Afghanistan Drought Early Warning Decision Support (AF-DEWS) Tool

Data Catalogue

• ERPAS
• IMERG
• CHIRPS
• MODIS
• SMAP
• LandScan
• Land Use Land Cover
• SWI

Indices

Climate, vegetation, and drought can be monitored many different ways through ground observations, gridded weather and climate data, and airborne and satellite remote sensing. Access to GEE’s satellite image and meteorological collections allow for efficient near real-time drought monitoring and early warning using indicators such as surface temperature, vegetation, precipitation, snow cover, surface water, soil moisture, evapotranspiration and many more.

Basic Indices

The product calculates accumulated rainfall using satellite-based rainfall i.e CHIRPS for a given period example weekly, monthly and seasonal.

Reference: Funk C.C., Peterson P.J., Landsfeld M.F., Pedreros D.H., Verdin J.P., Rowland J.D., Romero B.E., Husak G.J., Michaelsen J.C., and Verdin A.P. 2014. A quasi-global precipitation time series for drought monitoring: U.S. Geological Survey Data Series 832, 4 p.

Rainfall anomaly index (RAI) is a normalized precipitation values based upon the historical rainfall of a particular location. It is a comparison of current rainfall variation from the historical period.

RAI = (RFCurrent - RFMean)/(RFStdDev)

Where, RFCurrent is Current rainfall, RFMean and RFStdDev are the mean and standard deviation of long-term rainfall.

Reference: Kraus, E.B., 1977. Subtropical droughts and cross-equatorial energy transports. Monthly weather review, 105(8), pp.1009-1018.

The Soil Water Anomaly Drought Index (SWADI) monitors the changes of soil water anomaly with reference to long-term extremities. It is pixel based calculation that efficiently separates the short term changes in SWI from the long-term ecological changes. SWADI has proved to more efficient in identifying drought condition and following is the equation for calculating SWADI

SWADI_ijk = (SWI_ijk - SWI(mean)_ijn)/SWI(std)_ijn

Where, SWADI_ijk is SWADI for pixel i in composite j of year k; SWI_ijk is SWI for pixel i in composite j of year k; SWI(mean)_ijn is the long-term mean of SWI for pixel i in composite j and SWI(std)_ijx is the long-term standard deviation of SWI for pixel i in composite j. The value range varies between “-3” to “+3”. The value nearby “-2” reveals extreme drought situation and nearby +1 expresses healthy situation.

Although there are, several vegetation indices, one of the most widely used is the Normalized Difference Vegetation Index (NDVI). NDVI values range from +1.0 to -1.0. Areas of barren rock, sand, or snow usually show very low NDVI values (for example, 0.1 or less). Sparse vegetation such as shrubs and grasslands or senescing crops may result in moderate NDVI values (approximately 0.2 to 0.5).

NDVI = (NIR-RED)/(NIR+RED)

Where NIR = near infrared band, RED = red band

Reference: Rouse Jr, J.W., Haas, R.H., Schell, J.A. and Deering, D.W., 1973. Monitoring the vernal advancement and retrogradation (green wave effect) of natural vegetation.

NDVI anomaly is normalized vegetation values based upon the historical vegetation of a particular location. It is a comparison of current vegetation variation from the historical period.

NDVI Anomaly = (NDVI_Current - NDVI_Mean)/(NDVI_StdDev)

Where, NDVI_Current is current vegetation, NDVI_Mean and NDVI_StdDev are mean and slandered devotion of long-term vegetation condition

EVI can also be used to quantify vegetation greenness. However, EVI corrects for some atmospheric conditions and canopy background noise and is more sensitive in areas with dense vegetation or high biomass. EVI can be calculated as below

EVI = G * ((NIR - R) / (NIR + C1 * R – C2 * B + L))

The formula incorporates an “L” value to adjust for canopy background, “C” values as coefficients for atmospheric resistance, and values from the blue band (B). These enhancements allow for index calculation as a ratio between the R and NIR values, while reducing the background noise, atmospheric noise, and saturation in most cases.

Reference: Huete, A., Didan, K., Miura, T., Rodriguez, E.P., Gao, X. and Ferreira, L.G., 2002. Overview of the radiometric and biophysical performance of the MODIS vegetation indices. Remote sensing of environment, 83(1-2), pp.195-213.

EVI anomaly is normalized vegetation values based upon the historical vegetation of a particular location. It is a comparison of current vegetation variation from the historical period.

EVI Anomaly = (EVI_Current - EVI_Mean)/(EVI_StdDev)

Where, EVI_Current is current vegetation, EVI_Mean and EVI_StdDev are mean and slandered devotion of long-term vegetation condition

The normalized difference water index can be utilized for evaluating vegetation liquid water contents or water inundated areas (Gao, 1996). NDWI is useful for evaluating reflectance from vegetation canopies that have similar scattering properties, but slightly different liquid water absorption due to canopy water content. As a result, NDWI is sensitive to changes in liquid water content of vegetation canopies and open water areas. The common range of NDWI for green vegetation is -0.1 to 0.4

There are different way of NDWI calculation as follows

NDWI = (GREEN-NIR)/(GREEN+NIR)

Where NIR = near infrared band, GREEN = green band

Reference: Gao, B.C., 1996. NDWI - A normalized difference water index for remote sensing of vegetation liquid water from space. Remote Sensing of Environment, 58, 257–266.

NDWI anomaly is normalized NDWI values based upon the historical NDWI of a particular location. It is a comparison of current NDWI variation from the historical period.

NDWI Anomaly = (NDWI_Current - NDWI_Mean)/(NDWI_StdDev)

Where, NDWI_Current is current vegetation, NDWI_Mean and NDWI_StdDev are mean and slandered devotion of long-term vegetation condition

The near-infrared (NIR: 841–876 nm) and the shortwave infrared (SWIR1: 1628–1652 nm) bands are used to calculate LSWI. The NIR spectral region serves as a moisture reference band and the SWIR spectral domain is used as the moisture measuring band. SWI was found to correspond well with the drought severities in many studies. LSWI can be calculated as below

LSWI = (NIR-SWIR1)/(NIR+SWIR1)

Higher LSWI values represent wet condition and lower values indicated dry condition.

Reference: Xiao, X.; Hollinger, D.; Aber, J.; Goltz, M.; Davidson, E.A.; Zhang, Q.; Moore, B., III. Satellite-based modeling of gross primary production in an evergreen needle leaf forest. Remote Sens. Environ. 2004, 89, 519–534.

Calculation of the Normalize Difference Snow Index (NDSI) is mainly to understand presence of snow pixels in satellite images and more accurately, it is known as detection of Fractional Snow Cover (FSC).

NDSI = (Green-SWIR) / (Green+SWIR)

The pixels which are having NDSI value >0 is considered as to have some snow.

Reference: Hall D.K., Riggs, G.A., Salomonson V.V., DiGirolamo N.E., and Bayr K.J., 2002. MODIS snow-cover products. Remote Sensing of Environment 83: 181–194.

Intermediate Index

A dry spell is defined as the number of consecutive days with a daily precipitation amount below a certain threshold, such as 0.1, 1, 5, 10 mm, preceded and followed by at least one day with rainfall exceeding the threshold. The app uses rainfall product from CHIRPS to calculate the dry spell for specific time period ranging from few days to months.

Reference: Suppiah, R. and Hennessy, K.J., 1998. Trends in total rainfall, heavy rain events and number of dry days in Australia, 1910–1990. International Journal of Climatology: A Journal of the Royal Meteorological Society, 18(10), pp.1141-1164.

The Standardized Precipitation Index (SPI) the measure of the number of standard deviations of observed cumulative precipitation deviates from the climatological average. SPI calculation involves the estimation of probability distribution function over desired rainfall intervals. SPI can be calculated for any time scale vary from monthly to multi-months.

Reference: McKee T.B., Doesken N.J., and Kliest J. 1993. The relationship of drought frequency and duration to time scales. In Proceedings of the 8th Conference of Applied Climatology, 17-22 January, Anaheim, CA. American Meteorological Society, Boston, MA. 179-184.

The PCI compares present rainfall reference with long-term extreme rainfall values. The PCI value range varies between 0-100, where the value nearby 0 represents extreme stress, while values close to 100 expresses a healthy situation. RF_min and RF_max are the long-term minimum and maximum RF for a given pixel and the RF_i is the current RF for the same pixel.

PCI = ((RF_i - RF_min)/(RF_max - RF_min))X 100

Reference: Kogan F.N. 1995. Application of vegetation index and brightness temperature for drought detection. Advances in Space Research, 15(11): 91–100. DOI: 10.1016/0273-1177(95)00079-T.

TCI derived from long-term historical extremes of LST and it assigns the values as normalized ranging from 0-100. Where values close to 0 depicts stress condition for vegetation growth and values closer to 100 reflects favorable conditions for vegetation growth. LST_min and LST_max are the long-term minimum and maximum LST for a given pixel and the LST_i is the current LST for the same pixel.

LST = ((LST_i - LST_min)/(LST_max - LST_min))X 100

Reference: Kogan F.N. 1995. Application of vegetation index and brightness temperature for drought detection. Advances in Space Research, 15(11): 91–100. DOI: 10.1016/0273-1177(95)00079-T.

Pixel-based VCI calculation is more effective to identify the drought condition irrespective of the ecological region. The range of VCI values varies between 0-100 and the value 0 reveals extreme stress while 100 expresses healthy vegetation. NDVI_min and NDVI_max are the long-term minimum and maximum NDVI for given pixel and the NDVI_i is the current NDVI for the same pixel.

VCI = ((NDVI_i - NDVI_min)/(NDVI_max - NDVI_min))X 100

Reference: Kogan F.N. 1995. Application of vegetation index and brightness temperature for drought detection. Advances in Space Research, 15(11): 91–100. DOI: 10.1016/0273-1177(95)00079-T.

Vegetation Health Index (VHI) is an index characterizes the health of the vegetation by integrating NDVI and Temperature so that stressed conditions are linked to lower than normal NDVI and higher than normal temperature. The VHI and the sub-indices are used for various purposes, of which its applicability in detecting and monitoring the phenomenon of drought.

VHI = αVCI + (1-α)TCI

where VCI is the Vegetation Condition Index, TCI is the Thermal Condition Index and α is 0.5

Reference: Kogan F.N. 1990. Remote sensing of weather impacts on vegetation in non-homogeneous areas. International Journal of Remote Sensing, 11: 1405–1419. DOI: 10.1080/01431169008955102.

NDSI anomaly is normalized snow cover values based upon the historical representation of snow of a particular location. It is a comparison of current snow variation from the historical period. Snow Condition Index (SCI) represents snow condition similar like VCI and can be calculated as below

SCI = ((NDSI_i - NDSI_min)/(NDSI_max - NDSI_min))X 100

The SCI value range varies between 0-100, where the value nearby 0 represents extreme lower snow with respect to longterm, while values close to 100 expresses a greater snow condition. NDSI_min and NDSI_max are the long-term minimum and maximum NDSI for a given pixel and the NDSI_i is the current NDSI for the same pixel.

Soil Moisture Condition Index (SMCI) represents soil moisture condition similar like VCI and can be calculated as below

SMCI = ((SM_i - SM_min)/(SM_max - SM_min))

The SMCI value range varies between 0-100, where the value nearby 0 represents extreme soil moisture stress, while values close to 100 expresses a healthy situation. SM_min and SM_max are the long-term minimum and maximum SM for a given pixel and the SM_i is the current soil moisture for the same pixel.

Reference: Liu, Q., Zhang, S., Zhang, H., Bai, Y. and Zhang, J., 2020. Monitoring drought using composite drought indices based on remote sensing. Science of The Total Environment, 711, p.134585.

Moisture Adequacy Index (MAI) is the ratio of actual evapotranspiration (AET) to the potential evapotranspiration (PET). Thus, MAI represents moisture effectivity which has impact on vegetation in relation to the climate. The range of MAI values varies between 0-1; value near to 0 reveals extreme stress while 1 expresses ample amount of soil moisture availability.

MAI = AET/PET

Reference: Thornthwaite, C.W. and Mather, J.R., 1955. The water balance publications in Climatology, 8 (1). DIT, Laboratory of climatology, Centerton, NJ, USA.

Composite Index

IDSI is based on the integration of multi-scale VCI, PCI and TCI using data fusion techniques. The IDSI values range from 0 to 100, which is similar to all the prior indices produced. Values close to 0 represent extreme drought, whereas values near to 100 reveal healthy vegetation conditions (no drought). Integrated Drought Severity Index (IDSI) calculated using the following equation.

IDSI_ijk, VCI_ijk, TCI_ijk and PCI_ijk are IDSI, VCI, TCI and PCI values for a pixel (i) in a composite (j) of a year (k). L is the normalization factor to keep the output value in expected range and c is a constant to avoid a null in the denominator.