1. 5 severely dry <= -2 extremely dry
SPI: Its Merits and Pitfalls • SPI has many advantages over other indices in that it is simple and temporally flexible, thus allowing observation of water deficits at different time scales • However, SPI is calculated on site and based on precipitation only, hence it does not allow scientists to quantify the spatial extent of drought or vegetation response to water deficit • Consequently, SPI is best used in conjunction with other indices
SADC THEMA Drought Service Drought Monitoring Physical Basis of Remotely Sensed Drought Monitoring M. Masocha (Ph. D), A. Murwira (Ph. D), K. S. Murwira (Ph. D) I. Gwitira and M. D. Shekede Department of Geography and Environmental science University of Zimbabwe
Satellite-Based Remote Sensing Source: Wardlow, 2009
Monitoring Drought From Space: The Remote Sensing System radiation sensor image atmosphere target atmosphere reflectance transmission
Electromagnetic Spectrum
Optical Properties of Vegetation
Spectral Differentiation of Features on Earth
Basis of Vegetation monitoring NIR RED
Chivero Landsat TM NIR and R Scatter: Water NIR RED
Chivero Landsat TM, NIR and R Scatter: Water weed NIR RED
Chivero Landsat TM, NIR and R Scatter: Bare NIR RED
Drought Monitoring Indices: Normalized Difference Vegetation Index (NDVI) • Drought indices are important for detecting, monitoring and evaluating the magnitude of droughts • NDVI is a general measure of the state and health of vegetation (Tucker, 1979) and is one of the most widely used indices for monitoring droughts from landscape to global scales • Does NDVI increase or decrease during a drought event? Give reasons
NDVI-Rainfall Relationships: ETOSHA, Namibia (C. Sannier, Cranfield University, Silsoe, )
SADC THEMA: Drought Service Thank you A. Murwira (Ph. D), K. S. Murwira (Ph. D), M. Masocha (Ph. D), I. Gwitira and M. D. Shekede Department of Geography and Environmental science University of Zimbabwe and Scientific and Industrial Research and Development Centre, Geo-
SADC THEMA: Drought Service Remotely sensed Drought Indices DAY III
SADC THEMA Drought Service Drought Monitoring Remotely Sensed Drought Indices M. Masocha (Ph. D), I. Gwitira, A. Murwira (Ph. D), K. S. Murwira (Ph. D), M. D. Shekede Department of Geography and Environmental science University of Zimbabwe
Introduction • NDVI itself does not reflect drought or non-drought conditions but its deviation from the ‘normal’ is a useful tool for detecting and monitoring droughts • However, a number of indices based on the NDVI time series data for assessing drought have been proposed
Introduction • The severity of a drought may be defined as NDVI deviation from its long-term mean • A monthly NDVI time series for a drought year (1987) and a wet year (1993) compared to the NDVI long-term mean (averaged for the study area (pixel)
Vegetation Anomalies and Droughts Map source: http: //earthobservatory. nasa. gov/Natural. Hazards/view. php? id=18226 Jan-Mar
NDVI Difference • NDVIdiff =(NDVI – LT NDVI) • Where NDVIdiff is NDVI difference • LT NDVI is the long term average NDVI
Standardized difference vegetation index (SDVI) • SDVI =(NDVI – LT NDVI)/(STDEVNDVI for that period over time) • Where SDVI is standardized difference vegetation index • NDVI is the observed NDVI at in a particular dekad • STDEVNDVI is the Long term standard deviation of NDVI over the period
Vegetation Condition Index (VCI) • VCI shows how close the NDVI of the current time is to the minimum NDVI calculated from the long-term record for that given time (Kogan, 1995) and is calculated as: where, NDVImax and NDVImin are calculated from the long-term record (e. g. , 18 years) for that time (e. g. , week, month) and j is the index of the current time
VCI Interpretation • The condition of the vegetation presented by VCI is measured in percent • Different degrees of a drought severity are indicated by VCI values below 50% • A VCI threshold of 36% signifies an extreme drought condition (Liu and Kogan, 1996) but further research is needed to categorize the VCI by its severity in the range between 0% and 36%
VCI for Southern Africa: Frequency of Drought in January • We analysed VCI based extreme drought using a 25 year data series in Southern Africa (1982 -2006) and mapped areas extreme drought frequency
Temperature Condition Index (TCI) • TCI is based on brightness temperature and represents the deviation of the current month’s (week’s) temperature from the recorded maximum (Kogan 1995, 1997) • It is computed as: where, BT is the brightness temperature (e. g. , AVHRR band 4); BTmax and BTmin are the maximum and minimum values of respectively calculated from the long-term (e. g. , 18 years) record of remote-
Vegetation Health Index (VHI) • VHI integrates both Vegetation Condition Index (VCI) and thermal-based Temperature Condition Index (TCI) indicators derived from satellite data e. g. , Advanced Very High Resolution Radiometer (AVHRR) to depict drought stress as function of vegetation canopy greenness and temperature a and b are coefficients to quantify share of NDVI based VCI and TCI contribution in total vegetation health (Wardlow, 2009)
Percent of Average Seasonal Greenness (PASG) • PASG is a phenology metric which is based on time-series NDVI and is used for drought monitoring • where SGPn. Yn refers to the seasonal greenness (SG) for a ten-day period (Pn) of a specific year (Yn) and x. SGPn is the historical average for the same ten-day period
Relationships between VCI, PASG and SPI The trends of time-series SPOT VGTS NDVI for cropland correlation coefficients between VCI, PASG and multiscale SPIs in the Huang. Huai. Hai region, China (Zhou, L. et al. , 2010)
Discussion • As satellite data are now readily available, an operational drought-monitoring system based on remotely sensed drought indices has the potential to improve the efficiency of existing drought policies and management in Southern Africa • The satellite drought indices covered correlate well with key meteorological indicators of drought such as SPI
SADC THEMA: Drought Service Thank you A. Murwira (Ph. D), K. S. Murwira (Ph. D), M. Masocha (Ph. D), I. Gwitira and M. D. Shekede Department of Geography and Environmental science University of Zimbabwe and Scientific and Industrial Research and Development Centre, Geo-
SADC THEMA: Drought Service Relating Remotely Sensed Drought Indices to in-situ Crop and Rangeland Conditions DAY IV
SADC THEMA Drought Service Drought Monitoring Relating Remotely Sensed Drought Indices to in-situ Crop and Rangeland Conditions M. Masocha (Ph. D), M. D. Shekede, I. Gwitira and M. A. Murwira (Ph. D), K. S. Murwira (Ph. D) Department of Geography and Environmental science
Crop Assessment Results for Maize 2010, Zimbabwe
2009/2010 Vegetation Deviation from Average Dry dekad =(ndvi<(avg-(1. 5*std))
Cropland in Zimbabwe
Number of Dry Dekads Determine Crop Yield in Zimbabwe
SADC THEMA: Drought Service Thank you A. Murwira (Ph. D), K. S. Murwira (Ph. D), M. Masocha (Ph. D), I. Gwitira and M. D. Shekede Department of Geography and Environmental science University of Zimbabwe and Scientific and Industrial Research and Development Centre, Geo-
- Slides: 46