Distribution of Anophelinae (Diptera: Culicidae) and challenges for malaria elimination in Brazil

Abstract

In 1909, Arthur Neiva published an article titled “Contribuição para os estudos dos dipteros. Observação sobre a biolojia e sistematica das anofelinas brasileiras e suas relações com o impaludismo”, highlighting the biology, ecology, and distribution of Anophelinae mosquitoes and the need for more taxonomic studies in Brazil. This came 11 years after Ronald Ross and Grassi demonstrated mosquito roles in transmitting Plasmodium to birds and humans. Despite considerable advances in the understanding of Anophelinae species, knowledge remains insufficient given the complexity of Brazil’s ecosystems, the intensified anthropogenic environmental changes since the mid-20th century, and the persistent public health challenges posed by malaria. This perspective article presents the distribution of Plasmodium vectors and potential vector species in Brazil using climate variables and a maximum entropy model. Geographical distribution maps of Anophelinae species, including putative species, are provided. The article also discusses the current knowledge of vector species distribution in relation to Brazil’s malaria elimination plan, along with the ecological and anthropogenic factors influencing vector distribution.

In 1909, Arthur Neiva published an article in Portuguese and German titled “Contribuição para os estudos dos dipteros. Observação sobre a biolojia e sistematica das anofelinas brasileiras e suas relações com o impaludismo”. ¹ This followed the groundbreaking 1898 discovery by Sir Ronald Ross, who demonstrated the role of mosquitoes in bird malaria transmission, leading to significant research on the systematics, distribution, and bionomics of Anophelinae Grassi mosquitoes globally. At that time, the genera with representatives in Brazil were ― Myzomyia Blanchard (currently synonymous with Cellia Theobald), Cyclolepteron Theobald (currently synonymous with Anopheles Meigen), Stethomyia Theobald, Myzorynchella Theobald (currently synonymous with Nyssorhynchus Blanchard), Arribalzagia Blanchard (currently synonymous with Anopheles), Cellia, Chagasia Cruz and Manguinhosia Cruz (currently synonymous with Nyssorhynchus).

In his article, Neiva detailed the known geographic distribution of Brazilian mosquito species (Fig. 1) and discussed various aspects of their biology, ecology, and blood-feeding behaviors. Upon reviewing the article, two aspects stand out as particularly striking. First, the meticulous detail with which the knowledge of these mosquitoes was presented. Second, when addressing the challenges of locating larval habitats, Neiva emphasized the scarcity of studies on the biology of Anopheles species in Brazil - a gap that, 115 years later, remains largely unaddressed. An important early study worth highlighting is the field entomological investigations conducted from 1939 to 1944 by Deane, Causey, and Deane. Using various sampling methods, their research provided invaluable data on the biology, ecology, and distribution of 36 anopheline species identified in the northeastern and Amazon regions of Brazil, covering over 4.5 million square kilometers. ² Despite advances in scientific knowledge, research has predominantly focused on species of public health significance, leaving gaps in the broader understanding of Anophelinae taxonomy, biology, and ecology in Brazil.

Fig. 1: Neiva’s 1909 map detailing the known geographic distribution of Brazilian mosquito species.

In honor of the 115th anniversary of Memórias do Instituto Oswaldo Cruz, and with the respect Professor Neiva’s pioneering work deserves, this perspective article focuses on the potential geographic distribution of Nyssorhynchus darlingi (Root) and Kerteszia cruzii (Dyar & Knab) in Brazil. Additionally, distribution maps for other anopheline species are provided to showcase the diversity of species found in the country. The distribution analyses are primarily based on recent data from field collections, from published literature records, and specimens from the Entomological Collection of the Faculdade de Saúde Pública at the Universidade de São Paulo (CER-FSP), Brazil. The study also reflects on how gaps in knowledge of species distributions complicate malaria vector surveillance efforts, particularly in the context of Brazil’s malaria elimination initiatives.

Malaria as a public health concern

Malaria remains a significant public health concern in Brazil, particularly in the Amazon region, where over 99% of cases are reported. ³ Despite considerable efforts to control and reduce the disease incidence, malaria continues to affect several thousands of people annually, leading to loss of productivity, poverty, and jeopardizing the local health services in endemic countries. ^{4 , 5 , 6} Malaria impact is most pronounced in indigenous people, rural communities and remote illegal mining, where lack or poor access to fast diagnostic, antimalarial drugs, and molecular tools to monitor transmission, challenge the effective control and elimination of the disease. ^{7 , 8} In the Amazon, human migration patterns, especially the interstate travel network of infected people, can have an impact on the Brazil goals of malaria elimination by 2035. ³ Similarly, the risk of malaria parasite propagation is not negligible by nonimmune travelers within the Amazon and from the Amazon to outside regions where malaria was either eliminated or transmission is low. ⁹

Malaria has been a significant public health challenge in Brazil for centuries, with its history deeply intertwined with the country’s socio-economic development and environmental changes. ¹⁰ In the early 20th century, malaria was widespread throughout much of Brazil, particularly in the Parana River, Amazon River basin, and on the Atlantic coast, where the highest transmission rates occurred. ¹¹ The invasion of Anopheles (Cellia) gambiae Giles, later identified as Anopheles (Cellia) arabiensis Patton ¹² in the northeast state of Rio Grande do Norte and its spread to Ceará State was a major challenge to malaria control in the 1930s. ^{13 , 14} In the early 1940s, after approximately 14,000 deaths and 150,000 cases in the northeastern states, An. arabiensis was eliminated from Brazil. ^{15 , 16} Efforts to control malaria intensified in the mid-20th century, leading to notable declines in cases because of the widespread use of insecticide improved surveillance, and public health campaigns. ¹¹ The resurgence of cases in the 1980s highlighted the limitations of these strategies, particularly in regions undergoing rapid deforestation, infrastructure development, and population migration, primarily associated with the expansion of agrobusiness and exploitation of basic commodities for global trade. ^{10 , 17}

The Amazon region remains the epicenter of malaria transmission in Brazil, accounting for most cases nationwide. Within this region, continuous transmission zones are typically concentrated in remote, forested areas, particularly along rivers, where Ny. darlingi, the dominant mosquito vector, thrives. ¹⁸ These areas often coincide with regions experiencing rapid deforestation, illegal mining, agricultural expansion, and infrastructure development, all of which create favorable conditions for mosquito habitats. Migration and mobility of human populations further exacerbate transmission, as workers and settlers in newly developed areas can spread Plasmodium spp. to regions with limited healthcare access. ³ States such as Amazonas, Acre, and Pará consistently report the highest malaria incidences, with indigenous communities and rural populations being the most vulnerable. ¹⁹ Despite ongoing control efforts, such as the distribution of insecticide-treated nets and the expansion of diagnostic and treatment services, the complex socio-environmental dynamics of the Amazon region pose significant challenges to malaria elimination. ¹⁹

Taxonomy of Anophelinae

The subfamily Anophelinae Grassi typically includes the genera Anopheles Meigen, Bironella Theobald, and Chagasia Cruz. ²⁰ The species of the genus Anopheles were classified into the subgenera Anopheles Meigen, Baimaia Harbach, Rattanarithikul, Harrisson, Cellia Theobald, Christya Theobald, Kerteszia Theobald, Lophopodomyia Antunes, Nyssorhynchus Blanchard, and Stethomyia Theobald. However, Foster et al. ²¹ elevated Kerteszia, Lophopodomyia, Nyssorhynchus and Stethomyia to the genus level, resulting in a total of seven genera in Anophelinae. The genus Anopheles, in turn, now includes the subgenera Anopheles, Baimaia, Cellia, and Christya. The species of public health importance belong to the genus Anopheles, subgenera Anopheles (cosmopolitan) and Cellia (Old World), genus Kerteszia (Neotropics), and genus Nyssorhynchus (Neotropics and southern Nearctic).

Mosquitoes of the subfamily Anophelinae are found worldwide, except in few locations and islands where the ecological factors and temperature are inadequate for mosquitoes. ²² Females are responsible for transmitting four species and two subspecies of the protozoan genus Plasmodium Marchiafava & Celli that cause human malaria. ²³ Approximately 70 out of approximately 500 valid species are dominant vectors of Plasmodium. ²⁴ In addition, certain species play a role in transmitting arboviruses, such as O’nyong-nyong, ^{25 , 26} Guaroa, ²⁷ and might be efficient vectors of Mayaro and Sindbis viruses based on results of recent laboratory experiments. ²⁸ In the Americas, nine species or species of complexes are dominant vectors of Plasmodium spp.: Nyssorhynchus albimanus (Wiedemann), Nyssorhynchus albitarsis complex, Nyssorhynchus aquasalis (Curry), Ny. darlingi (Root), An. (Anopheles) freeborni Aitken, Nyssorhynchus marajoara (Galvão & Damasceno), Nyssorhynchus nuneztovari complex, An. (Anopheles) pseudopunctipennis complex, and An. (Anopheles) quadrimaculatus Say. ^{24 , 29}

In Brazil, Ny. darlingi is the dominant vector in areas across the Amazon River basin, Ny. aquasalis is an effective vector in areas on the Atlantic coast, while Ke. cruzii (Dyar & Knab), Ke. bellatrix (Dyar & Knab), and Kerteszia homunculus Komp are vectors in the Atlantic Forest biome. ³⁰ In addition, there are other species that were found naturally infected and may be local vectors of Plasmodium vivax (Grassi & Feletti), and/or Plasmodium falciparum (Welch): Anopheles (Anopheles) peryassui Dyar & Knab, Nyssorhynchus benarrochi B, Nyssorhynchus tadei (Saraiva & Scarpassa), Nyssorhynchus oswaldoi A, Nyssorhynchus rangeli (Gabaldón, Cova Garcia & Lopez), Nyssorhynchus triannulatus (Neiva & Pinto), among other species listed in Supplementary data ^{(470.6KB, pdf)} (Table I).

Drivers of Anophelinae distribution and malaria transmission

The geographic distribution of Anophelinae species is influenced by environmental factors such as precipitation, temperature, altitude, availability of larval habitats, relief and hydrology. ³¹ However, changes in the natural environment and land use can create micro and macroenvironmental conditions adequate for anopheline species proliferation and dispersion. Ny. darlingi is found in areas of extensive forest cover and along river networks from northern to southern Brazil, reaching its known south limit at the Foz do Iguaçu, Paraná State. In well preserved forest in Yanomami lands, in the Brazilian Amazon, the species is found in lakes associated with river floodplains, and old river paths with U-shaped form (oxbow lakes) that were formed by river isolation because of sedimentation or erosion. ^{32 , 33} In landscapes impacted by anthropic modifications in the natural environment, Ny. darlingi aquatic stages are found in lagoons, streams, streams combined with lagoons, streams combined with dams, and fishponds in the Brazilian Amazon. ^{34 , 35 , 36 , 37}

Kerteszia cruzii, Ke. bellatrix, and Ke. homunculus are sylvatic mosquitoes and vectors of malaria parasites in the Atlantic Forest biome in southeastern Brazil. These species occur in areas where bromeliad plants are abundant because they depend on these plant phytotelmas for egg laying and development of aquatic stages. ³⁸ Particularly, Ke. cruzii is a primary vector in areas where the anthropogenic changes are altering mosquito dynamics and malaria risk, especially in forested areas in São Paulo, Rio de Janeiro, Espírito Santo, Santa Catarina, and northeastern Rio Grande do Sul. ^{39 , 40 , 41} The first documented association of Ke. cruzii as a natural vector of simian Plasmodium simium Marchiafava & Celli and Plasmodium brasilianum (Gonder & Von Berenberg-Gossler) in the Atlantic Forest was reported by Deane et al. ⁴² However, both Plasmodium spp. can infect and cause infection and malaria in humans when they encroach on forest environments. ⁴³ The first report of P. simium malaria in humans was registered by Deane et al. ⁴⁴ in the forest reservation of Horto Florestal in Serra da Cantareira, north São Paulo municipality, São Paulo State. During two-year collections in the canopy and on the ground level, Deane and colleagues collected 30 mosquito species, among them eight anopheline species, including Ke. cruzii that was the most abundant on both level ground and in the canopy, and thus was considered a potential vector. Further studies in other states across the Atlantic Forest, Cerrado, and Amazon were crucial to bring to light new knowledge about the sylvatic reservoirs of the simian Plasmodium and mosquito vectors. ^{45 - 52} Under current and future climate scenarios, both P. vivax and P. falciparum may experience shorter extrinsic incubation periods, while their vector, Ke. cruzii, may increase survival rates because of rising temperatures. This could lead to the expansion of high-risk malaria areas, particularly in the southern Atlantic Forest. ⁵³

In northeast Brazil, Ny. aquasalis plays a role in transmitting malaria parasites in coastal communities, as well as in other coastal areas in South and Central America and Caribbean Island. ⁵⁴ Females lay their eggs in a mixture of fresh and saltwater typically found in coastal lagoons, estuaries, and mangroves. In the mangrove ecosystem, Ny. aquasalis larval and pupal stages can be found in stagnant or slow-moving waters in mangrove swamps, in full sun or partially shaded with mangrove vegetation, grasses and algae. ⁵⁵ Tidal pools with varying salinity levels are also used as larval and pupal stage habitats. Ny. aquasalis may also breed in human-made habitats such as fishponds, salt marshes, and drainage canals near coastal areas. ⁵⁵ The adaptability of the species to a range of salinity levels allows it to occupy a unique reproductive niche not occupied by other anopheline species in coastal regions of northeastern Brazil and other parts of South America. ^{38 , 55}

Anthropogenic changes in land use, particularly deforestation and agricultural expansion, have a profound impact on the distribution and dynamics of mosquito communities. ⁵⁶ Deforestation, habitat fragmentation, and the reduction of forest cover can drastically affect mosquito biodiversity, leading to the decline of some species while others, resilient to altered environments, become dominant. ⁵⁷ For example, in the northeastern Brazilian Amazon, Ny. darlingi, traditionally the primary malaria vector, was replaced by Ny. marajoara because of changes in land use, with Ny. marajoara emerging as the dominant vector species. ⁵⁸ Ny. marajoara is classified in the Albitarsis Complex that encompasses five valid species and five putative species to be formally named. Despite the difficulties in the identification of the species based solely on external characteristics, molecular analyses have confirmed the identification of distinct species within this complex. ⁵⁹ Members of this group are typically associated with aquatic habitats where their larvae develop, such as irrigated rice fields, dam, fishponds, lakes, ponds, streams in full sun, and partially shaded and shaded water. ^{36 , 60} In rural settlements in the Brazilian Amazon, Ny. darlingi benefits from deforestation, becoming more prevalent in areas with higher forest cover and lower border density. ⁶¹ These findings suggest that deforestation and forest fragmentation create environmental conditions that favor shifts in mosquito community structure, with significant changes in species dominance and distribution that may increase human exposure to vectors of malaria parasites, heightening the risk of vector-borne disease worldwide. ⁶²

Nyssorhynchus darlingi is, to some extent, resilient to the ecological conditions in human-modified environments. ⁶¹ As urbanization expands and natural habitats are modified, some anopheline mosquitoes are adjusting to new ecological settings by exploiting new habitats, such as human-made water bodies, drains, lagoons, lakes, and ponds in fish farms. ^{34 , 36 , 37} Ny. darlingi becomes increasingly abundant and dominant in peridomestic environments of human settlements because of anthropogenic factors such as deforestation, alterations in water ecosystems, the creation of standing water bodies, and other human-made structures. ⁶¹ The reduction of forest cover and the rise in edge density associated with deforestation result in decreased mosquito biodiversity, creating favorable conditions for the proliferation of Ny. darling. ⁶¹ One reason for the increased abundance of Ny. darlingi is the ecological advantage of its adaptation of the use of small dams in natural water bodies in forest fringes as larval habitats. ⁶³ The link between deforestation and Ny. darlingi was also found in areas with varying landscape composition in the Peruvian Amazon forest. Larvae of Ny. darlingi were found in water systems in areas with an average of 24.1% forest cover, contrasting with areas without the species that had 41.0% forest cover. Seasonality, presence of algae, water body size, presence of human populations, and the amount of forest and secondary growth were found to be significant for the presence of larval habitats and Ny. darlingi presence. ⁶⁴ In another study conducted in the Brazilian Amazon, spatial clustering of Ny. darlingi larvae was observed in areas where obstructions to river flow created slow-moving or stagnant pools of water. These conditions, combined with reduced sunlight exposure, favor Ny. darlingi larval habitats. ^{35 , 65}

Understanding the geographical distribution of Anophelinae species that are well-known and those that are potential vectors of malaria is crucial for the ongoing efforts toward malaria elimination. ^{66 , 67} Each species has a unique ecological niche, behavior, blood feeding behavior, and vectorial capacity, meaning that the risk of a susceptible person acquiring malaria varies significantly across different landscapes. ⁶⁸ Mapping vector species distribution may help identify areas of higher transmission risk, thus enabling targeted interventions, such as environmental management. ⁶⁹ Also, geospatial distribution modeling together with parasite serology can be employed to help to identify foci of residual malaria transmission. ⁷⁰

At the same time as malaria elimination advances, understanding the distribution of malaria vectors becomes critical for sustaining the success of interventions. ⁷¹ In areas where malaria has been largely eliminated but transmission remains uneven, the disease epidemiology becomes more complex. ^{71 , 72} Therefore, it is essential to establish an active vector surveillance program capable of detecting areas with residual transmission pockets, where malaria vector populations persist at low density. ⁷³ In addition, knowledge of Anophelinae distribution potential allows for early detection of shifts in vector populations because of changes in the environments, land use, urbanization, and climate. ⁷⁴ In regions where malaria transmission has been reduced or eliminated, monitoring the presence and expansion of mosquito vectors is critical to sustaining elimination achievements and preventing the resurgence of malaria. ⁷³ In addition, to achieve the long-term goal of malaria eradication it will be necessary to have greater funding, innovative solutions, and global cooperation. ⁷⁵

Climate change can drive shifts in the geographic distribution of mosquito vectors of Plasmodium and alter malaria-endemic areas. These changes will depend on the environment suitability for malaria transmission, thermal tolerance limits of both mosquito vectors and Plasmodium parasites, the availability of suitable habitats for aquatic life stages, and hydrological processes. ^{31 , 76} Additionally, climate changes can cause alterations in blood-feeding behavior, availability of vertebrate hosts, mosquito life cycle duration, female mosquito longevity, and the duration of Plasmodium extrinsic incubation period can influence malaria dynamics and seasonality. ^{77 , 78} Variation in temperature affects the development of Ny. darlingi in Brazil. Comparing three populations from Brazil, Amazon, Cerrado, and Atlantic Forest, it was found that higher temperatures accelerated larval development, but also shortened adult lifespan and overall longevity, while reducing body size at 28ºC compared to 20ºC. However, Ny. darlingi population from the Atlantic Forest exhibited faster development at warmer temperatures, while maintained a larger body size compared to other populations studied and showed no reduction in longevity. ⁷⁹ These factors together suggest that the vectorial capacity of Ny. darlingi from the Atlantic Forest population may increase under warmer conditions. Moreover, higher temperatures can expand the geographical range of Anophelinae mosquitoes into previously cooler areas, including highland regions, raising malaria risk in areas that were previously non-endemic. ^{80 , 81 , 82} In summary, climate change can exacerbate malaria transmission in some areas, while in others, it can reduce the disease.

Ecological distribution of Anophelinae in Brazil

Since Professor Neiva’s inspiring article in 1909, numerous anopheline species have been described and named. Over time, changes in nomenclature have been proposed, reducing the number of Anophelinae genera from 20 to three, and increasing to seven globally in 2017. ²¹ In Brazil, Neiva recorded eight genera in 1909, which later decreased to two and then increased to six by 2017. To align with the nomenclature proposed by Foster et al., ²¹ corrections to species names regarding gender and authority citations are necessary, as outlined in Tables I and II.

TABLE I. Valid species of the genus Anopheles subgenus Anopheles, and the genera Kerteszia, Lophopodomyia, and Stethomyia found in South America, grouped by subgenera and series, with updates to authority citations and corrections to gender classification. Species marked with an asterisk have been recorded in Brazil.

Genus/Series	Species, author, date
Anopheles Meigen, 1818 Series Arribalzagia	anchietai Corrêa & Ramalho, 1968*
	annulipalpis Lynch Arribálzaga, 1878
	apicimacula Dyar & Knab, 1906*
	bustamantei Galvão, 1955*
	calderoni Wilkerson, 1991
	costai da Fonseca & da Silva Ramos, 1940*
	evandroi da Costa Lima, 1937*
	fluminensis Root, 1927*
	forattinii Wilkerson & Sallum, 1999*
	guarao Anduze & Capdevielle, 1949*
	maculipes (Theobald, 1903)*
	malefactor Dyar & Knab, 1907
	mattogrossensis Lutz & Neiva, 1911*
	medialis Harbach, 2018 (new name for Anopheles intermedius Chagas, 1908)*
	mediopunctatus (Lutz, 1903)*
	minor da Costa Lima, 1929*
	neomaculipalpus Curry, 1931*
	peryassui Dyar & Knab, 1908*
	pseudomaculipes (Chagas in Peryassú, 1908)*
	punctimacula Dyar & Knab, 1906*
	rachoui Galvão, 1952*
	shannoni Davis, 1931*
	vestitipennis Dyar & Knab, 1906
Series Anopheles	eiseni eiseni Coquillett, 1902
	eiseni geometricus Corrêa, 1944*
	pseudopunctipennis levicastilloi Levi Castillo, 1944
	pseudopunctipennis neghmei Mann, 1950
	pseudopunctipennis noei Mann, 1950
	pseudopunctipennis patersoni Alvarado & Heredia, 1947
	pseudopunctipennis pseudopunctipennis Theobald, 1901
	pseudopunctipennis rivadeneirai Levi Castillo, 1945
	tibiamaculatus (Neiva, 1906)*
Kerteszia Theobald, 1905	auyantepuiensis (Harbach & Navarro, 1996)
	bambusicola (Komp, 1937)*
	bellatrix (Dyar & Knab, 1906)*
	boliviensis Theobald, 1905
	cruzii (Dyar & Knab, 1908)*
	gonzalezrinconesi (Cova-García, Pulido F. & Escalante de Ugueto, 1977)
	homunculus (Komp, 1937)*
	laneana (Corrêa & Cerqueira, 1944)*
	lepidota (Zavortink, 1973)*
	neivai (Howard, Dyar & Knab, 1913)*
	pholidota (Zavortink, 1973)
	rollai (Cova-García, Pulido F. & Escalante de Ugueto, 1977)
Lophopodomyia Antunes, 1937	gilesi (Neiva in Peryassú, 1908)*
	gomezdelatorrei (Leví-Castillo, 1955)
	oiketorakras (Osorno-Mesa, 1947)
	pseudotibiamaculata (Galvão & Barretto, 1941)*
	squamifemur (Antunes, 1937)*
	vargasi Gabaldón, (Cova García & López, 1941)
Stethomyia Theobald, 1902	acanthotoryna (Komp, 1937)
	canorii (Floch & Abonnenc, 1945)
	kompi (Edwards, 1930)*
	nimbus Theobald, 1902*
	thomasi (Shannon, 1933)*

TABLE II. Valid species of the genus Nyssorhynchus (subfamily Anophelinae) found in South America, grouped by series and informal groups, with updates to authority citations and corrections to gender classification. Species marked with an asterisk have been recorded in Brazil.

Series	Group	Subgroup	Complex	Species, author, date
Albimanus				albimanus (Wiedemann, 1820)
Oswaldoi	Oswaldoi	Oswaldoi		aquasalis (Curry, 1932)*
				evansae (Brèthes, 1926)*
				galvaoi (Causey, Deane & Deane, 1943)*
				ininii (Senevet & Abonnenc, 1938)*
				oswaldoi (Peryassú, 1922) (s.s.)*
				rangeli (Gabaldon, Cova Garcia & López, 1940)*
				sanctielii (Senevet & Abonnenc, 1938)
				trinkae (Faran, 1979)
			Konderi	konderi (Galvão & Damasceno, 1942) (s.s.)*
				tadei (Saraiva & Scarpassa, 2021)*
			Nuneztovari	dunhami (Causey, 1945)*
				goeldii (Rozeboom & Gabaldon, 1941)*
				jamariensis Sant’Ana & Sallum, 2024*
				nuneztovari (Gabaldon, 1940) (s.s.)*
		Strodei	Arthuri	albertoi (Unti, 1941)*
				arthuri (Unti, 1941) (s.s.)*
				ibiapabaensis Sant’Ana & Sallum, 2024*
				rondoni (Neiva & Pinto, 1922)*
				rondoniensis Sant’Ana & Sallum, 2024*
				striatus (Sant’Ana & Sallum, 2016)*
				strodei (Root, 1926)*
				untii Sant’Ana & Sallum, 2024*
			Benarrochi	benarrochi (Gabaldon, Cova Garcia & López, 1941) (s.s.)
		Triannulatus		halophylus (Silva-do-Nascimento & Lourenço-de-Oliveira, 2002)*
				triannulatus (Neiva & Pinto, 1922) (s.s.)*
Albitarsis	Albitarsis		Albitarsis	albitarsis (Lynch Arribálzaga, 1878) (s.s.)*
				deaneorum (Rosa-Freitas, 1989)*
				janconnae (Wilkerson & Sallum, 2009)*
				marajoara (Galvão & Damasceno, 1942)*
				oryzalimnetes (Wilkerson & Motoki, 2009)*
	Braziliensis			braziliensis (Chagas, 1907)*
Argyritarsis	Argyritarsis			argyritarsis (Robineau-Desvoidy, 1827)*
				sawyeri (Causey, Deane, Deane & Sampaio, 1943)*
	Darlingi			darlingi (Root, 1926)*
	Lanei			lanei (Galvão & Franco do Amaral, 1938)*
	Pictipennis			atacamensis (González & Sallum, 2010)
				pictipennis (Philippi, 1865)
Myzorhynchella				antunesi (Galvão & Franco do Amaral, 1940)*
				guarani (Shannon, 1928)*
				lutzii (Cruz, 1901) (s.s.)*
				nigritarsis (Chagas, 1907)*
				parvus (Chagas, 1907)*
				pristinus (Nagaki & Sallum, 2010)*

Currently, approximately 103 species of Anophelinae have been documented in Brazil, including 73 formally described species and 30 putative species identified through COI barcoding and supported by morphological evidence [Supplementary data ^{(470.6KB, pdf)} (Table I)]. These putative species, while recognized through genetic and morphological divergence, remain to be formally described [Supplementary data ^{(470.6KB, pdf)} (Table I)]. The species recorded in Brazil belong to the genera Anopheles, Chagasia, Kerteszia, Lophopodomyia, Nyssorhynchus, and Stethomyia. The genus Kerteszia includes seven formally named species and two additional putative species identified through COI barcoding of the mitochondrial genome. ⁸³ The genus Nyssorhynchus comprises 36 formally named species and at least 19 yet-to-be-described species, such as An. albitarsis F-J, ⁸⁴ An. oswaldoi A, B, An. benarrochi B, An. konderi A, C, D, An. triannulatus C, ⁸⁵ among others [Supplementary data ^{(470.6KB, pdf)} (Table I)]. Additionally, species of the Myzorhynchella subgenus require further study, as field collections across various Brazilian localities have revealed evidence supporting the existence of several new species, based on morphological characteristics of males, females, fourth-instar larvae, and COI barcode sequences. ^{86 , 87} The genus Anopheles includes 22 species, and 10 potential new species identified via COI barcoding, ^{88 , 89} some of which have proved to be experimentally competent to transmit malaria parasites ^{90 , 91} and/or found naturally infected. The least studied genera, Stethomyia, Lophopodomyia, and Chagasia, each have three species recorded in Brazil. These exclusively Neotropical genera are likely under-sampled and poorly understood due to challenges in collecting immature stages and adults, as well as their limited or unknown public health relevance as vectors of human Plasmodium.

The geographic distribution of Ke. cruzii is primarily restricted to the Atlantic Forest biome (Fig. 2A). However, the coastline extending from northern Santa Catarina to northern São Paulo states represents the areas with the highest climatic suitability for this species [Fig. 2C, Supplementary data ^{(470.6KB, pdf)} (Table II)]. Ny. darlingi has a wide distribution across Brazil and is found in all major biomes except the Pampas (Fig. 2B). Nevertheless, certain regions of Brazil exhibit low habitat suitability for Ny. darlingi [Fig. 2D, Supplementary data ^{(470.6KB, pdf)} (Table II)]. These areas include southern Brazil, particularly in Rio Grande do Sul, and transitional zones between the Pampas and Atlantic Forest biomes in Santa Catarina and southern Paraná states. Additionally, low suitability is noted in the Mantiqueira Mountain Range, located at the intersection of São Paulo, Minas Gerais, and Rio de Janeiro states. Further, Ny. darlingi shows low probability of occurrence in the Zona da Mata region, the tablelands of Bahia State, and the plateau regions of the western Amazon. On the other hand, areas with high susceptibility for Ny. darlingi occurrence include the Acaraú River valley near the Paulo Sarasate dam, at the foothills of the Ibiapaba Mountain Range in Ceará State. Other regions of high suitability include the Sertanejos Residual Plateaus of Seridó, Serra de Santana, and Vale do Açu, situated on the border between Paraíba and Rio Grande do Norte states (Fig. 2D).

Fig. 2: the probability of presence (ranging from 0 to 1) for Nyssorhynchus darlingi and Kerteszia cruzii was modeled based on the geographical areas with optimal climate conditions, as identified using climate variables from the WorldClim Global Climate Data. The analysis was conducted at a spatial resolution of 30 arc seconds (approximately 1 km), using a maximum entropy distribution model.

Mapping potential distribution of Anophelinae vector species and diversity

Currently, approximately 103 species of Anophelinae have been documented in Brazil, including 30 putative species that have yet to receive formal naming [Supplementary data ^{(470.6KB, pdf)} (Table I)]. Among these, 30 species and one species complex have been identified as dominant, local, or potential vectors. These species have either been found naturally infected, demonstrated vector competence in laboratory conditions, are known vectors in bordering countries, or have been detected infected in peridomestic environments, and forested areas with endemic transmission [Figs 2,3, Supplementary data ^{(470.6KB, pdf)} (Table I)].

Results of climate variables selection and principal components analyses (PCA) are in the Supplementary data ^{(470.6KB, pdf)} (Tables II-III). The MaxEnt modeling analysis incorporated the four most representative eigenvectors of climate variables relevant to the vector species under study [Supplementary data ^{(470.6KB, pdf)} (Table III)]. In addition, the MaxEnt analyses indicate that the key climatic factors influencing the distribution of Ny. darlingi and Ke. cruzii are the minimum temperature of the coldest month, temperature annual range, and precipitation-related variables, with some variation in the specific importance of these factors between species. While precipitation seasonality and the minimum temperature of the coldest month are particularly important for Ny. darlingi, isothermality and the precipitation of the driest month are the most influential factors for Ke. cruzii. MaxEnt modelling produced varying predictive accuracies, with an AUC score of 0.70 for Ny. darlingi (moderate accuracy) and 0.98 for Ke. cruzii (high accuracy) [Supplementary data ^{(470.6KB, pdf)} (Table IV)].

The potential geographic distribution of anopheline species with some role in malaria transmission dynamics, influenced by climatic factors, is highly heterogeneous. Most species are predominantly confined to the Amazon region, characterized by its warm and humid climate (Fig. 3). However, certain species have broader distributions across Brazilian biomes. These more adaptable, generalist species include An. medialis, An. peryassui, Ke. homunculus, Ny. darlingi, Ny. rangeli, Ny. rondoniensis, and Ny. triannulatus. Species such as Ke. bellatrix and Ke. cruzii thrives in the Atlantic coastal climate, especially in the northeast, where conditions are humid and tropical. The Atlantic Forest includes altitude tropical climates in southeastern Brazil, characterized by consistently high temperatures and humidity, with well-distributed rainfall throughout the year. Distinct from other vectors of Plasmodium, Ny. strodei is suited to both tropical and temperate climates. It occupies areas with mild average temperatures and dry winters, with its geographic distribution extending to the southernmost parts of Brazil.

Fig. 3: the probability of presence (ranging from 0 to 1) for 28 Anophelinae species or putative species with some role in malaria transmission dynamics was modeled based on the geographical areas with optimal climate conditions, as identified using climate variables from the WorldClim Global Climate Data. The analysis was conducted at a spatial resolution of 30 arc seconds (approximately 1 km), using a maximum entropy distribution model.

The Shannon-Wiener diversity map shows that regions with the highest diversity of anopheline vectors are largely concentrated in the Amazon (Fig. 4). The Brazilian Atlantic Coast, particularly rainforest areas, also exhibits high vector diversity, notably in the northeastern Zona da Mata. Other areas of significant diversity include Monte Pascoal National Park in Bahia and northern Espírito Santo. In the Southeast, regions with considerable diversity include Fluminense, Costa do Sol, and Guanabara Bay. Additionally, the border area between São Paulo and Paraná, which encompasses the São Paulo State parks of Lagamar de Cananéia and Ilha do Cardoso as well as Superagüi National Park in Paraná, hosts high anopheline diversity. The Mato Grosso and Rondônia ecotones between the Pantanal and Amazon regions, in western Mato Grosso and Rondônia, are areas of notable diversity where Plasmodium-infected anopheline species are commonly found.

Fig. 4: diversity of 28 Anophelinae species or putative species with potential role in malaria transmission, using Shannon-Wiener diversity index in Brazil.

Mapping geographical distribution of Anophelinae species in Brazil

The geographical distribution of species within the subfamily Anophelinae (Figs- 5 - 9) highlights their varying adaptability to ecological regions characterized by distinct biotic and abiotic factors, such as climate, hydrological features, and vegetation. While some species are generalists, thriving across multiple biomes, others display more specialized preferences [Supplementary data ^{(470.6KB, pdf)} (Table I)]. For instance, Ny. aquasalis is closely associated with saline waters along Atlantic coast, and Ny. rangeli is primarily found in the Amazon region (Fig. 5A). Species within the Konderi Complex are distributed throughout the Amazon region, with Ny. konderi extending southwest into western Paraná State. Ny. tadei reaches as far as Mato Grosso do Sul State, extending to southern of the Pantanal region, Ny. konderi A is restricted to Amazon forest, Ny. konderi C is found in a specific area in the southern Pantanal (Fig. 5B, C). Ny. oswaldoi A and Ny. oswaldoi B are both found in the Amazon rainforest and considered potential vectors of Plasmodium spp. [Fig. 5D, F, Supplementary data ^{(470.6KB, pdf)} (Table I)]. Additionally, specimens previously identified as Ny. oswaldoi or Ny. oswaldoi s.l. across the Amazon region may be Ny. oswaldoi A, likely due to misidentification (Fig. 5D-F). Among species distributed across Brazil’s major biomes, 26 have been collected exclusively within the Amazon rainforest, while 19 species have been reported solely in the Atlantic tropical rainforest. Within the Cerrado biome, four valid species - An. tibiamaculatus, Lp. gilesi, Ny. albertoi, and Ny. untii - have been documented exclusively, along with two potential new species, Ny. parvus Type 1 and Ny. parvus Type 2 [Supplementary data ^{(470.6KB, pdf)} (Table I)]. Two species were recorded in forested areas within the Caatinga biome: Ch. rozeboomi, found in Londa, near Crato, Ceará, at an elevation of 500 meters. This species inhabits a wooded area in a transition zone among the Caatinga, Cerrado, and Atlantic tropical rainforest in the Chapada do Araripe. ⁹² The second species, Ny. ibiapabaensis, was recently described from specimens collected at an altitude of 900 meters in the municipality of São Benedito. ⁹³ Ny. janconnae, a primary vector of human Plasmodium, ⁹⁴ appears to be a habitat specialist, as it has been found exclusively within the Savanna (known locally as Lavrado) zone near Boa Vista in Roraima State. Among the habitat-specialist species, Ch. fajardi/rozeboomi, Ke. laneana, Ny. antunesi Type 1, Ny. antunesi, Ny. lanei, and Ny. pristinus have been recorded only in the Serra da Mantiqueira region in São Paulo State. Meanwhile, Ny. antunesi Types 2, 3, and 4 have been found in mountainous areas along the southern boundary of the Atlantic tropical rainforest. Ny. antunesi Type 2 has been collected in both the Serra da Mantiqueira and along the southern boundary of the Atlantic rainforest, while Ny. pristinus Type 1 has been identified in a mountainous area that is part of the Atlantic Forest biome, within the mountain range of the Tapiraí region in Vale do Ribeira, São Paulo State. Forty-one species from the genera Anopheles, Chagasia, Nyssorhynchus, and Stethomyia are generalist mosquitoes distributed across various Brazilian biomes. Understanding this distribution requires careful consideration of potential species misidentification, stemming from limited taxonomic and ecological research.

Fig. 6: maps showing the distribution of the major Brazilian biomes, localities of collections of specimens of 18 species of the genus Nyssorhynchus. (A) Nyssorhynchus triannulatus, Nyssorhynchus halophylus; (B) Nyssorhynchus deaneorum, Nyssorhynchus albitarsis G. Ny. albitarsis H; (C) Ny. albitarsis, Ny. albitarsis l.s; (D) Nyssorhynchus oryzalimnetes, Nyssorhynchus marajoara, Nyssorhynchus janconnae; (E) Nyssorhynchus parvus, Nyssorhynchus guarani; (F) Nyssorhynchus antunesi, Nyssorhynchus lutzii, Nyssorhynchus nigritarsis; (G) Nyssorhynchus braziliensis, Nyssorhynchus sawyeri; (H) Nyssorhynchus argyritarsis; (I) Nyssorhynchus lanei.

Fig. 7: maps showing the distribution of the major Brazilian biomes, localities of collections of specimens of 22 species of the genera Nyssorhynchus, Kerteszia, Stethomyia. (A) Nyssorhynchus arthuri, Nyssorhynchus rondoni, Nyssorhynchus ibiapabaensis; (B) Nyssorhynchus strodei, Nyssorhynchus rondoniensis, Nyssorhynchus albertoi; (C) Nyssorhynchus striatus, Nyssorhynchus benarrochi B, Nyssorhynchus untii; (D) Kerteszia homunculus, Kerteszia laneana; (E) Kerteszia belatrix, Kerteszia bambusicola, Kerteszia lepidota; (F) Kerteszia neivai, Ke. neivai A, Ke. neivai B; (G) Stethomyia kompi, Stethomyia kompi/canorii; (H) Stethomyia nimbus; (I) Stethomyia thomasi; Stethomyia nimbus/thomasi.

Fig. 8: maps showing the distribution of the major Brazilian biomes, localities of collections of specimens of 18 species of the genus Anopheles. (A) Anopheles peryassui, An. rachoui; (B) Anopheles eiseni geometricus, Anopheles near costai, Anopheles near costai G2; (C) Anopheles anchietai, Anopheles near costai G4; Anopheles apicimacula; (D) Anopheles puntimacula, Anopheles near punctimacula, Anopheles near costai G3; (E) Anopheles maculipes, Anopheles near costai G1; (F) Anopheles near malefactor; (G) Anopheles near fluminensis; (H) Anopheles near fluminensis G1, Anopheles near fluminensis G2; (I) Anopheles near fluminensis G3.

Fig. 9: maps showing the distribution of the major Brazilian biomes, localities of collections of specimens of 21 species of the genera Anopheles, Chagasia, and Lophopodomyia. (A) Anopheles costai, Anopheles guarao; (B) Anopheles eiseni, Anopheles forattini, Anopheles minor; (C) Anopheles mattogrossensis, Anopheles medialis, Anopheles evandroi; (D) Anopheles medipunctatus, Anopheles tibiamaculatus, Anopheles bustamantei; (E) Anopheles fluminensis, Anopheles shannoni; Anopheles pseudomaculipes; (F) Ch. fajardi, Ch. rozeboomi; (G) Ch. fajardi/rozeboomi, Ch. bonneae; (H) Lp. gilesi, Lp. pseudotibiamaculata, Lp. squamifemur.

Fig. 5: maps showing the distribution of the major Brazilian biomes, localities of collections of specimens of 16 species of the genus Nyssorhynchus. (A) Nyssorhynchus rangeli and Nyssorhynchus aquasalis; (B) Nyssorhynchus konderi, Ny. konderi A, Ny. konderi C; (C) Nyssorhynchus tadei, Nyssorhynchus galvaoi; (D) Nyssorhynchus evansae, Nyssorhynchus oswaldoi B, Ny. oswaldoi l.s.; (E) Ny. oswaldoi; (F) Ny. oswaldoi SPForm, Ny. oswaldoi A; (G) Nyssorhynchus goeldii; (H) Nyssorhynchus nuneztovari, Nyssorhynchus jamariensis; (I) Nyssorhynchus dunhami.

For example, Ny. strodei, initially classified as a single species with a broad distribution, is now recognized as a complex comprising at least seven distinct species. ⁹³ Similarly, species within the Myzorhynchella subgenus of Nyssorhynchus remain under-sampled, largely because locating and accessing their larval habitats in forested regions is challenging. Enhanced sampling efforts in the Atlantic tropical rainforest and Cerrado biomes have revealed ten species previously unknown to science, alongside five recognized species. ⁸⁷ The broad geographical distribution observed in certain species of the genera Nyssorhynchus and Anopheles may be partially attributed to their preference for permanent larval habitats such as river basins, fishponds, ponds, lakes, canals, river margins, floodplains, and swamps - habitats that are relatively accessible for field investigations.

Field sampling biases, which influence perceived species distribution, were initially noted by Neiva ¹ and remain relevant for understanding Anophelinae spatial distribution in Brazil. To address potential biases in anopheline species distribution, several factors should be considered. Sampling effort bias, for example, can skew species occurrence data, as intensive focus on particular habitats can lead to underrepresentation of species from less-sampled environments, thus distorting the overall knowledge of species ranges. Temporal bias affects data collection across seasons and years, impacting the detection of species with seasonal or cyclical population dynamics. Taxonomic bias also plays a role; certain species, such as Ny. darlingi and species within the Albitarsis Complex, receive more research attention, while others remain understudied. Detection bias is similarly significant, as species inhabiting specialized environments, such as phytotelmata (plant-associated water bodies), Amazonian igapó forest, small forest ponds, or debris-covered habitats, are often overlooked. Geographical bias may account for data scarcity in under-sampled biomes such as the Pantanal, Pampas, and Caatinga. Additionally, environmental changes driven by habitat destruction and fragmentation alter mosquito communities, potentially misrepresenting species distributions, as seen in the Cerrado biome.

Vector surveillance for malaria elimination in Brazil

The World Malaria Report highlights a 27% reduction of global malaria incidence and mortality from 2000 to 2015. ⁹⁵ In subsequent years, progress has slowed, with a 2% reduction since 2015. Because of the stagnation of malaria control progress, the world faces a critical moment to prevent further setbacks, thus malaria control requires intensified and multifaceted interventions. ⁷⁵ The control for elimination strategies needs to address socioeconomic and environmental determinants of transmission dynamics, ⁹⁶ in addition to factors associated with malaria protozoans, and mosquito vectors. At a certain level, innovative technologies can provide alternatives for malaria and vector control. For instance, genetically modified mosquitoes that are refractory to Plasmodium infection can help to decrease the density of infective mosquitoes in a transmission setting, ⁹⁷ while geography and geospatial analysis will be important for monitoring malaria and the impacts of control strategies. ⁶⁹

Malaria in Brazil has decreased over the years, but it persists in remote areas, particularly in the Amazon region, where aggressive land occupation, extensive deforestation, and human migration in and out of these areas contribute to ongoing transmission. The most affected and vulnerable areas are mining sites, indigenous lands, and rural regions, where limited access and the sustainability of control measures are significant challenges. Additionally, deforestation and proximity to forested environments increase human exposure to mosquito vectors, as observed in rural settlements in the Amazon. ⁶¹ In 2023, P. falciparum and mixed malaria cases significantly increased in indigenous and mining areas. Delays in diagnosis and treatment exacerbate transmission and case severity. Ensuring accurate treatment data, continuous education for healthcare workers, and targeted health education for at-risk groups is essential to raise awareness and promote timely treatment. ⁹⁸

Monitoring the distribution of mosquito species that are vectors of Plasmodium spp. poses significant challenges for vector surveillance programs of countries where malaria is endemic. ⁷¹ Environmental changes, such as deforestation, urbanization, and agricultural expansion, can alter the distribution of habitats of Anophelinae mosquitoes, creating new habitats or diminishing existing ones. ⁹⁹ These dynamics make it difficult to predict and track mosquito populations over time, particularly in areas experiencing ecological shifts. ⁶⁹ Additionally, many Anophelinae species are difficult to differentiate morphologically, which complicates identification efforts in the field and increases the risk of misidentification in surveillance activities. ⁸⁶

Another challenge is the limited availability of comprehensive, up-to-date distribution data for many Anophelinae species. ²⁴ Although field data from multiple collections across Brazilian territories provide valuable information, there are still gaps in the knowledge of species distributions. These gaps can hinder the development of effective control strategies, as vector surveillance must be tailored to local ecological and epidemiological contexts. Moreover, many areas with high malaria transmission risk, particularly remote or understudied regions, lack adequate surveillance infrastructure, further complicating efforts to monitor species and implement timely interventions. ¹⁰⁰ To address these challenges, it is crucial to integrate multiple approaches of vector surveillance, including molecular techniques for species identification, geographic information systems (GIS) for mapping species distributions, and ecological niche modeling to predict potential vector habitats under different environmental scenarios. ^{101 , 102} Enhancing surveillance systems with these advanced tools will improve the accuracy of mosquito monitoring and support more effective interventions aimed at eliminating malaria transmission in endemic countries and in Brazil, as well.

In conclusion

Despite progress in understanding the distribution of Anophelinae species in Brazil since Neiva’s 1909 study, significant gaps remain, particularly in remote and ecologically complex biomes such as the Amazon, Cerrado, Pantanal, Pampas, and Caatinga. Limited data on the distribution, ecological preferences, and population dynamics of malaria vectors, especially species beyond Ny. darlingi and Ke. cruzii, from the Nyssorhynchus, Kerteszia, and Anopheles genera, hinder the development of targeted control strategies. These knowledge gaps impede efforts to accurately map transmission risk, predict vector behavior in diverse environments, design effective vector control interventions, and implement effective public health responses. For instance, sylvatic species capable of transmitting Plasmodium to human hosts in forested areas beyond human dwellings may continue to act as reservoirs, increasing the risk of outbreaks in both endemic and non-endemic areas. Addressing these knowledge gaps about vector species across all Brazilian biomes is essential for improving vector surveillance and control efforts, ultimately contributing to malaria elimination in Brazil.

Funding Statement

CNPq (grant number 303382/2022-8), NIH (grant R01 AI110112)