Integrated drought monitoring index: A tool to monitor agricultural drought by using time-series datasets of space-based earth observation satellites
Introduction
Food and agriculture are vital for achieving Sustainable Development Goals (SDGs) to end the poverty, and hunger (SDG1 and SDG2) and tackle the climate change (SDG11) by 2030 through maintain natural resource base (SDSN, 2013, FAO, 2016). Furthermore, to overcome the challenges like rising global population, global climate change, depletion of irrigation water, land degradation, it is necessary to increase the agricultural production, especially in rainfed arable land (Anderson et al., 2016a, Anderson et al., 2016b). Assessment of spatio-temporal dynamics of agricultural drought is of great interest as it has wide variability in time and space (Xue and Su, 2017) and has far-reaching impact on global food security and sustainable development (Hu et al., 2019, Kogan et al., 2019). Agricultural drought leads to decline in crop yields due to adverse weather conditions such as erratic rainfall pattern, the rise of global temperature (IPCC, 2018), and the associated decline of soil moisture content (Du et al., 2018). On the other side, the successful assessment and monitoring of agricultural drought requires frequent and internally consistent records of evidence on a range of biophysical variables (Kogan, 2001). Multi-variate analysis and geo-statistical methods commonly used in the assessment of spatio-temporal dynamics of regional droughts (Haining et al., 2010). The common approaches to depict the characteristics of agricultural drought are the region of influence approach (Zrinji and Burn, 1994), the entropy approach (Rajsekhar et al., 2012); the residuals method (Choquette, 1988) and the principal component analysis (PCA) method (Hazaymeh and Hassan, 2017). As agricultural drought generally begins with deficiencies in precipitation, then leads to deficiencies of soil moisture, higher land surface temperature, and at last, it adversely affects the vegetation growth. Therefore, the parameters derived from precipitation, soil and vegetation play a critical role to assess and monitor the agricultural drought. Traditional methods are not fully capable to assess and monitor the agricultural droughts (AghaKouchak and Nakhjiri, 2012) as they are limited in a region, often offers inaccurate measures, moreover, it is hard to get the near-real-time data to predict and quantify them (Easterling, 2013).
In recent years, the time-series datasets derived from space-based earth observation satellites play a key role in assessment and monitoring of agricultural drought as they provide wide and temporal coverage (Muthumanickam et al., 2011, Reddy et al., 2020). Remote sensing and Geographic Information System (GIS) techniques have been broadly used as an ideal tool in order to assess agricultural drought risk over a large area (Belal et al., 2012, Du et al., 2013, Sánchez et al., 2018). Moreover, which helps to monitor the agricultural drought at distinct stages like before, during or after the event (AghaKouchak et al., 2015, Himanshu et al., 2015). By means of satellite remote sensing and GIS technologies, many authors monitored the impact of drought on agriculture in both time and space (Gebrehiwot et al., 2011, Zhang et al., 2017). Moderate resolution imaging spectroradiometer (MODIS) provides near real-time remote sensing datasets and its integration with GIS offers an enhanced agricultural drought assessment (Qian et al., 2016, Zambrano et al., 2016, Santos et al., 2017, Reddy et al., 2020). With modern technological advances in the field of remote sensing, the latest indices of time-series satellite data that are real-time have fully toppled the conventional agricultural drought indices (Elhag and Zhang, 2018). In addition, remote sensing has a tremendous ability to provide comprehensive coverage across a large region with a spatial resolution that varies from a few meters to a few kilometres (Himanshu et al., 2015). Among the indices derived from remote sensing, the normalized difference vegetation index (NDVI) (Tucker et al., 2001) is the most robust and widely used index with the capabilities of measuring terrestrial vegetation's photosynthetic ability during the growing season (Karnieli et al., 2010, Son et al., 2012, Zhang et al., 2016, Okin et al., 2018). Agricultural drought has a relation with precipitation, NDVI, land surface temperature (LST), and soil moisture anomalies in different seasons and regions; however, in some cases, their relationship and correlation are not straight. Table 1 shows various drought indices derived from earth observation satellites and used in the assessment and monitoring of agricultural drought across the globe.
PCA is a linear transformation that eliminates uncertainty by converting the original component space, enabling the display of drought details without correlation in a new component space (MaChado-MaChado et al., 2011). It was widely used in all forms of analysis since it is a simple, non-parametric method of extracting relevant data from complex datasets (Wold et al., 1987, Zabiri et al., 2007). In recent years, the PCA method (Rencher, 1998) was adopted in developing drought monitoring tools through the aggregation of several hydro-climatic variables into a single drought indicator as it can effectively hold the characteristics of datasets through a lower dimension with a simplified structure (Keyantash and Dracup, 2004, Martins et al., 2012, Liu et al., 2016, Bayissa et al., 2018). Martins et al. (2012) used the PCA approach to study the spatial variance of drought in Portugal by reducing the size and extracting structured data from a large number of time-series drought indices. Liu et al. (2016) used the PCA approach to identify the specific locations and sub-regions, taking into account the drought characteristics to provide a regional view of drought conditions across the Loess Plateau, China. Bayissa et al. (2018) developed a vigorous tool for drought monitoring with the PCA by integrating conventional climate and satellite-based drought indices to describe the severity and spatial distribution of Ethiopia's historic drought events. Gocic and Trajkovic (2014) was used the PCA model to capture spatio-temporal drought patterns in Serbia by reducing dimensionality in a time-series of standardized precipitation index (SPI). These studies reported the temporal variation of agricultural drought as expressed in the PC scores obtained from the PCA and support the potential application of PCA approach in aggregating several input variables into a single aggregate drought index. The commonly used individual and many composite indices were not capable to monitor and assess the agricultural drought events. As rainfed agriculture is prone to seasonal variation of climate components, the integration of multiple climatic and agricultural drought variables derived from space-based earth observation satellite datasets provides a comprehensive view of drought conditions for its assessment and monitoring (Sepulcre-Canto et al., 2012).
In India, the agriculture sector is one of the primary sources of livelihood for about 68% of the total population (Dutta et al., 2015), and it contributes about 17% of the country's gross domestic product (GDP) (Arjun, 2013). Indian agriculture mainly depends on south-west monsoons (June to September) (Kumar et al., 2013) and any changes in monsoon pattern adversely affect the crop conditions and overall economy of the nation (Udmale et al., 2014). There is a growing concern over increasing weather aberrations, and drought patterns with the increasing climate variability in the semi-arid regions of India (Vyas and Bhattacharya, 2020). The Intergovernmental Panel on Climate Change (IPCC) predicted the increase in frequency of droughts over the semi-arid regions of India (IPCC, 2013). The agriculture in Tamil Nadu state, however, vastly depends on northeast monsoon (October to December), which contributes about 60% of the total annual rainfall of the state (Priya and Manimannan, 2014). Any scarcity of rainfall during northeast monsoon results acute water shortage and severe agricultural drought in the state. The present study was aimed firstly to develop comprehensive integrated drought monitoring index (IDMI) as a new tool by using time-series datasets derived from space-based earth observation satellites and PCA, secondly to demonstrate and validate its robustness in monitoring of spatio-temporal patterns of agricultural drought in Tamil Nadu state of Indian Peninsula.
Section snippets
Study area
Tamil Nadu state in the Indian peninsula lies between 08° 00′ and 13° 30′ northern latitudes and 76° 00′ and 80° 18′ of eastern longitudes with an area of 13.00 million hectare (Mha) and occupies about 3.96% of the country's total geographical area (TGA) (Fig. 1). The state is bounded by the Bay of Bengal on its east, Western Ghats on its west, Indian Ocean on its south, Andhra Pradesh state on north, and Karnataka state on its northwest. About 90% area of the state is under the semi-arid
Spatio-temporal variability of 3-month SPI
The spatio-temporal analysis of seasonal 3-month SPI for the period from 2000 to 2016 shows that study area was experienced above near-normal (−0.99 to 0.99) dry conditions in the majority of the years. However, the extreme wet condition was noticed in the year 2005, followed by 2004 and 2006 (Fig. 5). The analysis shows a recurring dry condition continuously from 2000 to 2003. During this period, the near-normal condition (−0.99 to 0.99) was noticed in large parts of the study area. During the
Intra-seasonal variability of IDMI and 3-month SPI in wet year
To understand the intra-seasonal variability of drought intensity of Tamil Nadu, the spatio-temporal variability of 3-month SPI and IDMI for the wet year (2005) during the northeast monsoon season was analysed. The analysis shows that moderately wet to very wet conditions with mean seasonal 3-month SPI of 1.7 were observed. During the northeast monsoon season, the intra-seasonal analysis of 3-month SPI shows the relatively near-normal conditions in the month of November as compared to the wet
Conclusions
The analysis of spatio-temporal variability of 3-month SPI for the dry year 2016 shows that moderate to extremely dry conditions (−2.0 and less) in the central part of the study area covers mainly in Kongu uplands and parts of the Cauvery delta region of the study area more particularly in the months of October and November. The analysis of IDMI for the period from 2000 to 2016 clearly shows moderate to extremely dry conditions during the dry year 2016, especially in central, northern and
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
The authors are thankful to NASA LPDAAC, ESA-CCI and CHC for providing time-series MODIS, soil moisture and precipitation datasets at free of cost for this research. The authors are thankful to the Director, ICAR-National Bureau of Soil Survey and Land Use Planning (NBSS&LUP), Nagpur for extending the facilities to carry out the work. The authors also thankful to Tamil Nadu Public Works Department (PWD), Chennai for providing the rainfall data. This research did not receive any specific grant
References (89)
- et al.
The evaporative stress index as an indicator of agricultural drought in Brazil: an assessment based on crop yield impacts
Remote Sens. Environ.
(2016) - et al.
Vegetation water content estimation for corn and soybeans using spectral indices derived from MODIS near-and short-wave infrared bands
Remote Sens. Environ.
(2005) - et al.
A simple method for reconstructing a high-quality NDVI time-series data set based on the Savitzky-Golay filter
Remote Sens. Environ.
(2004) ESA CCI Soil moisture for improved earth system understanding: state-of-the art and future directions
Remote Sens. Environ.
(2017)- et al.
A comprehensive drought monitoring method integrating MODIS and TRMM data
Int. J. Appl. Earth Obs. Geoinf.
(2013) - et al.
Assessment of agricultural drought in Rajasthan (India) using remote sensing derived Vegetation Condition Index (VCI) and Standardized Precipitation Index (SPI)
Egypt J. Remote Sens. Space Sci.
(2015) - et al.
Derivation of a shortwave infrared water stress index from MODIS near-and shortwave infrared data in a semiarid environment
Remote Sens. Environ.
(2003) NDWI- A normalized difference water index for remote sensing of vegetation liquid water from space
Remote Sens. Environ.
(1996)- et al.
Spatial and temporal assessment of drought in the northern highlands of Ethiopia
Int. J. Appl. Earth Obs. Geoinf.
(2011) - et al.
Spatiotemporal characteristics of drought in Serbia
J. Hydrol.
(2014)
Combination of multi-input remote sensing data for drought monitoring over Southwest China
Int. J. Appl. Earth Obs. Geoinf.
Agricultural drought monitoring using European Space Agency Sentinel 3A land surface temperature and normalized difference vegetation index imageries
Agr. Forest Meteorol.
Assessing vegetation response to drought in the northern Great Plains using vegetation and drought indices
Remote Sens. Environ.
Application of vegetation index and brightness temperature for drought detection
Adv. Space Res.
Spatiotemporal analysis of multiscalar drought characteristics across the loess plateau of China
J. Hydrol.
Estimating crop water deficit using the relation between surface-air temperature and spectral vegetation index
Remote Sens. Environ.
Monitoring agricultural drought for arid and humid regions using multi-sensor remote sensing data
Remote Sens. Environ.
Integrated remote sensing approach to global agricultural drought monitoring
Agr. Forest Meteorol.
A simple interpretation of the surface temperature/vegetation index space for assessment of surface moisture status
Remote Sens. Environ.
Monitoring agricultural drought in the Lower Mekong Basin using MODIS NDVI and land surface temperature data
Int. J. Appl. Earth Obs. Geoinf.
Spatiotemporal changes of vegetation and their responses to temperature and precipitation in upper Shiyang river basin
Adv. Space. Res.
Red and photographic infrared linear combinations for monitoring vegetation
Remote Sens. Environ.
Farmer’s perception of drought impacts, local adaptation and administrative mitigation measures in Maharashtra State, India
Int. J. Disaster Risk Reduct.
Principal component analysis
Chemom. Intell. Lab. Syst.
Drought monitoring and analysis in China based on the Integrated Surface Drought Index (ISDI)
Int. J. Appl. Earth Obs. Geoinf.
Characterization of forest types in North eastern China, using multi-temporal SPOT-4 VEGETATION sensor data
Remote Sens. Environ.
Response of vegetation to different time scales drought across China: spatiotemporal patterns, causes and implications
Glob. Planet. Chang.
Flood frequency-analysis for ungauged sites using a region of influence approach
J. Hydrol.
Remote sensing of drought: progress, challenges and opportunities
Rev. Geophys.
A near real-time satellite-based global drought climate data record
Environ. Res. Lett.
Addressing the yield gap in rainfed crops: a review
Agron. Sustain. Dev.
Indian agriculture-status, importance and role in Indian economy
Int. J. Agr. Food Sci. Tech.
Developing a satellite-based combined drought indicator to monitor agricultural drought: a case study for Ethiopia
Gisci. Remote Sens.
Drought risk assessment using remote sensing and GIS techniques
Arab. J. Geosci.
Satellite-based, multi-indices for evaluation of agricultural droughts in a highly dynamic tropical catchment, Central Vietnam
Water.
Monitoring and assessment of drought focused on its impact on sorghum yield over Sudan by using meteorological drought indices for the period 2001–2011
Remote Sens.
Using ArcGIS Geostatistical Analyst
Food and Agriculture: Key to achieving the 2030 agenda for sustainable development
Desvendandoos Mistérios do Coeficiente de Correlação de Pearson (r)
Revista Política Hoje
Correlação de Pearson e análise de trilhaidentificandovariáveis para caracterizar porta-enxerto de Pyruscommunis L
Ciência e Agrotecnologia.
A five-year analysis of MODIS NDVI and NDWI for grassland drought assessment over the central Great Plains of the United States
Geophys. Res. Lett.
Cited by (34)
Drought stress and its characteristics in China from 2001 to 2020 considering vegetation response and drought creep effect
2024, Journal of Hydrology: Regional StudiesVariability and drivers of grassland sensitivity to drought at different timescales using satellite image time series
2023, Agricultural and Forest MeteorologyA new integrated index for drought stress monitoring based on decomposed vegetation response factors
2023, Journal of HydrologyEvaluating the impact of different normalization strategies on the construction of drought condition indices
2022, Agricultural and Forest MeteorologyCitation Excerpt :The propagation of spatial and temporal effects from the raw data to single indices and then to Spat-CCIs and Temp-CCIs exhibit adverse impacts not only on the SCIs but also on the CCIs. CCIs can integrate multiple SCIs and provide a comprehensive assessment of drought conditions (Arun Kumar et al., 2021; Hao and Singh, 2015; Patel and Yadav, 2015). Nevertheless, CCI variants also accumulate the adverse effects of spatiotemporal normalization.