Chapter 1 Preamble

Nighttime Lights (NL) datasets were used into urban planning, econometric downscaling, disaster response, development, and socio-environmental studies since the very beginning of their public release (Pastor-Escuredo, Savy, & Luengo-Oroz, 2015). (Henderson, Storeygard, & Weil, 2012). (Otchia & Asongu, 2019)

In the malaria literature, most of studies used data extracted from NL datasets as a covariate to understand malaria epidemiology patterns (Lechthaler et al., 2019), most commonly used as part of an urbanicity/landscape indexes (Keiser et al., 2004; Kigozi et al., 2015; Siri et al., 2008; A. J. Tatem & Hay, 2004; Tewara, Mbah-Fongkimeh, Dayimu, Kang, & Xue, 2018). Another group of studies use NL datasets to estimate the fine-scale distribution of cases (Battle et al., 2014; Castro et al., 2005; Andrew J. Tatem, 2017; Weiss et al., 2018).

However, there is a scarcity of literature addressing the role of electrification in the dynamics of malaria epidemiology (Pellegrini & Tasciotti, 2016; Tasciotti, 2017). Moreover, most of this literature was focused in the African continent and no evidence of the electrification-malaria relation were reported in South America (SA). The landscape composition and mosquito biting behavior are two key components to hypothesize a contrasting relation in comparison to studies in Africa. In the Amazon Region, where 80% of malaria cases were located in SA (Recht et al., 2017), the rural, riverine context provides suitable breeding sites for Anopheles mosquitos allowing large heterogeneities in spatial and less marked seasonal patterns of malaria transmission. The primary vector in SA is An. darlingi that in comparison with An. gambiae in the African continent, has a more plastic selection of breeding sites, and increased anthropophilic (attracted to bite human) and exopahic (prefer to bite outdoors) behaviors (Conn, Grillet, Correa, & Sallum, 2018; Moreno et al., 2015).

In this context, electrification influence outdoor activities patterns, that in turn directly influence the malaria transmission risk, and also is a surrogate of income and urbanization in the villages. This study will use the Visible Infrared Imaging Radiometer Suite (VIIRS) dataset and a quasi-experimental design to quantify the effect of electrification in the village-level malaria epidemiology in the Peruvian amazon.