Adapting Multi-Index Remote Sensing for Drought Early Warning in Southeast Asia: Reflections on the Ethiopian Highland Study

Abstract

Effective and efficient drought monitoring reinforces food security, particularly under climate change. In the Philippines, this intersects with strategic plans under the Climate Change Commission’s NCCAP and the Department of Agriculture’s AMIA framework. A remotely sensed, index-based drought early warning system, rooted in the VDI-SPEI framework of Alito et al., could quicken the transition from reactive to anticipatory drought governance in the region. Alito et al. offer an exemplary remote sensing-based methodology that is highly transferable to Southeast Asian agroecosystems. Their findings advocate for the integration of multi-index drought detection systems into national disaster risk management and climate adaptation policies. Future scholarly research undertakings should focus on ground-truth validation of VDI in tropical cropping systems and assess its integration with real-time advisory platforms for farmers and local governments.

Dear Editor,

The scholarly work of Alito et al. (2025) entitled: “Characterization of Drought Detection With Remote Sensing Based Multiple Indices and SPIE in Northeastern Ethiopian Highland” is technically rigorous, timely, and highly commendable. The authors were able to integrate satellite-derived vegetation indices including, VDI, VHI, NDVI, VCI, and TCI with meteorological variables. They were also able to integrate the Standardized Precipitation-Evapotranspiration Index (SPEI) from 1981 to 2022. These critical data that they gathered offer a replicable framework for multi-temporal drought monitoring particularly to regions across Asia that are prone to drought. Many rural and sub-urban areas in the Philippines and other Southeast Asian countries can benefit from their findings, specifically those regions that are vulnerable to climate induced hydrological and agricultural drought.

According to the Climate Change Commission (2018), the Philippines, like Ethiopia, is chiefly dependent on rainfed agriculture and is thereby critically exposed to climate extremes such as La Niña that can cause extreme rainfall, or El Niño that can oppositely induce droughts. The authors, Alito, K. T., Kerebih, M. S. & Hailu, D. A., provided a practical and scalable approach through their utilization of MODIS NDVI (MOD13Q1) and LST (MOD11A2). These were crucial in data-scarce and fragmented landscapes and are essential in drought situations in Southeast Asia. In the Philippines being regarded as archipelagic, its urban and sub-urban regions have complex agroecologies and topographic gradients. The country’s urbanization is centralized in Metropolitan Manila, Metropolitan Cebu, and Metropolitan Davao where most of the country’s population is housed. Other complex topographic gradients may be observed in the Cordillera Administrative Region, the Mekong Dela, and Mindanao’s Bukidnon Plateu (Du et al., 2018).

A key finding of the Ethiopian study is the statistically inverse correlation between NDVI and LST (R = −.977, p < .01), confirming the vegetation health deteriorates with rising surface temperatures. This mirrors results from drought-affected rice paddies in Northern Luzon (Perez et al., 2016) and Central Vietnam (Pham et al., 2023), where NDVI declines sharply under thermal stress (Ryu et al., 2022). The observed positive correlation between VHI and rainfall (R² = .996) and with surface moisture (R² = .956) further validates the efficacy of remote sensing-based indices in estimating drought impacts on vegetation (Wei et al., 2021). These insights could improve predictive accuracy in Southeast Asian drought-vulnerable zones, where land-based meteorological networks remain sparse.

It can be emphasized that the integration of the Vegetation Drought Index (VDI), which highly improved the conventional VHI through the inclusion of NDWI and day-light LST differentials, resolves a crucial limitation in drought monitoring especially in the tropical region. The tropical drought monitoring limitation addressed was the lag between thermal stress onset and vegetation response (NDVI). This is crucial in the Philippine context, particularly in multi-cropping systems such as irrigated and rainfed rice, NDWI often reflects real-time moisture availability better than NDVI, which may lag behind actual stress. Thus, VDI’s more immediate responsiveness to soil water deficits offers critical improvements in short-term drought detection (Ha et al., 2022).

Alito et al.’s use of SPEI across multiple time scales (1-, 3-, and 6-months) highlights a critical nuance for Southeast Asian applications. Crops such as maize and rice exhibit phenological sensitivity to short-term water deficits, while long-cycle crops like sugarcane require cumulative water balance monitoring (Pagani, 2025). SPEI-3 and SPEI-6, as applied in this study, are especially useful in assessing cumulative drought stress during critical growth phases, an approach suitable for monsoonal agricultural systems in the ASEAN region.

The authors’ methodological emphasis on multiple linear regression to link remote sensing indices with temperature, soil moisture, LST, and precipitation aligns with data capabilities in many Southeast Asian countries. Agencies like the Thai Meteorological Department, BMKG (Indonesia), and PAGASA (Philippines) can adopt this model to operationalize drought forecasts. Moreover, spatial outputs from MODIS, at 250 m to 1 km resolution, can support localized advisories for provinces heavily reliant on rainfed farming (Gumma et al., 2011), such as Bohol, Isabela, and Cagayan.

Nevertheless, regional adaptation of the Ethiopian approach must consider certain constraints. First, persistent cloud cover in tropical zones may obscure MODIS data during the wet season. Hence, data fusion techniques combining MODIS, Sentinel-2, and ECOSTRESS could improve reliability (Xue et al., 2022). Second, land use heterogeneity is more pronounced in Southeast Asia, requiring recalibration of VCI and VDI thresholds to reflect paddy-upland interfaces, agroforestry mosaics, and mixed cropping systems. Machine learning-based regression models such as gradient boosting or random forest may also offer enhanced prediction accuracy in these complex environments (Tao et al., 2022).

We agree with the authors’ assertions that effective and efficient drought monitoring reinforces food security, particularly under climate change. In the Philippines, this intersects with strategic plans under the Climate Change Commission’s NCCAP and the Department of Agriculture’s AMIA framework. A remotely sensed, index-based drought early warning system, rooted in the VDI-SPEI framework of Alito et al., could quicken the transition from reactive to anticipatory drought governance in the region.

In conclusion, Alito et al. offer an exemplary remote sensing-based methodology that is highly transferable to Southeast Asian agroecosystems. Their findings advocate for the integration of multi-index drought detection systems into national disaster risk management and climate adaptation policies. Future scholarly research undertakings should focus on ground-truth validation of VDI in tropical cropping systems and assess its integration with real-time advisory platforms for farmers and local governments.