Spatio-temporal dynamics of Rift Valley fever virus circulation in sentinel livestock herds (2021-2024) before the initial fatal human outbreaks in Senegal Authors: Déthié Ngom, Elisabeth Thérèse Faye, Aliou Khoulé, Sokhna Mayemouna Diop, Ousseynou Sène, Mignane Ndiaye, Cheikh Talla, Cheikh Loucoubar, Oumar Faye, Mawlouth Diallo, Yamar Ba, Gamou Fall, Diawo Diallo Published: 2026 ABSTRACT BackgroundMultiple detections of Rift Valley fever virus (RVFV) infections in humans, animals and mosquito vectors suggest permanent circulation of RVFV in Senegal. This study investigated the spatial and temporal dynamics of RVFV circulation in central (Sahelian) and southeastern (Sudanian) Senegal from March 2021 to June 2024.MethodsA total of 976 initially seronegative sheep and goats from sentinel herds in 10 districts and 2 areas were monitored for 40 months. Serum samples were collected quarterly and tested by ELISA, while mosquitoes were collected monthly and screened for RVFV RNA by RT-PCR.ResultsActive RVFV circulation was detected in both areas during the four years preceding the 2025 outbreak. Overall, 2989 sera were tested, and 7.2% (95% CI: 5.6–9.0%) of animals seroconverted, while no RVFV RNA was detected from 65,830 mosquitoes. Seroconversion was higher in goats than in sheep (OR = 2.6; 95% CI: 1.5-4.6; p < 0.001) and during the rainy than the dry season (dry: 9.1% vs rainy: 5.6, p = 0.023).ConclusionEnzootic RVFV circulation preceded the 2025 outbreak. Sentinel herd surveillance, integrated with entomological and environmental monitoring, may improve RVFV risk assessment within a One Health framework. Introduction Zoonotic pathogens are major and growing health threat worldwide, due to expansion of hosts and vectors, increasing human exposure, amplified by climate-linked changes in temperature, precipitation patterns, and environmental conditions that favor pathogen transmission [[1], [2], [3], [4], [5], [6], [7]]. Rift Valley fever (RVF) is a major mosquito-borne zoonosis of public and veterinary health importance, responsible for recurrent outbreaks affecting both humans and animals [8]. RVF virus (RVFV), a Phlebovirus (order Hareavirales) [9], is transmitted primarily by mosquitoes but also through direct contact with infected animal tissues or products [[10], [11], [12], [13]]. In animals, RVFV causes abortions and high mortality in young sheep and goats [14,15]. Human infections range from mild febrile illness to severe disease and death [16,17], with veterinarians, livestock farmers, abattoir workers and herders at higher risk of infection [18]. In Senegal, RVFV has circulated since the first documented outbreak in 1987, with repeated detections in humans, livestock, and mosquitoes during both epidemic and inter-epidemic periods [11,12,[19], [20], [21], [22]]. The first major outbreak occurred in the Senegal River Valley following construction of the Diama Dam, affecting both Mauritania and Senegal and resulting in widespread livestock epizootics and human cases [23]. Subsequent outbreaks were reported in 1993, 1998, 2003, 2012–2013, and 2014 [11,20,21,[24], [25], [26]]. In 2013, RVFV transmission expanded beyond its traditional northern foci to central, urban, and peri-urban areas, with human cases confirmed in Thiès and Dakar, highlighting an unprecedented geographical spread [26]. Following 2014, RVFV circulation persisted, although only sporadic human cases or small outbreaks were reported. During the 2020 outbreak, six of the nine confirmed human cases occurred in the northern regions of Saint-Louis and Matam [27]. A recent serosurvey in Matam further demonstrated high RVFV IgG seroprevalence in sheep (66.7%) and humans (20.8%), confirming sustained virus circulation in this border region [28]. These observations indicate that RVFV remains endemic in Senegal, with marked spatial and temporal heterogeneity, particularly in the Sahelian and Sudano-Guinean ecological zones [[19], [20], [21],[24], [25], [26]]. In September 2025, Senegal experienced its first fatal RVF outbreak. By 24 December, 554 laboratory-confirmed human cases and 31 deaths had been reported [29]. However, despite evidence of enzootic circulation, longitudinal data documenting silent viral transmission prior to major outbreaks are lacking. Small ruminants are good sentinels for monitoring zoonotic risk due to their high susceptibility to infection, their close proximity to humans, and their greater exposure to vectors [[30], [31], [32]]. This study assessed longitudinal exposure of sentinel small ruminant herds in southeastern and central Senegal between 2021 and 2024, using integrated entomological and serological surveillance. The findings provide new insights into the spatio-temporal dynamics of RVFV circulation, and the importance of integrated One Health surveillance for early detection and prevention of future outbreaks. Materials and methods Study sites The study was carried out from March 2021 to June 2024 in two different bioclimatic zones, central (Linguère) and southeastern (Kédougou) Senegal, West Africa (Fig. 1). In these areas, livestock production is one of the major economic sectors of the population, providing products such as food and income. Kédougou, bordered to the northwest by the abundant natural reserve forest of Niokolo-Kobo National Park, to the east by Mali and to the south by Guinea, has a Sudanian climate, with heavy precipitations from May to October (averaging 1100 to 1450 mm per year) [33,34]. Linguère is located in the central part of the Sahelian Ferlo region of Senegal, characterized by shrub savannah, a hot dry climate and a short rainy season that generally occurs between July and September. Annual rainfall is approximately 300 mm [35]. Temporary ponds of various sizes are flooded with the first rains and the large ponds remain flooded for a long period. The study included the districts of Dodji, Gassane, Thiel, Thiargny and Barkedji in the Linguère area; and the districts of Bandafassi, Dindefelo, Dimboli, Fongolimbi and Tomboronkoto in the Kédougou area. These districts were purposively selected due to huge presence of different RVFV hosts (mosquitoes, humans and animals).Fig. 1Study area with different districts. Study area with different districts. Animal sentinel herds Continuous seasonal cross-sectional serosurveys were conducted using sentinel herd comprising 15 sheep and 15 goats, installed in each of the 10 aforementioned districts in Linguère and Kédougou. Animal herds were composed of naive young females at least 5 months old, ideally between 6 and 12 months. Animal age was estimated by dental examination [36] and grouped into two categories: 05–18 months for young and > 18 months for adults. Animals were permanently identified, monitored clinically, and sampled every three months for RVFV-specific IgG and IgM antibodies. Seroconversion was defined as a change from seronegative to IgG and/or IgM positive. Seroconverted animals were replaced with seronegative young animals (2–4 months old) to maintain a susceptible cohort. Only the first confirmed seroconversion for each animal was included in the analysis. Blood sampling and processing On each ruminant, a venous blood sample (approximately 4 mL) was collected in plain Vacutainer tubes (venipuncture) by local veterinarians, then vitamins were given for the treatment and antibiotics for certain animals affected by wounds or infections of blood parasites. After centrifugation at 5000 × g for 10 min, the serums were aliquoted in 2 mL sterile Nunc tubes (Nunc, Denmark). Sera were transported to the laboratory in liquid nitrogen and stored at −80°C until further processing. Serological assay for the detection of antibodies against RVFV Serum samples were tested for both IgM or IgG antibodies detection using ID Screen® Rift Valley fever Competition Multi-species (ID Vet. Montpellier, France) commercial Enzyme-Linked Immunosorbent Assay kit. Competitive ELISA was performed according to the manufacturer's instructions. This diagnostic kit is designed to detect antibodies directed against the RVFV nucleoprotein (NP) in serum or plasma. Briefly, in each well, 50 μL of dilution buffer was mixed with 50 μL of either samples or controls (positive and negative) in duplicate, and incubated at 37°C for 1 h. After three washes, 100 μL of the conjugate solution was added, and the plate was incubated at room temperature (21°C) for 30 min. Following another washing step, 100 μL of substrate solution was added. The plate was incubated for 15 min at room temperature in the dark, and the reaction was halted by adding 100 μL of stop solution. Optical densities (OD) were read at 450 nm using an ELISA microplate reader. The assay was validated if the mean negative control OD was > 0.7 and the ratio of the OD of the positive control to the negative control was < 0.3. Results were expressed as a percentage of the sample to negative control ratio (S/N%). Samples with a S/N% less or equal to 40% were considered positive for RVFV antibodies, between 40% and 50% as doubtful, and > 50% as negative. The same ELISA kit and testing procedure were used consistently throughout the entire study. Seroconversion was defined as a transition from RVFV antibody-negative status at baseline to antibody-positive status in a subsequent sampling round. Animals with equivocal results were retested according to the manufacturer's instructions, and only confirmed positive results were classified as seroconversions. Mosquito sample collection and processing Mosquitoes were collected monthly from districts where animal sentinel herds were placed, using CO2-CDC light, dry ice-baited BG-Sentinel-2, water-baited BG-Sentinel-2 and foldable mosquito net traps baited with CO2 (200x150 × 150 cm) [[37], [38], [39]] placed in animal shelters, ponds, bare land and human dwellings. Mosquitoes were killed by freezing at −20°C, sorted on a chill table using an available identification key [[40], [41], [42]] and pooled (≤100 mosquitoes) by species, sex, date and capture method. Mosquitoes were transported to the laboratory in liquid nitrogen and stored at −80°C until further processing. RNA extraction and qRT-PCR testing A super pooling strategy was adopted to reduce the expense of large number of unfed mosquito pools to be tested. Mosquito pools were homogenized in 3 ml of L-15 medium (Gibco BRL, GrandIsland, NY, USA) supplemented with 20% of fetal bovine serum and clarified by centrifugation at 1500g, 4°C for 10 min. The supernatant was then filtered using a 1 ml syringe (Artsana, Como, Italy) and sterilized with 0.20 μm filters (Sartorius, Göttingen, Germany) [22]. Then, super pools of mosquito samples were prepared using the supernatants from the different pools. The pooling from 2 to 10 individual mosquito pool was done to obtain super-pools based on the mosquito species, origin and date of collection. Briefly, we mixed 100 μl of supernatant from each pool to make the super pool. RNA was extracted from the super pools using the QIAamp viral RNA mini kit according to the manufacturer's instructions (Qiagen, Hilden, Germany). RT-PCRs were performed for the RVFV [43]. In case of positive sample, the different pools that constitute the positive super pools were subsequently submitted to RNA extraction and RT-PCRs were performed to identify the primary positive mosquito pools. This super-pooling protocol was successful to detect 42 isolates of 7 viruses in a previous study [39]. Data analyses Data were entered, cleaned and validated in MS Office Excel® 2016, then exported into R software v 4.2.3 [44] for analyses. The chi-square test was used to assess the association between sentinel animal seroconversion and categorical variables. Additionally, the 95% confidence intervals were calculated for proportions using a binomial distribution. Logistic regression analyses were performed using seropositivity as the dependent variables and potential risk factors (species, age, sex, districts and areas) as explanatory variables. The strength of the association between categorical variables were presented as odds ratios (ORs) and 95% confidence interval (CI), p-value < 0.05 was considered significant. Results Spatial and seasonal abundance of mosquito vectors A total of 65,830 mosquitoes were collected, comprising more than 68 species across 12 genera. The most abundant species were Culex quinquefasciatus (22.5%), Cx. cinereus (18.8%), Cx. nebulosus (12.3%), Cx. poicilipes (10.8%) and Cx. neavei (9.9%). Major RVFV vectors (Cx. poicilipes, Cx. tritaeniorhynchus, Aedes vexans, Ae. dalzieli, Ae. ochraeus, Mansonia africana and Ma. uniformis) represented 20.1% of all collected mosquitoes. Among the RVFV vectors, the most abundant species were Cx. poicilipes (53.9%), Ae. vexans (16.5%), Cx. tritaeniorhynchus (13.5), Ma. uniformis (6.6%) and Ae. dalzieli (5.6%). Vectors abundance showed spatial and seasonal variations. Cx. poicilipes, Ae. vexans, Cx. tritaeniorhynchus and Ma. uniformis predominated in the central area, representing 60.7%, 16.2%, 14.7% and 3.9% of RVFV vectors, respectively (Fig. 2). Conversely, the southeastern area was dominated by Ae. dalzieli (43.1%), Ma. uniformis (27.4%), and Ae. vexans (19.5%).Fig. 2Abundance of mosquitoes, central and southeastern Senegal, 2021-2024Cx. = Culex, Ae. = Aedes, Ma. = Mansonia, An. = Anopheles, RVFV = Rift Valley fever virus. Abundance of mosquitoes, central and southeastern Senegal, 2021-2024 Cx. = Culex, Ae. = Aedes, Ma. = Mansonia, An. = Anopheles, RVFV = Rift Valley fever virus. Seasonal fluctuation of RVFV vectors (Table 1) showed higher proportion of these vectors during the rainy season (23.8%) compared with the dry season (8.93%; p-value = 0.0002). The population of Cx. poicilipes, was over 11 times higher in rainy season than in the dry season. RVFV was not isolated from 7326 pools of 65,830 unfed mosquitoes tested.Table 1Seasonal fluctuation of mosquitoes, central and southeastern Senegal, 2021-2024.Table 1SEASONMosquito categoriesMosquito speciesPercent (n)Dry seasonRVFV vectors8.9% (1640/18368)Ae. vexans3.1 (570)Cx. poicilipes3.0 (555)Ae. dalzieli1.9 (339)Ae. ochraceus0.4 (75)Ma. uniformis0.4 (74)Ma. africana0.2 (6)Others91.1% (16728/18368)Cx. quinquefasciatus28.7 (5274)Cx. cinereus24.5 (4499)Cx. nebulosus13.3 (2449)An. gambiae2.2 (400)Ae. aegypti2.1 (392)Rainy seasonRVFV vectors23.8% (11306/47462)Cx. poicilipes13.8 (6569)Ae. vexans3.4 (1613)Cx. tritaeniorhynchus3.2 (1497)Ma. uniformis1.7 (805)Ae. dalzieli0.8 (399)Ae. ochraceus0.8 (358)Ma. africana0.1 (36)Others76.2% (36156/47462)Cx. quinquefasciatus20.2 (9564)Cx. cinereus16.6 (7890)Cx. neavei13.5 (6422)Cx. nebulosus11.9 (5627)An. coustani1.8 (862)An. gambiae1.6 (738)Cx. = Culex, Ae. = Aedes, Ma. = Mansonia, An. = Anopheles, RVFV = Rift Valley fever virus, n = number. Table 1Seasonal fluctuation of mosquitoes, central and southeastern Senegal, 2021-2024.Table 1SEASONMosquito categoriesMosquito speciesPercent (n)Dry seasonRVFV vectors8.9% (1640/18368)Ae. vexans3.1 (570)Cx. poicilipes3.0 (555)Ae. dalzieli1.9 (339)Ae. ochraceus0.4 (75)Ma. uniformis0.4 (74)Ma. africana0.2 (6)Others91.1% (16728/18368)Cx. quinquefasciatus28.7 (5274)Cx. cinereus24.5 (4499)Cx. nebulosus13.3 (2449)An. gambiae2.2 (400)Ae. aegypti2.1 (392)Rainy seasonRVFV vectors23.8% (11306/47462)Cx. poicilipes13.8 (6569)Ae. vexans3.4 (1613)Cx. tritaeniorhynchus3.2 (1497)Ma. uniformis1.7 (805)Ae. dalzieli0.8 (399)Ae. ochraceus0.8 (358)Ma. africana0.1 (36)Others76.2% (36156/47462)Cx. quinquefasciatus20.2 (9564)Cx. cinereus16.6 (7890)Cx. neavei13.5 (6422)Cx. nebulosus11.9 (5627)An. coustani1.8 (862)An. gambiae1.6 (738)Cx. = Culex, Ae. = Aedes, Ma. = Mansonia, An. = Anopheles, RVFV = Rift Valley fever virus, n = number. Seasonal fluctuation of mosquitoes, central and southeastern Senegal, 2021-2024. Cx. = Culex, Ae. = Aedes, Ma. = Mansonia, An. = Anopheles, RVFV = Rift Valley fever virus, n = number. Seroconversion of animal sentinel herds In total, 2989 serum samples from 976 animals were tested for RVFV antibodies (Table 2), showing an overall seroconversion proportion of 7.2% (95% CI: 5.6-9.0%). No significant differences in proportions of seroconverted animals (RVFVSP) were observed between areas (Linguère: 8.0%, 95% CI: 5.8-10.7% vs Kédougou: 6.3%, CI: 4.2-8.9%; p = 0.071) and sex (male: 3.6%, 95% CI: 1.3-7.7% vs female: 7.9%, 95% CI: 6.1-10.0%; p = 0.47). However, RVFVSP was significantly higher in goats (9.5%, 95% CI: 7.1-12.4%) than sheep (4.6%, 95% CI: 2.9-6.9%; p = 0.004), with goats showing a OR of 2.6 (95% CI: 1.5-4.6; p = 0.001). RVFVSP differed significantly across districts in both Kédougou (p = 0.004) and Linguère (p = 0.038). The highest RVFVSP were recorded in Dimboli (12.5%, 95% CI: 6.8-20.4; p = 0.011) in the Kédougou area and Barkedji (12.6%, 95% CI: 7.5-19.4; p = 0.038) in the Linguère area.Table 2Seroconversion of animal sentinel herds for Rift Valley fever virus and potential risk factors, central and southeastern Senegal, 2021-2024.Table 2RISK FACTORSNaive animalSeroconverted animalOR (95%CI)ap*nn%95%CIp*Seropositivity976707.25.6 - 9.0<0.001AreaLinguère513418.05.8 - 10.70.357Kédougou463296.34.2 - 8.9Age (month)Young (5-18)577274.73.1 - 6.7<0.001Ref.Adult (>18)3994310.87.9 - 14.22.9 (1.7-5.0)<0.001SexMale16763.61.3 - 7.70.071Female809647.96.2 - 10.0SpeciesSheep460214.62.9 - 6.90.004Ref.Goats516499.57.1 - 12.42.6 (1.5-4.6)0.001Districts of Kédougou463296.34.2 - 8.9<0.001Tomboronkoto9211.10.0 - 5.90.011Ref.Bandafassi7334.10.9 - 11.53.9 (0.5-9.8)0.243Dimboli1041312.56.8 - 20.413.0 (2.5-38.6)0.014Dindefelo11254.51.5 - 10.14.3 (0.7-8.3)0.190Fongolimbi8278.53.5 - 16.88.5 (1.5-16.6)0.048Districts of Linguère513418.05.8 - 10.7<0.001Gassane7633.90.8 - 11.10.038Ref.Barkedji1351712.67.5 - 19.44.7 (1.5-21.1)0.019Dodji12086.72.9 - 12.72.3 (0.6-10.9)0.242Thiargny10798.43.9 - 15.42.8 (0.8-13.1)0.141Thiel7545.31.5 - 13.11.5 (0.3-8.2)0.590* The p-values reported in the first part of table correspond to univariable comparisons (chi-square tests), whereas the odds ratios (ORs), 95% confidence intervals, and associated p-values correspond to the multivariable logistic regression analysis.a95% CI based on Clopper-Pearson intervals; CI = confidence interval; OR = odds ratio; Ref. = reference category. Table 2Seroconversion of animal sentinel herds for Rift Valley fever virus and potential risk factors, central and southeastern Senegal, 2021-2024.Table 2RISK FACTORSNaive animalSeroconverted animalOR (95%CI)ap*nn%95%CIp*Seropositivity976707.25.6 - 9.0<0.001AreaLinguère513418.05.8 - 10.70.357Kédougou463296.34.2 - 8.9Age (month)Young (5-18)577274.73.1 - 6.7<0.001Ref.Adult (>18)3994310.87.9 - 14.22.9 (1.7-5.0)<0.001SexMale16763.61.3 - 7.70.071Female809647.96.2 - 10.0SpeciesSheep460214.62.9 - 6.90.004Ref.Goats516499.57.1 - 12.42.6 (1.5-4.6)0.001Districts of Kédougou463296.34.2 - 8.9<0.001Tomboronkoto9211.10.0 - 5.90.011Ref.Bandafassi7334.10.9 - 11.53.9 (0.5-9.8)0.243Dimboli1041312.56.8 - 20.413.0 (2.5-38.6)0.014Dindefelo11254.51.5 - 10.14.3 (0.7-8.3)0.190Fongolimbi8278.53.5 - 16.88.5 (1.5-16.6)0.048Districts of Linguère513418.05.8 - 10.7<0.001Gassane7633.90.8 - 11.10.038Ref.Barkedji1351712.67.5 - 19.44.7 (1.5-21.1)0.019Dodji12086.72.9 - 12.72.3 (0.6-10.9)0.242Thiargny10798.43.9 - 15.42.8 (0.8-13.1)0.141Thiel7545.31.5 - 13.11.5 (0.3-8.2)0.590* The p-values reported in the first part of table correspond to univariable comparisons (chi-square tests), whereas the odds ratios (ORs), 95% confidence intervals, and associated p-values correspond to the multivariable logistic regression analysis.a95% CI based on Clopper-Pearson intervals; CI = confidence interval; OR = odds ratio; Ref. = reference category. Seroconversion of animal sentinel herds for Rift Valley fever virus and potential risk factors, central and southeastern Senegal, 2021-2024. * The p-values reported in the first part of table correspond to univariable comparisons (chi-square tests), whereas the odds ratios (ORs), 95% confidence intervals, and associated p-values correspond to the multivariable logistic regression analysis. 95% CI based on Clopper-Pearson intervals; CI = confidence interval; OR = odds ratio; Ref. = reference category. Seasonal and interannual dynamics of Rift Valley fever virus infection The seroconversion for RVFV varied significantly by season (p = 0.023) and year (p < 0.001) (Table 3). It was higher during the rainy season (9.1%, 95% CI: 6.7–12.0%) than the dry season (5.2%, 95% CI: 3.4–7.6%). Seroconversion peaked in 2021–2022 (12.2 and 11.7%), and declined thereafter. In Kédougou, the highest seroconversion occurred in 2021, whereas in Linguère it peaked in 2022 (16.9%, 95% CI: 10.9–24.5%).Table 3Inter-annual dynamics of animal seroconversion, central (Linguère) and southeastern (Kédougou) Senegal, 2021-2024.Table 3YearOverall seroconversionSeroconversion in LinguèreSeroconversion in Kédougou202112.2 (7.7 - 18.1)4.4 (0.9 - 12.4)17.3 (10.6 - 26.0)202211.7 (7.9 - 16.3)16.9 (10.9 - 24.5)5.9 (2.4 - 11.7)20234.7 (2.4 - 8.3)6.2 (2.9 - 11.4)2.3 (0.3 - 8.2)20242.8 (1.3 - 5.2)4.1 (1.7 - 8.4)1.3 (0.2 - 4.6)p-value<0.001<0.001<0.001 Table 3Inter-annual dynamics of animal seroconversion, central (Linguère) and southeastern (Kédougou) Senegal, 2021-2024.Table 3YearOverall seroconversionSeroconversion in LinguèreSeroconversion in Kédougou202112.2 (7.7 - 18.1)4.4 (0.9 - 12.4)17.3 (10.6 - 26.0)202211.7 (7.9 - 16.3)16.9 (10.9 - 24.5)5.9 (2.4 - 11.7)20234.7 (2.4 - 8.3)6.2 (2.9 - 11.4)2.3 (0.3 - 8.2)20242.8 (1.3 - 5.2)4.1 (1.7 - 8.4)1.3 (0.2 - 4.6)p-value<0.001<0.001<0.001 Inter-annual dynamics of animal seroconversion, central (Linguère) and southeastern (Kédougou) Senegal, 2021-2024. Discussion Although sporadic human cases, animal infections and detection of RVFV in mosquito vectors have been frequently reported in Senegal [12,19,22,38,45,46], the spatial and temporal dynamics of RVFV circulation remain poorly characterized. This study reports the first continuous seasonal combined entomological and serological surveillance using naive sheep and goats as sentinel herds, in close to human populations, conducted between 2021 and 2024. Our findings demonstrated an active RVFV circulation in both central and southeastern Senegal during the inter-epidemic period (2021–2024), approximately one year before the 2025 RVF outbreak in Senegal. This serological evidence, despite the absence of viral RNA detection in mosquitoes, demonstrated ongoing low-level virus circulation in livestock. The subsequent detection of RVFV in mosquito super-pools during the 2025 outbreak in affected regions of the country suggests that a marked increase in viral amplification occurred after the surveillance period, highlighting the complementary value of longitudinal serological surveillance for identifying areas at risk before clinical outbreaks become apparent. The declining seroconversion proportion observed between 2021 (12.2%) and 2024 (2.8%) is consistent with reduced RVFV transmission during the inter-epidemic period rather than the absence of viral circulation, potentially maintained by the periodic hatching of vertically infected floodwater Aedes eggs [11] and limited transmission among susceptible vertebrate hosts. Moreover, the progressive decline in animal seroconversion suggests an increasing proportion of immunologically naive livestock within the population. Previous studies have shown that a reduction in herd immunity during inter-epidemic periods may increase the risk of RVFV amplification once favorable environmental conditions occur, thereby increasing the likelihood of outbreaks in both animals and humans [47]. The exceptional rainfall and flooding recorded in 2025 may have provided these favorable conditions, triggering extensive vector proliferation [47]. Conversely, in hyperendemic areas where RVFV circulates more continuously, repeated exposure of livestock may result in high levels of herd immunity. Because livestock are the principal amplifying hosts and the major source of human infection [8], high herd immunity may reduce the likelihood and magnitude of epizootics and consequently decrease the risk of human infection, despite ongoing low-level viral circulation [48]. Our data showed a high abundance of mosquito species known as RVFV vectors and repeated seroconversions in sentinel animals. Furthermore, several abortions were reported in small ruminants, including sentinel animals, supporting evidence of active RVFV circulation. Indeed, the overall seroconversion proportion of 7.2% across both areas, indicates active enzootic transmission of the virus [14,32,49,50]. The absence of infected mosquitoes in our collection, despite the detection of RVFV circulation in animal herds, could be explained by low level of infection in mosquito populations, which is a common challenge in arbovirus surveillance. During inter-epidemic periods or in areas with low-level virus transmission, mosquito infection rates are often extremely low, requiring the screening of very large numbers of vectors to detect a small number of infected individuals. Although super-pooling may reduce analytical sensitivity because viral RNA from infected mosquitoes can be diluted, our entomological surveillance system successfully detected RVFV RNA in super-pools composed of up to 5 pools of 100 mosquitoes each during the 2025 RVF outbreak in other regions of Senegal. Moreover, 42 isolates of 7 viruses were detected in mosquito populations processed using this super-pooling protocol, demonstrating the efficacy of this approach [39, WHO Collaborating Centre for Arboviruses and Hemorrhagic Fever Viruses, unpublished data]. Therefore, the absence of RVFV RNA detection in the present study is more likely attributable to the extremely low infection prevalence in mosquito populations during the inter-epidemic period than to limitations of the super-pooling strategy itself. Goats showed significantly higher infection rates than sheep, in agreement with earlier findings [51,52], probably reflecting higher preference of RVFV vectors for goats [53]. Thus, goats should be prioritized in RVFV surveillance systems. Seroconversion increased significantly with age of animal, reaching over two times higher in older than in younger animals. This trend can be explained by a cumulative regular exposure of animals to RVFV with age, increasing progressively the RVFVSP. The results are in agreement with previous studies [32,46]. Higher RVFVSP during the rainy season concorded with peaks in mosquito vector abundance, driven by more favorable ecological conditions including flooding of temporary ponds with dense vegetation around (breeding and resting places of vectors) [19,22,38,54,55]. There was no significant difference for RVFVSP between central and southeastern areas; contrary to previous comparative animal serology studies indicating highest RVFVSP in the central area [11,56]. The observed geographical change in RVFV infection patterns may be explained by evolving ecological and anthropogenic factors that influence virus transmission. In southeastern Senegal, changes in land use, vegetation cover, vector abundance, and livestock transhumance may have altered the local epidemiology of RVFV. Over the past two decades, central and northern Senegal have experienced substantial fluctuations in vegetation indices, temperature, and rainfall, whereas the southern part of the country has remained comparatively stable, with more persistent vegetation cover and climatic conditions favorable for mosquito populations [57]. The absence of significant geographical differences should not be interpreted as evidence of a permanent shift in RVFV distribution but rather as an indication of the highly dynamic nature of RVFV transmission in Senegal. This apparent discordance may be explained by differences in study period, surveillance design, ecological conditions, and transmission intensity. Our study was conducted during an inter-epidemic period, when RVFV circulation is expected to be spatially heterogeneous and strongly influenced by local vector dynamics, livestock movements, herd immunity, and environmental conditions. Continued longitudinal surveillance across multiple ecological zones will be essential to determine whether these observations represent transient fluctuations or an emerging change in RVFV transmission patterns. Furthermore, within southeastern Senegal, most seroconversions were recorded in the districts of Dimboli and Fongolimbi, which accounted for 69.0% (20/29) of all seroconversions detected in the region. These districts, located southeast of Kédougou and bordering Mali and the Republic of Guinea, lie within a transboundary area where RVFV is known to circulate [[58], [59], [60], [61]]. Frequent cross-border movement of people, including workers involved in artisanal gold mining, together with seasonal livestock transhumance, may facilitate the introduction and maintenance of RVFV through the movement of infected animals and increased opportunities for vector-host contact. These factors may have contributed to the relatively higher RVFV activity observed in this part of southeastern Senegal. Moreover, compared to Tomboronkoto, the district of Dimboli is separated from Kédougou town by the Gambia River. The riverine ecosystem, characterized by dense tree and grass vegetation, provides abundant resting sites for mosquitoes, while the permanent water body and associated floodplains create favorable conditions for vector survival and breeding. These ecological conditions may increase vector abundance, extend livestock exposure to mosquito bites, and thus facilitate local RVFV transmission and amplification. In the Linguère area, Barkedji exhibited the highest seroconversion proportion among naive animals, consistent with previous studies identifying this district as a major hotspot of RVFV activity in Senegal [11,20,22]. This district is characterized by many temporary ponds, constituting natural habitats of many species of birds, reptiles and rodents, and the breeding and resting sites for RVFV mosquito vectors. During the rainy season, livestock congregate around these ponds for water and grazing, increasing their exposure to infected mosquitoes and promoting viral amplification in an area where RVFV is known to be endemic [10,11,53]. Overall, the marked heterogeneity in seroconversion observed between neighboring districts suggests that RVFV transmission is strongly influenced by local environmental and anthropogenic factors operating at a fine spatial scale. In addition to ecological conditions, factors such as livestock density, grazing practices, seasonal transhumance, and cross-border movement of animals and people may contribute to these localized differences. Future studies integrating environmental, entomological, livestock movement, and socio-ecological data are needed to better understand the drivers of this small-scale spatial heterogeneity. The highest RVFV seroconversion proportion was observed in 2021 rather than in 2022, that was the wettest year of our study period, with an annual rainfall of 917 mm compared to 769 mm in 2020, 718 mm in 2021, 783 mm in 2023 and 779 mm in 2024 [62]. This discrepancy suggests that inter-annual variation in RVFV seroconversion was not directly associated with total annual rainfall, supporting previous observations that the timing, distribution, and intensity of rainfall events are more influential than cumulative yearly rainfall in shaping RVFV transmission dynamic [63]. In contrast, 2025 was characterized by exceptional hydroclimatic conditions. On 14 August 2025, a rainfall of 150 mm was recorded, substantially exceeding the long-term average for August (115.5 mm), and was accompanied by persistently above-average rainfall, rising river levels, and water levels in the Senegal River Basin exceeding critical retention thresholds [64]. These conditions resulted in controlled water releases and extensive flooding across northern and central Senegal [64]. Such flooding likely created favorable ecological conditions for the mass hatching of desiccation-resistant eggs of floodwater Aedes spp., some of which may have been vertically infected with RVFV. The emergence of infected adult mosquitoes could have initiated virus amplification among livestock, with subsequent transmission by secondary vectors, particularly Culex spp., leading to widespread infection in ruminants and spillover to humans [11,47]. These observations further emphasize that the temporal pattern and hydrological consequences of rainfall, rather than annual rainfall totals alone, are key determinants of RVFV emergence. The widespread 2025 RVF outbreak was likely driven by sustained enzootic RVFV circulation, declining herd immunity that increased livestock susceptibility, and exceptional rainfall and flooding that promoted massive vector proliferation and virus transmission. Vertically infected floodwater Aedes mosquitoes may contribute to RVFV maintenance and amplification, however, our data do not allow assessment of their relative contribution compared with increased vector abundance and host susceptibility. Although rainfall is a key environmental factor of RVFV transmission through its influence on mosquito breeding habitats, other climatic factors, including temperature, relative humidity, vegetation dynamics, and surface flooding, also affect vector abundance and RVFV transmission. The absence of these data represents a limitation of the present study. Future studies integrating climatic, hydrological, entomological, and serological data will improve understanding of RVFV transmission dynamics and improve outbreak prediction. Conclusion This study shows high seroconversion proportions in sentinel herds across central and southern Senegal, demonstrating the existence of multiple, active enzootic RVFV transmission foci in the different bioclimatic areas of the country in the years preceding the 2025 major outbreak. The results showed that RVFV circulates annually, with clear spatio-temporal variations, host species, age and small-scall geographical levels. Although sentinel herd surveillance alone cannot predict the timing or occurrence of outbreaks, its integration with entomological, climatic, hydrological, and livestock movement surveillance could substantially strengthen RVFV risk assessment and contribute to earlier identification of conditions favorable for epidemic emergence. CRediT authorship contribution statement Déthié Ngom: Writing – review & editing, Writing – original draft, Visualization, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Elisabeth Thérèse Faye: Writing – review & editing, Methodology, Formal analysis, Data curation. Aliou Khoulé: Writing – original draft, Methodology, Investigation, Data curation. Sokhna Mayemouna Diop: Writing – original draft, Methodology, Formal analysis, Data curation. Ousseynou Sène: Writing – original draft, Methodology, Formal analysis, Data curation. Mignane Ndiaye: Writing – original draft, Methodology, Formal analysis. Cheikh Talla: Writing – original draft, Software, Methodology. Cheikh Loucoubar: Writing – original draft, Software, Methodology. Oumar Faye: Writing – original draft, Validation, Supervision, Project administration, Methodology, Conceptualization. Mawlouth Diallo: Writing – original draft, Resources, Methodology, Funding acquisition, Conceptualization. Yamar Ba: Writing – original draft, Validation, Supervision, Project administration, Methodology, Investigation, Funding acquisition, Data curation, Conceptualization. Gamou Fall: Writing – original draft, Visualization, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Conceptualization. Diawo Diallo: Writing – original draft, Visualization, Validation, Supervision, Project administration, Methodology, Investigation, Data curation, Conceptualization. Funding This work was supported by the National Institute of Allergy and Infectious Diseases (NIAID) of the National Institutes of Health (NIH) [grant numbers U01AI151758]. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. 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.