1 Introduction

Flooding and flood-related menaces have continuously plagued the global world annually. Of all the natural disasters, flooding has offered the most devastating impact on human existence in recent times [29]. Flooding affects approximately 170 million people annually and has accounted for around 8 million deaths in the past 30 years [2, 55, 57]. Approximately 2.3 billion lives have been swayed by flood impacts, resulting in about USD 386 billion in economic depletion worldwide [23].

Flooding simply means an excessive water flow, which occurs when the water flow velocity is higher than the average volume of a river, usually during episodes of heavy rainfall, dam failures, or melting piles of snow. Flooding is a direct consequence of climate change which affects the entire continent, and this is responsible for the high severity and prevalence of tropical cyclones, typhoons, and hurricanes, as well as various other water-related severe weather spectacles [22, 44]. Extremely high global temperatures facilitate the formation of convection currents, which produce excessive rainfall as its end-product. Global warming is responsible for the continuous melting of ice caps in the polar regions of the world and the consequent global increase in sea levels which contributes to coastal and river flooding [16].

Nigeria is not left out on the list of countries affected by flooding disasters; Nigeria is one of the most affected developing countries in Africa regarding flooding incidences. There is virtually no state in the country that does not experience one or two flooding incidences annually. Over the last ten years, reported flooding cases in Nigeria showed that virtually all the thirty-six states within the country are susceptible to flood disasters [33]. A total of 1550 people died from flooding in 2010, while the disaster displaced approximately 258,000 people [6, 41]. In 2012, 361 lives were lost due to flooding, and 3.8 million people were displaced, costing a total of US$ 6.5 billion in damages [10, 56]. 2015 to 2021 witnessed a far more devastating flooding menace than the previous years, primarily due to the failure of governments and policymakers to learn from past flooding occurrences and the inability to put proper flood forecasting and disaster management systems in place.

The leading mechanisms for flooding in Nigeria include lack of proper drainage systems, inundation of drainage channels because of obstructions, emergency dam releases, and heavy rainfall. Further, rapid population growth, urbanization, and rising sea-levels are the key factors responsible for floods in Nigeria and globally [14, 18].

Ilorin, which is the capital city of Kwara state, north central Nigeria, together with its neighboring towns within the state have continued to witness yearly flooding events. Spontaneous population increase within the state has led to competition for land, increasing waste generation, substandard and limited drainage systems, and poor waste management habits. Flood disasters in the area between the years 1997 and 2021 have mostly occurred within Ilorin, Edu, Patigi, Kpada, Gbogbondogi, Moro, and Lafiaji. Several lives have been lost due to flooding within the state, and multimillion naira worth of properties have been destroyed due to flood disasters [36, 39, 43]. It was reported that flooding in Ilorin city mainly occurs in the rainy season when increased precipitation results in an overflow of water from the Asa River [4, 27] and its tributaries. Effective flood mitigating activities have not been effectively implemented to curb this menace, hence, the yearly reoccurrence of flood disasters within Ilorin.

A major step in mitigating flood occurrences is flood-risk assessment. While several authors have analyzed flooding within the Kwara State [4, 39, 42, 43, 52], there is still inadequacy in the adoption of geospatial (Remote Sensing (RS) and Geographic Information System (GIS)) technology for flood-risk assessment. GIS and RS techniques offer robust flood disaster forecasting and occurrence prediction tools. It has a powerful and reliable tool that provides valuable guidance during discussions aimed at solving flood disasters [12]. Actions geared towards ameliorating flood disasters might only be successful with a proper flood vulnerability map. Several authors have used GIS and RS techniques in flood vulnerability analyses globally [1, 7, 15, 19, 30, 45, 48, 60], and in some parts of the Nigeria [2, 3, 5, 8, 13, 17, 24, 34, 35, 38, 40, 53, 54].

Hence, this present study applied GIS and RS techniques in flood-risk mapping of Ilorin and its environs via Analytical Hierarchy Process (AHP). AHP offers a rapid, cost effective, accurate, and mathematically proven scientific methodologies which are applicable to a wide spectrum of geoscience applications [9]. It is mostly applicable as a first-call tool for a quick delineation and mapping of risk zones in environmental disaster monitoring. For this reason, this method has been adopted in this research and combined with field investigation data for better accuracy. Also, Land Surface Temperature and Soil Moisture Index, which are rarely considered in flood investigations have been applied in this study.

2 Study Area Description

The area of investigation sits within the Nigerian North central geopolitical zone. It covers the City of Ilorin, the capital of Kwara State, and the surrounding major towns that lie within Ilorin South, West and East, Moro, and Asa Local Government Areas of Kwara State. The study area is sandwiched within latitudes 8°15’and 9°10’North of the Equator and longitudes 4°15’ and 4°50ʹEast of the Greenwich Meridian and it has an approximate average altitude value of 300 m (Fig. 1).

Fig. 1
figure 1

The study area: Ilorin and its environs

Full size image

Ilorin and environs fall within the transitional zone between savanna and forest regions of Nigeria; it has an annual mean rainfall value of 1200 mm and an average monthly temperature that ranges between 25 °C and 29.5 °C, with March having the highest 30 °C [32]. The study area also forms part of Nigeria's Lower Niger River Basin, covering Kwara state, southern parts of Niger state, and Kogi State. Ilorin area is drained by the tributaries of River Niger in the north and across the western, eastern, and southern margins. Major rivers within the study area include Asa, Agba, Osere, Alalubosa, Okun, and Aluko. Three significant dams exist in the study area: Asa, Agba, and Sobi. According to the National Population Census of 2006 [31], the region's resident population was 1,015,317, and it covered an area of about 3,158.9 km2. Oko-Erin, Unity, Bode-Saadu, Ganmo, Fufu, Oke-Oyi, Shao, Idofian, Malete, Alakuko, Oloru, and Onikoko are notable locations within the research area. Geologically, the research area is found in the Southwestern Nigeria’s Basement Complex terrain. Igneous and metamorphic rocks such as gneisses, migmatites, granites, quartzites, schists, and amphibolites are the major rock groups in the area.

3 Materials and Method

3.1 Materials

For this investigation, primary and secondary datasets were used. The primary data consists of locations of the places that have flooded within the Ilorin and its environs which was collected using the Global Positioning System (GPS) gadget. The area's Digital Elevation Model (DEM) and LANDSAT 8 Operational Land Imager (OLI) data are the secondary data. Satellite data were acquired from earthexplorer.usgs.gov. Details of the data applied for this research are given in Table 1.

Table 1 Primary and secondary data types and their sources
Full size table

3.2 Method

For a GIS-assisted flood-risk mapping of Ilorin and its environs, ten (10) factors of substantial influence on flooding were considered. The factors are; Land Surface Temperature (LST), Elevation, Soil Moisture Index (SMI), Distance to Stream (DS), Drainage Density (DD), Stream Power Index (SPI), Normalized Difference Vegetation Index (NDVI), Slope, Land Use Land Cover (LULC), and Topographic Wetness Index (TWI). The combined process of flood vulnerability assessment was primarily divided into four (Fig. 2). Processing and analysis were carried out via the use of GIS and RS software such as ENVI and ArcGIS.

Fig. 2
figure 2

Stages in flood-risk assessment of Ilorin and its environs

Full size image

The study area’s elevation map was produced via the merged DEM data, elevation map was further used to generate the slope map. Both the elevation and slope maps were further processed to produce SPI and TWI maps. The DD map was computed via streamlines generated from the DEM with the use of ArcHydro tool. DS was also computed using the spatial analyst tool in ArcGIS to buffer areas close to and far away from significant streams. Merged LANDSAT 8 OLI data of the area was processed for basic corrections using ENVI software, and the processed data was used to compute SMI, NDVI, and LST maps. LULC mapping were achieved via a supervised classification technique in GIS.

The classification of the layers and weighting was done using the AHP approach, and the resulting thematic layer maps were created in a pairwise comparison table. Validation was done by plotting coordinates of affected areas in the state on the output flood-risk map of Ilorin and its environs.

3.3 Mechanisms of Flooding Considered in this Study

3.3.1 Elevation and Slope

The elevation of an area refers to the height to which the area is elevated or rises, usually above sea level. Elevation largely influences the water discharge rate given the fact that water flows downhill from high elevations to low elevations; therefore, areas in low-elevation regions have higher flood risk than high-elevation regions [9, 37]. The slope of an area conveys the landscape steepness in that area. It provides basic information about the geodynamic processes at work in the area. Slope directly affects water surface discharge and infiltration potency during runoff,therefore, a higher slope leads to rapid runoff and lesser infiltration, while a lower slope contributes to significantly lower runoff and higher infiltration rate during rainfall. Elevation values in the studied part of Ilorin and environs were extracted from the processed SRTM DEM map of the area (Fig. 3a), and ranged from 71 to 566 m. Average elevation value of Osogbo is 320 m [32], therefore, derived elevation values from the DEM map of the study area were divided into five categories: very low elevation (71 m to 203 m), low elevation (203 m to 270 m), moderate elevation (270 m to 321 m), high elevation (321 m to 371 m), and very high elevation (566 m or above). While 42% of the overall area of focus sits in the high-elevation zones, approximately 58% of it is located in moderate to low-elevation zones. High-elevation places have less of a chance of flooding, while low-elevation areas have a high likelihood of flood threats. Slope angle values were also derived from the SRTM DEM data of the study area using ArcHydro tool in ArcMAP and ranged between 0–36° (Fig. 3b). The slope values were reclassified into five, where 0° to 2° represents a very low slope angle, 2° to 3°, 3° to 6°, 6° to 12°, 12° to 36° signifies low, moderate, high, and very high slope angles, respectively. The moderate and low slope angle zones make up around 85% of the research area, whereas the high slope angle zone makes up 15% (Table 2).

Fig. 3
figure 3

a Map showing elevation variations within Ilorin and its environs, b Map showing variation of Slopes angles across the study area

Full size image
Table 2 Pairwise comparison matrix
Full size table

3.3.2 Drainage Density (DD) and Distance to Stream (DS)

Drainage Density describes the level to which a drainage basin is drained by stream channels and therefore forms a crucial factor in the flood-risk assessment. It is computed by dividing the area of the drainage basin by the total length of all the rivers and streams within the drainage basin. The closer streams are together, the higher the drainage density value; flood danger rises as DD rises and vice versa [20]. The DD map of the focus area (Fig. 4a) was derived from the SRTM DEM data of the area by applying the line density tool on the extracted drainage lines in ArcMAP. Figure 4a shows that DD values ranged from 0 to 0.74 km/km2. While only 26% of the area under investigation has high to very high DD, the remaining 74% has moderate to low DD values (Table 2), which is anticipated to increase floods during periods of severe rainfall. The chance of flooding in a location after a heavy downpour is also influenced by how far away it is from a big stream [50]. Because river-induced floods occur more frequently and with greater intensity closer to the drainage line, considering the closeness of streams is essential when determining flood risk. DS was computed using Euclidean distance tool in ArcMAP with a maximum allowed distance of 2 km. Areas between 0 to 0.2 km of major streams are considered close to a major stream, while areas between According to the study area’s DS map (Fig. 4b), 46% of the area is close to a major stream and has a high risk of flooding, while the remaining 54% is far away and has a reduced risk of flooding.

Fig. 4
figure 4

a Drainage Density distribution within the study area, b Distance to Streams (m) variation in the study area

Full size image

3.3.3 TWI (Topographic Wetness Index) and SPI (Stream Power Index)

TWI demonstrates the regional distribution of wetness, soil moisture, saturation zone, groundwater level, and flow aggregation. A high TWI rating implies wet, extremely susceptible to flooding land. Equation 1 below was used to compute the TWI (after [11]:

$$TWI=\mathrm{ln}(\frac{\alpha }{tan\beta })$$
(1)

where tan β is the slope angle at the location and α represents the total upslope area draining via a single spot (per unit contour length). Compared to 19% of the area, which has high TWI values and is likely prone to flooding, 81% of the Ilorin and its environs have low TWI values and less likely to flood during periods of severe rainfall (Fig. 5a). SPI describes the volume of water flowing over a watershed as contrasted to TWI [47]. It illustrates how frequently flooding occurs. Low SPI levels indicate minimal to no downstream water movement, which is more likely to cause floods, while high SPI values suggest swift downstream water flow which implies low flood danger.

$$SPI=Astan\beta$$
(2)

where β is the slope radiant (in °), and As depicts the catchment area (in m2 m−1). The SPI map (Fig. 5b) for the study area was segmented into five (5) groups, with low SPI accounting for 98% of the region. As a result of the remaining 2%'s high SPI, the majority of the locations have low SPI values and are vulnerable to flooding.

Fig. 5
figure 5

a Topographic Wetness Index variation in the study area, b Power Index of streams within Ilorin and its environs

Full size image

3.3.4 Normalized Difference Vegetation Index (NDVI) and Land Use Land Cover (LULC)

NDVI map displays precise locations of areas with thriving vegetation and areas with little or no vegetation; this is important in flood-risk assessment because vegetation influences water movements, making it a conservational property of flooding, which is useful when assessing plant communities' growth and development [9, 46]. NDVI values generally vary between -1 and + 1. NDVI was computed using Eq. 3:

$$NDVI=\frac{\left(NIR-R\right)}{(NIR+R)}$$
(3)

where R represents Red bands (Band 4 in Landsat 8), and NIR indicates Near Infrared Red bands (band 5 in Landsat 8).

Zero and negative NDVI values show water and barren land, NDVI values between 0.2 to 0.4 show grassland, while > 0.5 represent forest land [26]. NDVI map of Ilorin and its environs (Fig. 6a) showed that the Vegetation index is high in most of the remote parts of the studied area, hence a low likelihood of flooding. Low indices are seen in the built-up parts of Ilorin, where industrializations are taking place, these areas have little to no vegetation within them that can slow down water movement during excess rainfall and hence, have a high flood risk.

Fig. 6
figure 6

a Normalized Difference Vegetation Index variation across Ilorin and its environs, b Land Use and Land Cover types in the study area

Full size image

Another critical factor for consideration in flood-risk assessment is LULC; it controls the rate of infiltration of surface and groundwater on land. Vegetated and open land permit easy flow and infiltration of water, thereby reducing flood risk, while built-up areas do not, increasing flood risk. LULC map of Ilorin and its environs was derived from the area’s Landsat 8 data using the Maximum Likelihood technique in GIS. Equation 4 [25] was used to calculate the output accuracy as follows:

$$\mathrm{Overall\;accuracy\;}\left(\mathrm{OA}\right)=\frac{\mathrm{No}.\mathrm{\;of\;correct\;observations\;}}{\mathrm{Total\;no}.\mathrm{\;of\;observations}}$$
(4)
$$\mathrm{OA}=\frac{71}{80}=88.7\%$$

Observation data were derived from field investigation across the study area. This was done to ascertain the LULC pattern on ground in the study area.

A multivariate technique called kappa (k) analysis uses Khat statistics, which indicate the degree of accuracy, to assess output map accuracy levels. The Kappa coefficient was compiled using Eq. 5 [37, 58].

$$\mathrm{K}=\frac{\mathrm{\%\;overall\;correct\;value\;}-\mathrm{\%\;correct\;agreement\;to\;observed\;values}}{\mathrm{Total\;no}.\mathrm{\;of\;class\;}-\mathrm{\%\;correct\;agreement\;to\;observed\;values}}$$
(5)

An overall accuracy (OA) value of 88.7% suggests a strong association. The calculated Kappa coefficient (K) value was 0.82, which further demonstrates the overall precision of the observed LULC in the research area. Five categories were identified from the LULC map of Ilorin and its environs (Fig. 6a): open/barren land, built-up regions, vegetated land, water bodies, and rocky areas. Open/bare land makes up 52.5% overall, followed by vegetated land (34.4%), rocky terrain (9.8%), built-up regions (3.8%), and water bodies (0.25%). While vegetation-covered and rocky locations are least harmed by floods, they mostly affect built-up regions and bodies of water.

3.3.5 LST (Land Surface Temperature) and SMI (Soil Moisture Index)

LST is one factor rarely used in flood-risk assessment. It gauges the amount of thermal radiation that is emitted from the ground when sun energy interacts with and warms the bare earth or vegetation [21]. LST is a crucial tool in studying the physical processes of the land surface,it provides insight into how LST variation influences hydrological and meteorological systems in an area, hence its application in flood-risk potential assessment. As the average temperature of land increases, more evaporation occurs, leading to an increase in overall precipitation. Urban areas are characterized by high surface temperatures, while water bodies and vegetated lands have low LST. Thermal Infrared Sensor (TIR) band of Landsat 8 was used as an input band together with K1 and K2 values provided in the Landsat 8 metadata. LST was computed in ArcMap using Eq. 6. The Landsat 8 data was acquired in the month of December 2018 when vegetation is less abundant.

$$LST= \frac{BT}{1+w}\times \frac{BT}{p}\times In(e)$$
(6)

BT represents At-Sensor brightness temperature, w represents λ of emitted radiance,

$$p\;\mathrm{is\;calculated\;as\;}h\times \frac{c}{s}(1.438\times {10}^{-2}mK)$$
H:

is the Plank’s constant which is \(6.626\times {10}^{-34}Js\)

S:

represents Boltzmann constant which is \(1.38\times {10}^{-23}J/K\)

C:

represents speed of Light which is \(2.998\times {10}^{8}m/s\)

e:

Land Surface Emissivity

The LST values in the area (Fig. 7a) ranged from 60 °C to 73 °C. While 47% of the whole research area has an LST value between low and very low, approximately 53% of it falls between moderate and very high.

Fig. 7
figure 7

a Land Surface Temperature pattern within Ilorin and its environs, b Soil Moistures Index variations in the study area

Full size image

SMI is another factor rarely used in the flood-risk potential assessment. The ratio of the difference between the current soil moisture and the permanent wilting point to the field capacity and the remaining soil moisture is known as the SMI. This index varies from 0 to 1, where 0 signifies exceptionally dry soil and 1 denotes exceptionally wet soil conditions [49]. The soil's wet condition indicates the soil's ability to hold water during runoff, which helps reduce flooding. In contrast, dry soil conditions show the soil's inability to retain water, thereby contributing to flooding. Figure 7b shows the SMI map of Ilorin and its environs, 74% overall falls in the moderate and low SMI, and 26% falls in the high soil moisture index. SMI was computed using Eq. 7 (after [59].

$$SMI= \frac{(LSTmax-LST)}{(LSTmax-LSTmin)}$$
(7)

where LSTmin is the minimum LST of a given NDVI, and LSTmax is the maximum LST of a given NDVI.

A comparison between SMI and LST shows that areas with high LST have low SMI, while areas with low LST have high SMI; this implies that as surface temperature increases in soils, moisture level reduces because water is lost to the atmosphere during evaporation. Soils with good water retention capacity will retain some excess water during heavy rainfall, thereby limiting the effect of flooding. In contrast, dry soils tend to be soggy and flooded during excess rainfall.

3.4 Determining Criteria Weight

One of the key methods for Multi-Criteria Decision Making’s (MCDM) balancing is AHP, which was developed by Saaty [51]. The MCDM technique helps compare various factors simultaneously via a matrix of pairwise comparison that orders the proportionate influence of one criterion over another. The importance ranking runs from 1 to 9, where 9 denotes the greatest significance and 1 denotes the equal significance. Due to the AHP method's effectiveness in this aspect, it is applicable in various industries globally, including transportation, education, and healthcare [28].

The Analytical Hierarchy Process (AHP) makes it possible to arrange criteria hierarchically and contributes to the extent of logical reasoning supported by a conceptual model [50]. The pairwise comparison matrix (Table 2) was computed based on relative scale of importance. It was further normalized to derive the Criteria Weights (CW) and Weighted Sums (WS). Equation 8 was used to determine the Consistency Index (CI), while Eq. 9 was used to derive the Consistency Ratio (CR).

$$\mathrm{CI}=\frac{{\lambda }_{max}-k}{k-1}$$
(8)
$$\mathrm{CR}=\frac{CI}{RI}$$
(9)

The calculated CR for the study area was 0.09, while the CI was 0.12. A consistency ratio value < 1 implies a logical perception, but when it is > 0.1, the assessment needs to be amended. The AHP readings are consistent because the CR value is within the permitted range.

By using a weighted index overlay to combine all the thematic layers, a flood-risk map was produced (Eq. 10).

$$\mathrm{Flood}-\mathrm{risk\;map\;}\left(\mathrm{FR}\right)={\sum }_{i,j=1}^{8}WiXj$$
(10)

where; \(Wi=\%\) weight thematic map

$$Xj=\mathrm{reclassified\;map}$$

Weight assignment and arrangement for the prepared thematic layers for flood-risk assessment are given in Table 3. The ten (10) flood causative factors considered in the study were reclassified into 5 groups, where 1 represents low flood-risk and 9 represents high flood-risk. The thematic maps were rescaled and reclassified into equal spatial resolution for better accuracy. Weight assignment and classifications were based on expert opinions, literature study, and previous works on similar methodology. Processing was done in a GIS environment. The layer values were classified into five categories based on influence on flood-risk which varies from very low to very high. The area covered by each category is also given in percent and kilometers.

Table 3 Weight assessment and arrangement for flood-risk mapping
Full size table

4 Results and Discussion

4.1 Flood-Risk Zones Mapping

The primary motive for flood-risk assessment is to produce a map showing the various areas with high and low flood risks within an area. The ten flood factors were merged via Weighted Overlay Index in ArcMAP to produce the flood-risk map of Ilorin and its environs. The output flood-risk map (Fig. 8) showed the flood-risk zonation pattern in Ilorin area. The flood-risk map was categorized into three viz-a-viz low, moderate, and high flood-risk zones. Areas with flood-risk index of 1 to 2 are classified as low, while 3 and 4 to 5 are classified as moderate and high, respectively. Table 4 shows the areal coverage of each zone. It can be observed that 12.49% of the total landmass under investigation lies in the high flood-risk zone, while 54.69% and 32.82% sit in the low and moderate flood-risk zones, respectively. The majority of the area of study is located within the moderate and low flood-risks zones. The regions with a high danger of flooding are those that are near large streams, waterways, and dams.

Fig. 8
figure 8

Flood-risk map of Ilorin and its environs

Full size image
Table 4 Flood-risk level with areal extent
Full size table

During times of heavy rainfall or flooding, communities located in high flood-risk zones are expected to suffer significantly. Additionally, places with low and extremely low flood risks have a lower likelihood of experiencing flooding. When determining flood risk, elevation and distance from streams are important considerations. High flood-risk zones as indicated on the flood-risk map are areas with low elevation and closer to major streams. Further, drainage density values are also high in high flood-risk areas. High LST and SMI values also contributed to the increased flooding in these areas. Low flood-risk areas are mainly within the high-elevation regions and farther away from major streams; drainage density values are also low in such areas.

4.2 Validation of Flood Vulnerable Zones Map

Sixty (60) points were collected from areas of reported flood and no flood occurrences within the studied part of the Ilorin and environs to validate the flood-risk map produced. To determine the level of accuracy of the findings, these sites were overlaid on the flood-risk map generated. Figure 9 shows the plots of the reported flood cases across Ilorin on the flood-risk map of Ilorin and its environs. It can be observed that the majority of the reported incidences, which span the years 2011 to 2021, mainly fell within moderate and high flood-risk zones on the map; this confirms the validity of the findings of this research. Further, points collected from flood affected areas and areas of little or no flooding events were computed to access the overall accuracy of the output Flood-Risk map. The error matrix details are given in Table 5; the Overall Accuracy (OA) and Kappa coefficient (K) value were derived as 86.7% and 81%, respectively.

Fig. 9
figure 9

Validation points on Flood-Risk map of Ilorin and its environs

Full size image
Table 5 Error matrix of flood-risk map in Ilorin and environs
Full size table

5 Discussion

Identification of flood-vulnerable zones using GIS and RS techniques has significantly contributed to flooding mitigation strategies in recent years. From the results of this research, the ten flood causative factors all played different roles in the intensity and frequency of flooding that has continued to ravage Ilorin and its environs. Table 3 revealed that 33.5% and 19.29% of the study area are in moderate and high flood-risk zones, respectively, while 47.2% of the region is in a low flood-risk zone. The studied part of Ilorin lies on a relatively high-elevation terrain, which is why most places away from the major streams have a low flood risk. Most areas affected across the states are areas close to major streams; a heavy downpour of rainfall brings about overflow of these streams flowing through culverts and drainage channels. This overflow makes its way into nearby residents, especially houses within a 500 m radius of the major streams. Elevation and slope angles affect the rate of movement of excess water from the stream channels. The low elevation along the river channels and low slope angles makes it easy for water to find their ways to residential areas as water moves from high heights too low land areas. Vegetation indices, soil moisture index, and land surface temperatures are factors that have moderate effects on the control of flooding. In contrast, topographic wetness and stream power indices have minimal control over flooding in the area. Several dams within the study area have been reported to be a major contributing factor to flooding within Ilorin and its environs [42]. Asa dam is located within Ilorin and overflows from the dam during excess water control floods nearby residents. Alimi et al. [9] have previously recommended GIS as a practical tool for flood-risk mapping, their study also showed that distance to streams and elevation are highly influential flood causative factors, this is similar to the results derived in this present study. Oriola and Chibuike [43] also reported in part of Kwara state, that distance to major drainage channels affect flooding within the state. Other factors such as precipitation and socioeconomic status were also recommended by Hussain et al. [23]. In the present study, elevation, distance to stream, and drainage density, as seen in Table 3, are the most influential flood-risk factors in the study area.

The ability to visualize flood-risk zones makes mitigation strategies and planning an easy task. The importance of a flood-risk map in conversations geared towards ameliorating flooding incidences cannot be overemphasized. In the present study area, there is a pressing call for genuine land use control strategies. Land acquisition in high flood-risk zones needs to be controlled; this will cushion the yearly loss of life and properties that arise during flooding events. Further, the construction of river embarkments along stream channels can prevent water overflow to residential areas during periods of excessive rainfall. The embarkments will control the movement of water and slow the rate of excessive spreads into the residential areas. Also, there is an urgent need for flood warning systems that alert people to the likelihood of a flood in the future; this will enable proper preparations and prevents sudden loss of life and properties.

6 Conclusions and Recommendation

The seasonal reoccurrence of flooding within Ilorin, the Kwara state capital, mandated the urgent need for flood-risk assessments in some parts of the state. Flood-risk potential assessment is a useful pre-disaster management technique, especially when assessing the risk of flooding in an area against all flooding mechanisms, such as groundwater, pluvial, fluvial, and tidal. It also involves flood mitigation measures. Most parts of Ilorin and its environs are within the low flood-risk zones, and areas of high flood occurrence are extremely close to major river channels that tend to overflow during periods of heavy downpours. This could have been averted by preventing land acquisition near the rivers. GIS and remote sensing techniques offer cheap and user-friendly practical flood-risk mapping tools, which can be used to map flood-risk zones quickly. This is effective in this study. The MCDA technique might go a long way in providing valuable information for effective flood mitigation controls. To avoid occupying high flood-risk lands, It is strongly advised that locals be actively discouraged from erecting homes close to important waterways. Additionally, the government needs to install proper flood warning systems and flood-resistant buildings throughout flood-prone locations.