There are two sources of surface data used to describe surface characteristics and for comparison. The NASA POWER data shown in Figure 2(a) and the in-situ NiMET data (Figure 2(b)). Both Figures represent plots of wind speed and direction (top), relative and specific humidity (middle), and temperature and surface pressure (bottom). On 26^th May 2019, Owu-Kuta in Osun State Nigeria experienced a cold pool wind gust. As indicated by the continuous vertical red line, there is a sharp drop in the wind direction for more than 1-hr as reflected in both Figure 2(a) and Figure 2(b). The temperature, at this time, was already decreasing and the pressure was rising (increasing) with corresponding increasing wind speed. The relative humidity was increasing until close to saturation (100%) during the same period. The above environment established the presence of cold pools based on [31]. The maximum wind gust was around 2 m/s on the average from the two different sources (NASA POWER and NiMET) during the period of the cold pool.

On 19^th March 2020, the city of Akure Ondo State Nigeria experienced cold pool wind gust during the evening hours, which lead to destruction of properties. Figure 3(a) and Figure 3(b) showed a sudden drop in the wind direction (top), which lasted more than 1-hr. the temperature was decreasing also during this period, whereas, the surface pressure was increasing, based on the NASA POWER observation, but surface pressure was constant as observed by the in-situ NiMET observation. The relative humidity was increasing and reached near saturation (100%) during the same period. The wind speed was fluctuating and exceeds 3 m/s on the average from the two different sources. As in the first case event, parameters that described the Akure cold pool wind gust showed

conformity with [31]. In each case event, the black dashed lines indicated the boundary/transition into another day.

The cloud parameters analysed are the Brightness Temperature (Tb), Rainfall amount, Lightning and thunderstorm amount. The May 26^th 2019 cold pool event was observed to have great impact in a community in Osun State of Nigeria. Owu-Kuta is situated at Latitude 7.65˚ and Longitude 4.28˚ Osun State Southwestern Nigeria. The NOAA brightness temperature (Figure 4) clearly indicated the presence of cold cloud (core < 200 K) in the vicinity where the cold pool was observed. The core of the cloud shield was not centered over the impacted

community at the time of the satellite snapshot (21Z) on 26^th May 2019, however, clouds over West Africa are known to migrate westward [33] [34] [35], therefore it is expected that the cloud would have migrated over the impacted community few minutes/hours later.

The cloud shield spread over several kilometers across some states in southwestern Nigeria of which the peak 3-hr accumulated rainfall amount (~60 mm), as shown in Figure 5(a), was observed over Akure (Latitude 7.23, Longitude 5.21) and environs in Ondo State. The system, which is already dissipating at the time it was capture, indicate that it had initiated in the mid-afternoon [36]. The satellite imagery (Figure 5(b)) of the CMORPH rainfall indicated that the system had migrated slightly westward 3-hr later and accumulated rainfall amount peaked at 60 mm∙3hr⁻¹ during 21:00 - 00:00 UTC, except for a location north of 8˚ Latitude.

In Figure 6(a) and Figure 6(c), representing the case event of 26^th May 2019, the in-situ NIMET observed (locally observed) rainfall amount and number of thunderstorms present indicated that the hourly rainfall peaks at 15 Local Time (LT), that is, 16 UTC in Figure 6(a), and three thunderstorms were observed to be present at exactly the same time in the impacted community, as shown in Figure 6(c). The WWLLN stroke density discussed in the next paragraph, and shown in Figure 7(a), does not depict actual number of thunderstorms, but presents patterns and possible location/presence of thunderstorms. Separate thunderstorms can be determined using clustering algorithm to cluster strokes in time and space [37], which is beyond the scope of this work. Thus, this work has not compared the number of locally (NIMET) observed thunderstorms with the WWLLN observations of stroke density but presented the results separately. Although, Figure 7(a) indicated relatively few possible presences (about two or three) of thunderstorms, similar to the locally (NIMET) observed number of thunderstorms, this cannot be ascertained since clustering algorithm was not

performed on the strokes. The rainfall amount had dropped from its peak value of 17.5 mm to about 5 mm at mid-night indicated by the dashed black vertical line. The thunderstorms present had also reduced to only one, meaning that two thunderstorms had decayed during the pass hours. In Figure 6(b) and Figure 6(d), representing the case event of 19^th March 2020, the in-situ NIMET observed (locally observed) rainfall amount and number of thunderstorms present indicated that the hourly rainfall has it highest value at 17 Local Time (LT), that is 18 UTC in Figure 6(b), whereas there are three thunderstorms present at 15 (LT, synoptic hour) as shown in Figure 6(d). It was observed that two thunderstorms had decayed as at mid-night, the remaining one was not producing rainfall at mid-night. Although there were clusters of thunderstorms, as shown by the WWLLN observation (around ten localized and separated stroke density maxima, which could mean possible isolated and separated thunderstorms) in Figure 9(a), again this number could not be ascertained since clustering algorithm was not performed on the strokes to determine the actual number of separate thunderstorms from WWLLN observation. The local (NIMET) observations were recorded on synoptic hours whereas the WWLLN observations represents the whole day for both case events.

Lightning observations from WWLLN showed distributions, with 0.5˚ × 0.5˚ resolution, of stroke density (strokes∙km⁻²∙day⁻¹: since the plot is observation for one day) in locations where lightning were observed. Figure 7(a) showed stroke density of about 400 stroke∙km⁻²∙day⁻¹ around or in close proximity to Owu-Kuta as indicated by the arrow. This also justified the presence of thunderstorm around or in close proximity to the impacted area (Owu-Kuta). There were observed higher stroke densities over the Land-Ocean boundaries (Figure 7(a)). The diurnal

cycle, in Figure 7(b), showed peak activities and hence highest stroke counts at around 14 - 16 UTC (daytime), whereas the minimum lightning activities were observed at night and morning hours. This is consistent with [23], based on TRMM and WWLLN data, that maximum diurnal lightning activities over Benin peaks around 18:00 UTC and minimum frequencies are observed during morning hours (10:00 - 11:00 UTC). Balogun et al. [38] justified more stratiform precipitation (precipitation with little or no flashes) during night and morning hours and more convective storms (thunderstorms and deep convective systems) during afternoon and evening hours over West African sub-regions. The highest observed occurrence of lightning was about 1750 stroke counts during the 14:00 UTC. About 1600 and 1700 stroke counts were observed during the 15:00 and 16:00 UTCs, which represented the highest during that particular day.

Results for the case event on 19^th March 2020, which occurred in Akure the Ondo State capital, indicated the presence of cloud (Tb < 200 K and substantial rainfall amount observed) over Akure (Latitude 7.23, Longitude 5.21). Satellite observations, using geostationary weather images, are known to be appropriate for identifying thunderstorms and hence MCSs [39] [40]. The core of the cloud shield, indicated by the brightness temperature in Figure 8(a), is centered over Akure with value as low as 170 K at 18 UTC on 19^th March 2020. The system was also producing rainfall, as shown in Figure 8(b), with 3-hr accumulated rainfall amount of 20 mm between 15 - 18 UTC on same day over the impacted location (Akure).

The WWLLN lightning observations showed distributions over Akure, with 0.1˚ × 0.1˚ resolution, of stroke density (strokes km⁻² day⁻¹: since the plot is observation for one day) in locations where lightning were observed. In Figure 9(a),

the stroke density is above 50 strokes∙km⁻²∙day⁻¹ in the vicinity of the impacted area (Akure) indicated by the arrow. The stroke density resolution used for Akure (0.1˚ × 0.1˚) is different from the case event over Owu-Kuta (0.5˚ × 0.5˚). Another location, left of Akure city, within or outskirt of Akure indicated stronger stroke density, which could also have strong influence over major Akure communities. These are scattered thunderstorms in and around the impacted area (Akure). During this particular day, there were no observed stroke densities over the Land-Ocean boundaries (Figure 9(a)), unlike in May 26^th 2019. In Figure 9(b), the diurnal cycle showed peak activities and hence highest stroke counts at around 17 UTC (evening time), whereas the minimum lightning activities were observed at night and morning hours, as was observed for the Owu-Kuta case event. This result is also consistent with [23] and [38] as discussed for the Owu-Kuta case event. The highest observed occurrence of lightning was about 1000 stroke counts during the 17:00 UTC, whereas, the morning hours recorded little or no stroke counts up until about 13:00 UTC, after which the strokes begin to increase.

Local communities experience disasters most acutely [41]. The local impacts of the cold pool wind gust are presented in Figure 10 for the case event of 26^th May 2019, which occurred in Owu-Kuta Osun State Nigeria. The wind gust broke down mounted structures, which fell on parked vehicles (Figure 10: left wing), and it also opened the roof of the Oba’s palace in Owu-Kuta (Figure 10: right wing). The Hope newspaper [42] gathered that about sixteen buildings, besides the palace of Oba Adekunle Oyelude Makama, the Olowu (king/paramount ruler) of Owu Kuta, had their roofs fully blown off by the windstorm. The windstorm blew off a part of the roof of the Oba’s palace and some other properties, such as furniture, where damaged inside the palace. The blown-off roof fell on the traditional ruler’s car, which resulted in some damages on the car. These observed and usually high impacts on communities, particularly in this part of the world, are caused by poor (little or zero) mitigation approach or complete absence of disaster risk reduction strategy [43].

Another source, Nigerian Tribune [44] gathered that antecedent of the destructive rainfall occurred in the atmosphere over the impacted community few minutes earlier when an unusually dark cloud, presumably a Nimbostratus cloud, suddenly occupy the sky at about 3:30 pm, which caused anxiety among dwellers. While the residents were hoping for a moderate rain that would suppress the hot temperature on their bodies, what they experienced was a destructive windstorm, violently blowing-off roofs of buildings. Other infrastructure affected was electricity poles, and the windstorm uprooted trees, of which some explained the event as spiritual.

However, the traditional ruler dismissed the argument that the windstorm was caused by spiritual factor or by certain forces. He stated that it was a natural disaster, on the ground that the storm affected two nearby communities like Ikoyi and Iwo. According to him, a public school’s roof was blown off in Ikoyi and in Iwo, a well-known place known as Odori was strongly hit by the windstorm. He further said, “Just take a look at this place well and tell me the number of trees standing here. This is an open place which needs trees to shield the buildings from heavy winds as we just experienced. The world has developed beyond attributing every bad occurrence to the gods or anger by the ancestors. If the ancestors have the opportunity of returning to the world, they would not want to do things the way they did it in their first coming. What I am saying is that an incident like this is simply a wake-up call and a reminder that we should endeavor to plant trees. This deforestation is counter-productive despite the fact that it is evidence of development. But when we cut trees in the forest we should always endeavor to plant more”.

Gencer [45] wrote that increasing density of human population, deforestation, expansion of slums, blocking of natural derange, soil erosion, among others frequently increased the risk of disasters. Others responsible factors include inadequate governance, inappropriate use of resources, and poor preparedness. According to [46], increasing environmental degradation, risk of geological events and extreme weather as a result of growing population encourage vulnerability and reduced resilience to risk disasters.

The traditional ruler used the opportunity to call on the Nigerian Meteorological Agency (NIMET) to learn from developed nations where natural disasters are forecasted before they occur, and precautionary measures taken to prevent it. “NIMET is saddled with that responsibility to guide us on the behaviour of the weather. We all know that the entire world is experiencing climatic change. That is more reason the agency should always update us so that we prepare before calamity comes”, he said.

The second event investigated occurred in Akure City and impacts are presented in Figure 11. The windstorm rendered some residents homeless, some injured, and some dead. The Hope Newspaper [47] recounted some of the damages as follows; an eye witness explained that the windstorm carried a container and the owner inside, which resulted in several injuries on the fellow inside the container. Also, the wind lifted a car slightly above the ground from the road, which made the passengers alighted and took to their heels in the heavy downpour. Dwellers in Abusoro/OluFoam, Onigari area of Shagari village, and Igoba communities in Akure had their electricity poles broken and pulled down. A forty year old electrician whose building almost collapsed to the unfortunate incidence, informed The Hope Newspaper in Olufoam community that the wind removed the roof of his building and destroyed the roof of another house in the impacted community, and the storm later pulled down almost all his entire building. An inhabitant in the impacted community narrated that his three children left the house to stay with a neighbor as a result of fear due to the intensity

of the wind, which later wreak havoc on his building. As suggested by [48] and [49], most climate disasters can be mitigated (reduced) by community-based collective response and building resistance in other to cope with the event at the time and location it occurred.

Kayode Jide, a bricklayer with four children, whose house was pulled down to its foundation. In his words; “I was in town on that day but the rain which fell in town was not destructive. I was surprised when I received a call that my house had been pulled down by the wind and my children were inside the house. I thank God because He used my neighbours to rescue the children from the scene. The wind was just too much. I have never seen that kind of wind before in my entire life”.

According to [50], the people most affected by windstorms/rainstorms are the poor (low-income earners). For example, a 63-year-old widow recounted her losses while speaking with The Hope Newspaper. In her words; “I trade in garri (cassava flour) at Odokoyi market and I built this 10 bedroom face me I face you apartment after the demise of my husband some five years ago. I was in the house with six other people including my daughter-in-law who had a six-month-old baby with her. It was raining and suddenly, the wind began to operate. We started praying to God for mercy and the next thing we saw was a block that fell right in our midst in the room before the whole ceiling fell on us. Nobody was injured and we all escaped through the window before most parts of the house were pulled down. All our roofs which were removed during the rain spread across the road preventing vehicular movement”. The finding by [51] indicated that natural disasters often increase poverty due to human and material losses.