Our study was limited to the southern part of Nigeria (Fig 1), as about 94% of the confirmed cases were within this area. Only 12 out of 199 confirmed monkeypox cases by state, September 2017- September 2021 were from the northern part of Nigeria. Rainfall is often the key climatic variable, and there is a marked distinction of wet and dry seasons in most areas. Two air masses control rainfall-moist northward-moving maritime air coming from the Atlantic Ocean and dry continental air coming south from the African landmass. The climate becomes drier as one moves north from the coast. The estimated human population in southern Nigeria was 91,442,452 in 2021 [24].

A total of 68 climatic and human population density variables (S2 Table) were used which was extracted from the WorldClim version 1.4 dataset from 1950–2000 at 30 arc–second resolution [32] and worldpop dataset available at www.worldpop.org. WorldPop dataset are available to download in Geotiff and ASCII XYZ format at a resolution of 30 arc-seconds. Also, it is the unconstrained individual countries 2000–2020 population count datasets. We explored the existence of multicollinearity using the variance inflation factor (VIF). Multicollinearity often arises in statistical models, and can be a severe problem for parameter estimation because it inflates the variance of regression parameters and hence potentially leads to the wrong identification of relevant predictors in an ecological model [33]. The Variance inflation factor represents the amount of variability of a covariate which is explained by other covariates. A VIF >10 has been considered to indicate highly correlated variables which are then typically removed from the input data set [34]. Of the 68 covariates initially tested for multicollinearity, thirty-six (36) were excluded from further analysis. All remaining covariates were considered to be independent and were included in the analysis. The multicollinearity test was implemented using the statistical package for social sciences (spss v22.0). Furthermore, the remaining bioclimatic variables, human population density and elevation were subsequently processed in MaxEnt algorithm, three methods; Jackknife test, backward stepwise variable elimination, and the variable response curves were selected to identify the relative contribution of predictor variables to the model. Variables with a percentage contribution less than one percent were removed from the model.

A MaxEnt model was fitted using 100 bootstrap runs and a 70/30 partition percentage for the training/testing datasets. The advanced options in MaxEnt that were selected include the maximum iteration set to 5000 to allow the models to have enough time to reach convergence at 0.00001 [35], and 90% sensitivity was set within the MaxEnt model for determining suitability. The area under the Receiver Operating Characteristics [36] was used to assess the accuracy of the model. In the MaxEnt model, the Area Under the Curve (AUC) of the receiver operating characteristic plot was used as an evaluation criterion to assess the accuracy of the model [37]. We reclassified the MaxEnt spatial model output into three risk classes, namely high, moderate and low risk areas.

In this study, an ecological niche modeling technique was used to predict the potential suitable habitat distribution of the monkeypox virus in southern Nigeria. Our study presents the first risk assessment of ecologically suitable areas for monkeypox disease in Nigeria. Precipitation of the driest quarter (15-25mm) is the most important predictive variable in our model, with a contribution of 47%. Other study have shown that monkeypox disease outbreaks are associated with precipitation as the chief environmental predictor in the model [38]. Also, during the comparison between the Central and West African monkeypox forms, it was observed that monkeypox in West Africa occurs in a wetter and hotter condition which is a subtle difference in ecological condition to its Central African form [38]. The second most important predictive variable is elevation (below 800 m. a.s.l). The relatively low elevation in the south could be the primary reason for a higher number of monkeypox cases when compared to the northern counterpart. Our model predicted human population density as an important variable in monkeypox distribution, which may be caused by increased human-environment interaction. Minimum temperature in December (above 22°C) and maximum temperature in March (32°C) are the least predictive variable in our model. Temperature has been suggested to play an important role in transmitting and spreading infectious diseases [39]. However, the response curves remind us that the appropriate temperature (minimum temperature in December = above 22°C) and (maximum temperature in March = 32°C) can increase the risk of MPX disease. These alert us to pay close attention to temperature and precipitation seasonality in preventing MPX in Nigeria. Our study reveals that climatic factors and elevation are fundamental conditions that influence MPXV distribution in southern Nigeria. Also, human population density and other related human activities are specific factors reshaping the spatial distribution of MPXV. The predicted variables don’t translate to singly influencing the disease distribution or suitability. Some variables might act in proxy or directly influencing the host abundance, behaviour or activities. Two or more variables might jointly act to foster disease outbreak.

Our model predicted areas with suitable environmental conditions for MPXV around southern Nigeria. The high-risk area covered the southwest, southeast and south-south part of the region, which include Lagos, Delta, Bayelsa, Rivers, Imo, Abia, Akwa Ibom, and Cross River.

The predicted geographic distribution of MPX defines some suitable areas with no historical outbreak records in Ondo, Osun, and Ebonyi state (only one confirmed case), which may either be over-prediction of the model or poor surveillance systems, leading to severe under-reporting and hence misleading disease burden information. However, it is important to note that over-predicting the geographic distribution of a species does not necessarily infer prediction error. The potentially over-predicted geographical distribution areas may represent an accurate illustration of the spatial disease extent [40], despite the lack of presence records that could be used for testng the accuracy of our model in those states.

Monkeypox has a low fatality rate (0–11%), with the highest rates occurring in children and young persons who were not part of the population vaccinated against smallpox. Previous study once suggests that consumption of bush meats could be a potential risk factor in the transmission of the infection, as could caring for an infected patient [15]. A new vaccine was produced in 2019 for the prevention of monkeypox. However, there is an ongoing study to validate the effectiveness and safety of the Modified Vaccinia Ankara (MVA) vaccine in human cases as it is believed to confer immunity against all orthopoxviruses [41, 42].