The study area, Asemoku, is in Ndokwa East Local Government Area (LGA) of Delta State. Ndokwa East is situated in the Eastern part of Delta State, Nigeria. It is bordered to the North by Aniocha south and Oshimili South LGAs respectively. The area is bordered in the East by River Niger and to the South by Isoko south and to the West Isoko North and Ndokwa West as shown in Figure 1. The Nigeria Agip Oil Company (NAOC) pipeline under study, traverses various communities. This study area lies between latitudes 5˚55' and 5˚69' North and longitudes 6˚40' and 6˚56' East. The spill point is Long. X_ 6.5616111, Lat.Y_5.6505833.

The area lies on the tropical monsoon climate, which is characterised by the rainy and dry seasons. The annual rainfall ranges between 1895 mm and 2105 mm annually. The annual mean temperature ranges between 27˚C and 30˚C respectively. In its original state, the climate supports agriculture and fishing economic activities [39] . The discovery of oil and the haphazard nature of its mining has resulted in serious migration and change of economic activities by the locals, due to poor productivity of land resources—a consequences of years of spills with no corresponding clean-up [40] .

This research used primary and secondary data (Table 1). The data was provided through government sources and databases from other organizations. The raw spatial data and satellite images used in the research came from the United States Geological Surveys (USGS), Google Earth Pro, Oil Spill Incident data from the National Oil Spill Detection and Response Agency (NOSDRA). Published oil spill records (https://oilspillmonitor.ng/) NOSDRA is a Nigerian Government Agency tasked with capturing all oil spill incidents both on marineand terrestrial ecosystems across the country [24] .

Based on fieldwork data collected coupled with images downloaded from Google Earth with geo-referencing on ArcGIS 10.8, a classification pattern was developed to enhance the classification of the land-use and land-cover inland habitat in the study area [41] . The system of classification is shown in Table 2.

Source: Authors compilation (2023).

The supervised classification (SC) was deployed for the classification mechanics of this study [42] . Supervised classification entails the imagery user using developed skill to deploy spectral known categories, i.e rural, industrial, or forest, and thereafter manipulates the software by assigning pixels of the image to landcover which matches the afore classifications. This technique has been termed the most frequently used [43] , and has been deployed by [44] . However, the SC was deployed after the demarcated for the study was determined (training-class TC). Three training classes for used for a single category. This was done in agreement with the imagery. To define the TC the area for area land cover class (LCC) was determined [45] . Thereafter, extraction of signatures was done (ES) in the ERDAS imaging V16.6 environment. The SC was applied after the TCs were determined. Two or more TCs were deployed for a single class. The non-parametric (NP) technique was deployed for the classification. The image after these was classified into natural vegetation, farmland, water body, wetland, Build-up areas and bare surfaces (Table 2).

The main criteria considered to establish the degree of sensitivity to oil spill and other stress factor of an ecological class include its biological productivity, oil/ecology interaction, ease of clean up, social, economic and human importance [46] . Fasona et al. [47] adopted similar criteria in their study (Table 3).

It is needful that after classification, accuracy of such classification is validated. This is a quantitatively assesses the efficiency of the sampled pixels and how the match reality. The matching of the classed data and the ground truth (GT) are shown in Table 4. Herein, the validation statistics deployed for ESI sensitivity validation was the Kappa statistic using the following equation:

Sensitivity = a e a e + b e

Specificity = d e b e + d e

errorcommission = 1 − s p e c i f i c i t y

Omission = 1 − s e n s i t i v i t y

+ predictivecapacity = a e a e + b e ( u s e r a c c u r a c y e q u i v a l e n )

− predictivecapacity = a e c e + d e

where:

ae = agreement between classification and observed values

be = frequency of X not observed to be X

ce = frequency of times X classified was X observed

de = frequency of times X was classified and not observed.

Total points = N = (ae + be+ ce + de).

Table 4. The buffered zones of the land use area.

The KAPPA analysis represents a multivariate statistics deployable when the test for accuracy is the need [48] . See equation bellow:

K = N ∑ i = 1 r x i i − ∑ i = 1 r ( x i + X x + 1 ) N 2 − ∑ i = 1 r ( x i + X x + 1 )

where;

r = total errors that exist is the rows cum columns of the matrix,

N = all observations

X_ii = observation in row cum column i

X_i+ = borderline total of row i, and X + i = marginal total of column i

Generally, where a Kappa output is at 1, this represents total agreement, and the further it is closer to zero, it means disagreement.

The buffering operation helps in knowing the proximity of resources that are vulnerable or sensitive to oil spill [48] . It is also used to assess the hazard areas along a risk zone (i.e. oil pipeline) as it affects the land-use and the land-cover. Buffer distances of 50 m, 100 m, 200 m, 300 m and 400 m (Table 3) were adopted; as used by Onosemuode et al. [46] .

Emergency response zone is a location strategically positioned at area considered to be easily accessible, between the area where the inland habitat features are likely to suffer great harm and where responders and equipment can easily be deployed within the shortest of time after oil spill incident has been reported [49] . The emergency response zones along the study area were chosen by considering the following factors a) the most delicate inland habitat feature and b) the proximity and accessibility of required responders and equipment deployment along the pipeline route (see [48] ).

The results of the image processing and ecological classification of the Landuse/cover buffer standards of 100 m, 200 m, 300 m and 400 m respectively, was used for the establishment of the various landuse/cover, that at risk of being affected in the event of oil spill in each buffered zone. The 100 m buffered zone is the off-set of 100 m on each side of the pipe line spill point as shown in colour red in Figure 2. The 200 m buffer zone is the 200 m off-set of the pipe line spill point is as shown in colour dark blue in Figure 2. The 300 m buffer zone is the 300 m off-set of the pipeline spill point and it is shown in Purple colour and the 400 m buffer zone is 400 m off set of the pipeline spill point is shown in light blue colour. The various buffered zones helped in determining the spread of the spill and how it affects the various classified landuse/cover as shown in Table 2.

The various landuse/cover identified in the study and their ranking and classification are discussed below (see Table 5).

The natural vegetation is the landuse that occupied the most and largest area of land use/cover within the buffer zones. It comprises of grassland, shrubs, rain forest etc. It is also home tofauna species like rodent, Rabbits, squirrel and grass-cutter, snakes etc. [50] . Natural vegetation occupies a total land area of 38.641 hectares (71.99%), 90.704 hectares (73.75%), 135.979 hectares (73.96%), and 180.213 hectares (73.76%) within the 100 m, 200 m, 300 m and 400 m buffer zones respectively (Table 4 and Figure 3).

Farmlands are areas where agricultural activities are carried out with the aim of producing different crops for personal consumption and also as a source of income [51] . The farmland area occupies a total land area of 2.469 hectares (4.60%), 6.494 hectares (5.28%), 136.979 hectares (73.96%), and 180.213 hectares (73.76%) within the 100m, 200m, 300m and 400m buffer zones respectively (Table 4 and Figure 4).

The water body within the buffer zones comprise of rivers, streams, ponds and creeks [7] . The water body serves as a source of drinking water, bathing water and for other domestic purposes for some people in the Asemoku community. It also serves as habitat to aquatic plants and animals of various diversities like water lilies, water hyacinths, frog, toads, shrimps, turtles, fishes, and some reptiles [52] . There is water body within the 100m buffer zone. It is also recorded that water bodies occupy a total land area of 0.491 hectares (0.91%), 1.315 hectares (1.07%), and 2.613 hectares (1.42%) and 4.210 hectares (1.72%), within the 100 m, 200 m, 300 m and 400 m buffer zones respectively (Table 4 and Figure 5).

The wetland areas comprise of ponds, marshes, forested freshwater, wet grassland and swamps [53] which is dominant in the study area. Wetland has a very rich unique biodiversity of flora and fauna species. Wetlands support populations of fish, amphibians, reptiles, birds, and animals, with many species reliant upon wetlands for their reproduction and early life stages when they are most sensitive to oil [54] . Migratory water-birds depend heavily on wetlands as is the

case in the study area. The wetland area occupies a total land area of 9.396 hectares (17.51%), 11.969 hectares (9.73%), 13.974hectares (7.601%), and 16.637 hectares (6.81%) within the 100 m, 200 m, 300 m and 400 m buffer zones respectively (Table 4 and Figure 6).

The built up areas comprises mainly of residential, utility, commercial, religious and educational structures [55] . This area comprises of diverse floras like orange, mango trees, coconut and palm trees, cocoyam, maize water leaf, bitter leaf, scent leaf plants and shrubs [56] , which serve as food options for the locals. The faunas consist mainly of domestic animals such as dog, goat, fowls, rat, wall gecko, lizard, frog, cats, insects and microbes that may not visible. The built up area occupies a total land area of 6.494 hectares (5.28%), 13.770hectares (7.49%), and 23.205 hectares (9.49%) within the 200 m, 300 m and 400 m buffer zones respectively (Table 4 and Figure 7).

Bare surfaces are exposed surfaces which can be attributed to natural processes and human activities [54] . It hardly supports plants growth because of the limited nutrients in it [57] . The bare surfaces occupy a total land area of 2.677 hectares (4.98%), 12.317 hectares (10.01%), 15.348 hectares (8.34%), and 16.130 hectares (6.60%) within the 100 m, 200 m, 300 m and 400 m buffer zones respectively as (Table 4 and Figure 8).

The Emergency Response Zone (ERZ) was proposed to be strategically positioned in the area where the inland habitat and Landuse/cover is likely to suffer great harm in the event of an oil spill and also where responders and equipment such as hard booms, skimmers, storages, fire extinguishers and vehicles can easily be deployed within shortest of time; following Onosemuode et al. [46] . The emergency response zone (ERZ) in the study area was proposed to be situated at Asemoku community. The site chosen was within 50 m buffer zone (Figure 9). This area was chosen, due to its proximity to the pipeline. Another reason for siting the ERZ in Asemoku community is its accessible road network to the pipeline [48] . This will enable response team deploy response equipment with ease in case of incident of spill along the pipeline [58] .