Assessment of the sustainability of landcovers due to artisanal mining in Jos area, Nigeria

Environmental sustainability ensures that minerals are responsibly exploited to meet the present needs without depriving the future generations the ability to meet their needs. Unfortunately, environmental sustainability and artisanal mining seem unattainable in recent years with aggressive mining operations. It is on this premise that geospatial techniques with proven role in environmental studies were employed to assess the environmental sustainability due to artisanal mining in Jos area, Plateau State, Nigeria. Land surface temperature (LST) and seven geospatial indices used for land use land cover (LULC) estimation were generated. The mean LST values showed a steady increasing pattern from 23.98 to 25.88 °C and 29.46 °C in 1984, 2002, and 2020 respectively, as a result of exposed outcrops occasioned by mining and the expansion of mining communities. The mean value of the Normalized Difference Vegetation Index (NDVI) depicts a considerable increase from 0.179 in 1984 to 0.458 in 2002 and a slight decline to 0.438 in 2020. This is congruent with the mean Optimized Soil-Adjusted Vegetation Index (OSAVI) values, thus revealed that the Jos area is not densely vegetated implying that the course to revegetate the region has not been achieved to the tune of woodland populated vegetation. The resulting maps from Modified Normalized Difference Water Index (MNDWI) revealed a consistent decline in the mean values − 0.349, − 0.391, and − 0.411 in 1984, 2002, and 2020 respectively. Mineral recovery, mineral processing, and seasonal variations could be some of the reasons waterbodies are one of the most stressed natural resources in the study area. The mean values of Enhanced Built-up and Bareness Index (EBBI) decreased from 0.282 in 1984 to 0.202 in 2002, but increased from 0.202 in 2002 to 0.230 in 2020, which corroborated with the findings of NDBI. It was deduced that built-up areas in the study area are very low. This may be attributed to urban migration and the migration of artisanal miners to new mining sites across the state or country. The values of bare land mapped using Dry Bare Soil Index (DBSI) corresponds with that of Normalized Difference Bareness Index (NDBaI) and showed that bare land has reduced on the Jos Plateau due to improved vegetation growth. This study provided essential input and referential information for proper decision making on environmental sustenance, environmental management, and mineral resource conservation.


Introduction
The continuous surge in the earth population, increased urbanization, technological advancements, and industrialization have increased the global demand for minerals. Minerals are crucial for promoting economic development and sustaining the level of living and the functionality of today's society. This evidently shows that the mineral industry is fundamental in the world's modernization, notwithstanding palpable balance must be achieved between environmental sustainability and the prevailing economic needs of the modern world. The mineral industry encompasses a diverse set of enterprises, ranging from artisanal, low-tech operations to large-scale global firms using advanced technologies to obtain and recover minerals (Litvinenko 2020).

Responsible Editor: Philippe Garrigues
Mining, which entails obtaining minerals from the earth's surface in an environmentally acceptable manner is recognized as a key anthropogenic activity that has considerably contributed to high levels of land degradation (Davou 2013;Owolabi 2013Owolabi , 2020Bubnova 2018;Adewumi and Laniyan 2020). According to Adekeye (2010), mining in Nigeria, until this century was characterized by small-scale activities that were usually rudimentary in terms of technology and harmful to the health and safety of human beings while still providing society with the required minerals.
Minerals are exploited by artisanal miners using only the most basic tools on readily available deposits. This primitive method of mining is irrefutably the foremost mining method employed in Nigeria despite the abundant proven mineral deposits. Mallo (2012) pointed out that artisanal and small-scale mining of gold, tin or columbite, lead or zinc, coal, etc., began in Nigeria when colonial mining was introduced around 1902 with over 95% of mining activities in Nigeria being artisanal and another 95% of these being illegal. On the Jos Plateau, artisanal mining is currently taking place on abandoned low-grade 0.1 g/t deposits of minerals such as cassiterite (tin) and columbite (Mallo and Aluwong 2012). According to Adegboye (2012), tin mining has been responsible for significant changes in the Plateau's terrain and social economic structure.
The extraction of minerals from the soil necessitates and induces the removal of land covers, which host communities depend upon for livelihood. Lands which were previously potential for grazing, gardens, natural forest, agriculture, natural herbs, and many other uses have been used for mining activities for over a century since mining commenced on the Plateau. Potential environmental hazards that accompany mining are not given utmost significance and urgency due to instant focus on short-term economic benefits, consequently ruining the natural environment extensively. The methods, tools, and products of artisanal mining are peculiar in nature when compared to other types of mining due to their simplicity and crudeness.
Artisanal mining approaches in Jos area have been highlighted by Ahmed and Oruonye (2016), Aluwong et al. (2017), Owolabi (2017), and Melodi and Mary (2020) to include open cast mining (which requires the utilization of basic equipment such as diggers, hoes, and hand shovels), loto (vertical), underground mining, and panning. Artisanal miners are locked in a cycle of primitive activities that have both environmental and health consequences attributable to poverty, lack of education, and expertise (Stocklin-Weinberg et al. 2019).
Land cover changes, resulting from artisanal mining activities, have serious consequences on the local environment and thus need continuous monitoring to ensure environmental sustainability (Owolabi et al. 2021). Modern geospatial techniques like remote sensing (RS) and geographic information system (GIS) are peculiar geospatial technologies that provide accurate and timely geospatial information to assess the past and present environmental condition of land covers, which has thus been addressed by few studies in Plateau State, Nigeria. There is a dearth of studies and consequently salient data on the diverse impacts artisanal mining operations exert on surface temperature, vegetation, water, built-up, and bare land of the study area, especially from a geospatial perspective, which provides more extensive coverage of large areas than conventional methods and has a proven role in environmental studies. The use of conventional or traditional method to assess the environmental impacts artisanal mining exerts on the natural world is time-consuming and not cost-effective. The limited coverage it provides, human disturbance, and the country's current security vulnerability make a much-needed alternative indispensable. Bubnova (2018) claimed that remotely sensed data is currently nearly the only means of collecting objective and credible information about the state of mining landscapes in large areas. Jos South and Barkin Ladi Local Government Areas of Plateau State are extensively mined terrain and need updated and continuous monitoring of the terrain dynamics for sustainability of the environment. This research work is therefore crucial and imperative to employ geospatial technique to assess the effects of artisanal mining activities on the land covers of the study area, ascertain the trend in changes in the land cover, and also evaluate the relationship between the surface temperature and the vegetation, water, built-up, and bare land of the study area.

Environmental sustainability and artisanal mining
Although the mining industry clearly plays an important role in global modernisation, a balance must be reached between environmental sustainability and the modern world's socioeconomic necessities. At local, regional, and global levels, the activities of the sector are identified to be detrimental to public health and safety, as well as causing damage to the ecosystem (Suh et al. 2017). This contributes to the industry's poor image. Morelli (2011) described environmental sustainability as addressing the necessities of existing and future generations without threatening the wellbeing of the ecosystems by which they are derived. Environmental sustainability ensures that the minerals are responsibly exploited to meet the country's current needs without endangering the potential needs of future generations. Artisanal mining activities and environmental sustenance in Nigeria seem unachievable with intensified conflicts in recent years. Mallo and Wazoh (2014) observed that all mining endeavors have significant environmental consequences. Regardless of how vital mining is for meeting a broad range of necessities, it is generally acknowledged that mining activities have resulted in vast volumes of hazardous wastes that may include potential injurious elements of high concentrations (Del Rio-Salas et al. 2019). Mine sites are abandoned when the grade and quantity of minerals that can be safely and economically extracted have been reached because minerals are finite and exhaustible resources. However, the effects of mining activities are usually irreversible and last long after the mine is closed.

Effects of mining on the environment
For more than a century since mining began on the Plateau, mined out areas have suffered environmental degradation considerably because of mining activities (Mallo and Wazoh 2014). Burrow pits are frequently left unreclaimed, resulting in water filling them and creating dangerous habitats for reptiles that can harm humans (Ako et al. 2014). Burrow pits, pools, and ponds created as a result of artisanal mining of minerals have a tremendous effect on both man and his environs. Mining ponds lead to water accumulation in pits, thus becoming good breeding grounds for pathogens.
According to Sun et al. (2020), among the most difficult hurdles in the global mineral industry is effectively dealing with mine waste. However, tailings from some mineral processing operations are emptied into rivers, polluting and making them inaccessible to the community who rely on them for their livelihood. Ideriah and Abere (2017) discovered that amended soil system was employed around Jos Plateau because the soils physicochemical properties revealed acidity and certain deficiencies needed for crop growth and nutrition. Aigbedion (2005) reported that the people who are suspected to have used monazite (an important source of thorium and cerium) soil for house construction have died tragically and mysteriously.

Mitigation methods in mineral extraction
The key causal environmental effects of mineral extraction and mineral processing operations are connected to soil quality, water quality, and air quality. Mitigating environmental impacts from mining operations is crucial, especially for host communities to guarantee environmental, economic, and social stability. In mining contexts, mitigation measures have been identified, tested, and remedied in many former and existing mining operations by reclaiming, stabilizing, and restoring disturbed soil, treating contaminated water, and controlling air or gas emissions.

Light-colored concrete and cool roofs
Buildings, car parks, pavements, bare lands, roads, and other infrastructure absorb heat from the sun more efficiently than ecosystem service providers such as water bodies and woodlands; hence, heat-inhibiting building materials are favored in mining areas. Dark-colored roofs absorb and emit more energy as heat to buildings, which results in higher surface temperatures. Extreme temperatures are reduced by lighter land surfaces. White or lighter colored roofs tend to reflect solar radiation that strikes the surface of the roofs (Fayad et al. 2021). Cool or reflective roofs help to reflect sunlight and heat away from buildings by reducing the temperature of the roof. The use of light-colored concrete and white roofs, which reflect up to 50% more light and lower the temperature, is significantly more effective in mitigating the problem of rising LST (Wavin 2022). These strategies can offer great solutions in reducing the surface temperature and urban heat island effect.

Green roofs and vegetation covers
Green roof is a vegetative layer grown on a rooftop. It entails planting vegetation on a roof in the same way that it would be done in a garden. Plants on the roof serve as perfect insulators during dry season and decrease the overall LST. Green roofs insulate buildings from heat and cool the air through evapotranspiration, reducing temperatures of the roof surface and the surrounding air. Air quality is improved as the plants absorb carbon dioxide and produce oxygen. Green roofs are sustainable option for reducing the impact of LST.

Afforestation and reforestation
Trees provide shade, absorb carbon dioxide, and release oxygen and fresh air thereby provide a cooling effect. Planting trees and other vegetation lowers surface temperatures by providing shade and cooling through evapotranspiration. The practice of planting and replanting trees in bare land and deforested region is significant to reducing the volume and rate of runoff because trees intercept and slow down precipitation before it hits the ground. This protects soil from erosion and reduces flooding.

Water harvest technologies
Water harvesting promotes environmental resilience and contributes significantly to maintaining water sustainability in many countries. Water harvesting technologies include CO 2 absorption from the atmosphere, earth dam, and rain water harvesting. Rainwater harvesting is the accumulation and deposition of rainwater for reuse onsite, rather than allowing it to run off (Kumar et al. 2016). Loveson and Misra (2004) suggested rainwater harvesting in open areas and rooftops to manage acute water problems in summer seasons as community development programs in rural areas around mining zones. Proper recharge of harvested water can augment the ground water storage and increase the ground water level (Kumar et al. 2016). Harvested rainwater can serve diverse purposes especially during dry or harmattan season.

Soil improvement farming
Adegboye (2012) discovered that tin mining activities resulted in low productivity in crop farming and fragmentation of land. Legumes are especially important for revegetating mined lands because they can use nitrogen from the air to meet their nitrogen nutrition requirements, and they can transfer this fixed nitrogen to other components of the plant or soil system (Skousen and Zipper 2018). Tephrosia vogelii, which is a native leguminous plant in many African countries, has advantages in helping post-mining land reclamation. Organic amendments (compost, wood chips, biochar, animal manure, straw, husk, geotextile, and sewage manure) and inorganic or mineral amendments (gypsum, superabsorbent polymers (SAPs), fly ash) are the two basic soil amendment types that have been used to improve soil fertility and stabilize mine site conditions, according to Maiti and Ahirwal (2019). Page-Dumroese et al. (2018) discovered that locally sourced organic amendments (biochar, wood chips, biosolids) were effective at increasing soil cover and plant growth in a mine site in North-eastern Oregon, USA.

Geospatial indices
Spectral bands of multispectral satellite images have been combined to estimate land cover for over four decades. These empirical indicators formulated by combining two or more spectral band of satellite multispectral images for the assessment of land cover are called geospatial or spectral indices. The abundance or lack of various land cover is assessed by generating geospatial indices otherwise called land use land cover (LULC) indices. Discrimination and classification of different land cover increase when spectral indices are employed, as they act as LULC indicators.

Material and methods
The study area is a combination of two mining regions in Plateau State, Nigeria, otherwise called the Jos Plateau. Landsat images of 1984, 2002, and 2020 were pre-processed with ENVI 5.2 and ArcCatalog 10.8 and processed with ArcMap 10.8 before LST and the spectral indices were generated. Google Earth images of 1984, 2002, and 2020 were used to identify various land covers and validate the maps generated with the Landsat images.

The study area
The study area as shown in Fig. 1 comprises two local government areas (Jos South and Barkin Ladi) in Plateau State, Nigeria.

Location and human settlement
Jos South is a local government area in Plateau State, Nigeria, that is home to the Governor's office in Rayfield and hence serves as the state's de facto capital. It is located between latitudes 9° 36′ 41.5915″ N and 9° 51′ 14.2973″ N and longitudes 8° 38′ 24.4785″ E and 8° 57′ 14.0240″ E. It has its headquarters in Bukuru, some 15 km from Jos, the state capital, in the north-western portion of the state. Kuru, Gyel, Du, and Vwang are the four districts of Jos South. It has a population of 311, 392 people (Federal Republic of Nigeria Official Gazette 2009) and a total land area of around 1037 km 2 .
Barkin Ladi is a local government area in Plateau State, Nigeria, with latitudes ranging from 9° 19′ 0.0890″ N to 9° 50′ 4.2664″ N and longitudes ranging from 8° 40′ 27.6936″ E to 9° 5′ 11.1122″ E. Barkin Ladi, Kuru Jenta, Heipang, Bisichi, Foron, and Kassa are among the communities that make up this area. It is about 47 km from Jos, the state capital. It covers an entire land mass of 1032 km 2 and has a population of 179, 805 (Federal Republic of Nigeria Official Gazette 2009).

Geology of the study area
The Jos Plateau is part of Nigeria's largest region with a height of over 1000 m which comprises a clearly defined highland area, standing above surrounding lowlands (Odunuga and Badru 2015). The study area is part of the Jos-Bukuru Complex, which is dominated by biotite-granite. The Jos Plateau mainly consists of granites that are part of the Precambrian Basement Complex (One Earth 2021). Distinctive alkaline younger granites (Jurassic to Triassic) comprise the second group. The older and newer basalts (quaternary) make up the third group (Ozoko 2014). The younger granite is the source of economic quantities of tantalite, cassiterite, columbite, zircon, monazite, ilmenite, thorite, molybdenite, and pyrochlore (Owolabi 2018). Tantalite, cassiterite, and columbite mining, mostly from alluvial deposits, and processing the ores have been taking place for over a hundred years on the Jos-Plateau, Nigeria (Opafunso and Owolabi 2016).

Relief, drainage, and hydrography
The Jos Plateau is the main hydrological hotspot in Nigeria, supporting agriculture, tourism, and trade. The Plateau's unique physical attributes are its high relief, especially in the north, with undulations dominating the area and ranging between 915 and 1300 m, with an average of about 1250 m, which are the source points for many rivers in northern Nigeria. The formation of its high relief makes Plateau State one of the country's mineral-rich states (Falconer 1911;Lar 2007 andMallo andWazoh 2014).
The Jos Plateau as an isolated upland massif has naturally developed a radial pattern of drainage which flows into four major river systems: Chad (largely Delimi River), Gongola, Benue, and Kaduna (Turner 1989). However, drainage within the studied area is mainly dentritic and partially trellis in pattern, with minor stream channels flowing from the highlands and linking-up with bigger streams in the lowlands (Ramadan and Haruna 2018). Most of the streams also connect with the abandoned mine ponds within the study area (Ramadan and Haruna 2018). Odunuga and Badru (2015) further described the drainage pattern of the Plateau to be radial and opined the Plateau to be the source of numerous rivers, including the Kaduna, Karami, and N'gell, which feed the Niger River; the Mada, Ankwe, Dep, Shamanker, and Wase, which flow into the Benue; the Lere, Maijuju, and Bagei, supplying the Gongola; and the Kano, Delimi, Bunga, Jamaari, and Misau, which intermittently nourish Lake Chad.

Image and map acquisition
Acquired datasets comprise three scenes of multi-temporal and cloud-free Landsat data, including Thematic Mapper (TM) of 1984, Enhanced Thematic Mapper Plus (ETM +) of 2002, and Operational Land Imager (OLI)/Thermal Infrared Sensors (TIRS) of 2020. They were freely downloaded from path 188 and row 53 in the United States Geological Survey (USGS) image archive. Datasets at an interval of 18 years were deliberately acquired to ensure uniformity between the time-nodes. Furthermore, administrative maps and local administrations from which the study area was extracted were obtained from the Centre for Space Research and Applications (CESRA), Federal University of Technology Akure, Ondo State, Nigeria.

Image pre-processing
The data provider made the imagery available in a georeferenced format and projected to Universal Transverse Mercator (UTM) coordinate system, World Geodetic System 1984 (WGS-84) datum, and Zone 32 North. Radiometric calibration and atmospheric correction of datasets were carried out using the dark of subtraction (DOS) in ENVI 5.2 where Landsat Collection1 Level-1 data products consisting of quantized and calibrated scaled digital numbers (DN) representing the multispectral image data were converted to surface reflectance (SR), attributed to the fact that DN values have no physical meaning, and time series analysis is to be studied. ENVI 5.2 was also used to convert thermal bands to at-sensor brightness temperature and normalized to correspond with other bands ranging between 0 and 1. Images converted to surface reflectance in ENVI 5.2 were extracted by mask in ArcMap 10.8 to generate different spectral bands. ArcCatalog 10.8, a unique component of the ArcGIS suite, was used to calculate the band statistics of each image before LST, and the geospatial indices of each dataset were generated.

Image processing and analysis
Maps were produced to portray the spatial distribution of vegetation, water, built-up, and bare land for the investigation region based on the three scenes of imagery utilized.
Color composite for precise distinguishing proof and the interpretation of the images were carried out by layer stacking most of the bands for each image (Fig. 2). Land covers were all clearly identifiable in the band color composite. Google Earth images of 1984, 2002, and 2020 were also used to aid the recognition, validation, and assessment of the accuracy of LST and the derived spectral indices due to their higher resolution than Landsat images.

Normalized difference vegetation index (ndvi) estimation
NDVI is the most used vegetation indicator for observing vegetation generally attributable to its favorable relationship with plant condition and abundance characteristics (Al-doski et al. 2013). It ranges from − 1 to + 1 and serves as a measure of the health or status of vegetation. NDVI is determined using Eq. 1 (Rouse et al. 1973).
In Landsat 5, TM and Landsat 7 ETM + , NIR (near infrared) is band 4 reflectance value, and RED is band 3 reflectance value. NIR is band 5 reflectance value in the Landsat 8 OLI, while RED is band 4 reflectance value.

Land surface temperature (LST) retrieval
LST represents the temperature value of an object on the earth's surface (Erthalia et al. 2019). It is used to ascertain how temperatures are changing on a global, regional, and (1) NDVI = NIR − RED NIR + RED Fig. 2 Agriculture and color infrared band combinations local level (Orhan et al., 2014). Six steps were used to extract LST from thermal bands 6 (TM and ETM +) and 10 (TIRS), as well as the spectral bands for NDVI retrieval.
1. DN conversion to top of atmosphere (TOA) spectral radiance (L λ ) The foremost step in retrieving LST involves the conversion of thermal band DN to TOA spectral radiance using Eq. 2 for Band 10 TIRS and Eq. 3 for Band 6 TM/ ETM + (U.S. Geological Survey 2019a; U.S. Geological Survey 2019b). The values of terms used in the formulae were obtained from the metadata file of the imagery.
where L λ is the TOA spectral radiance (watts/ (m 2 ·srad·µm)), M L is the band-specific multiplicative rescaling factor, Qcal is the band 10 DN image, and A L is the band-specific additive rescaling factor.
where L MINλ is the spectral radiance scaled to Q CALMIN (watts/(m 2 ·srad·µm)), L MAXλ is the spectral radiance scaled to Q CALMAX (watts/(m 2 ·srad·µm)), Qcal is the band 6 DN image, Q CALMIN is the minimum quantized calibrated pixel value, and Q CALMAX is the maximum quantized calibrated pixel value. 2. Spectral radiance (L λ ) conversion to at-sensor brightness temperature (BT) Thermal constants were used to convert the spectral radiance to at-sensor brightness temperature, as shown in Eq. 4. (U.S. Geological Survey 2019a; U.S. Geological Survey 2019b).
Absolute zero which is approximately equal to − 273.15 °C was added to obtain the result in Celsius. 3. Calculation of NDVI NDVI, as already stated in Eq. 1 must be generated in order to compute the proportion of vegetation (P v ) and land surface emissivity (ε).

Calculation of the proportion of vegetation (P v )
P v was computed from NDVI threshold values using Eq. 5 (Pal and Ziaul 2015).
where NDVI is the processed NDVI; NDVI min and NDVI max are the minimum and maximum values of the processed NDVI image respectively. 5. Calculation of the land surface emissivity (LSE) LST retrieval is greatly influenced by LSE, which is a measure of a surface's ability to release energy through radiation (Yang et al. 2014). It is a crucial metric used to compute the surface temperature of a terrain since it expresses the ability of a substance to emit thermal energy. LSE was calculated using Eq. 6 (Pal and Ziaul 2015; Sobrino et al. 2004). P V is already obtained using Eq. 5. 6. Calculation of the land surface temperature (LST) Ultimately, Eq. 7 was used to obtain the emissivity-corrected LST values (Artis and Carnahan, 1982).
where, T s is the LST in Celsius (°C), BT is at-sensor brightness temperature (°C), is the emitted radiance wavelength ( = 11.5 µm for Landsat 5 and 7 band 6 and 10.8 µm for Landsat 8 band 10), and ε is the LSE while ρ is obtained from Eq. 8.

Optimized Soil-Adjusted Vegetation Index (OSAVI) estimation
OSAVI maps canopy density variations, is unaffected by soil brightness (in the presence of various soil types), and improves sensitivity to vegetation cover higher than 50% (MicaSense, 2020). This indicator works best in locations with scant vegetation, when the soil can be seen through the canopy and the NDVI is high. The range of OSAVI values is − 1 to 1. OSAVI is determined by utilizing Eq. 9 (Rondeaux et al. 1996).

Modified Normalized Difference Water Index (MNDWI) estimation
MNDWI is preferable for improving and recovering water details for a water area with a background marked by builtup land areas, because of its superiority in minimizing and even eliminating built-up land noise over the NDWI (Xu 2006). It has a range of − 1 to + 1. It can be calculated by using Eq. 10.
where ρ GREEN is band 2 reflectance value and SWIR (short wave infrared) is band 5 reflectance value in Landsat 5 TM and Landsat 7 ETM + . In Landsat 8 OLI, ρ GREEN is band 3 reflectance value while SWIR is band 6 reflectance value.

Normalized Difference Built-up Index (NDBI) estimation
In 2003, the Normalized Difference Built-up Index (NDBI) was presented as a novel method for mapping built-up regions automatically (Zha et al., 2003). The values of NDBI range from − 1 to + 1. Built-up areas have positive values, while vegetation and waterbodies have negative values. It is calculated using Eq. 11 according to Zha et al. (2003).
where SWIR is band 5 reflectance value while ρ NIR is band 4 reflectance value in Landsat 5 TM and Landsat 7 ETM + . In Landsat 8 OLI, ρ SWIR is band 6 reflectance value and ρ NIR is band 5 reflectance value.

Dry Bare Soil Index (DBSI) estimation
DBSI was proposed by Rasul et al. (2018) for mapping urban areas and bare soil in Erbil, Iraq. DBSI values range from − 2 to + 2, with the larger values indicating more bare soil. DBSI is calculated using Eq. 12 (Rasul et al., 2018).
where ρ GREEN is band 2 reflectance value and ρ SWIR is band 5 reflectance value in Landsat 5 TM and Landsat 7 ETM + . In Landsat 8 OLI, ρ GREEN is band 3 reflectance value while ρ SWIR is band 6 reflectance value. where SWIR is band 5 reflectance value, NIR is band 4 reflectance value, and TIRS (thermal infrared) is band 6 brightness temperature value in Landsat 5 TM and Landsat 7 ETM + . In Landsat 8 OLI, SWIR is band 6 reflectance value, NIR is band 5 reflectance value, and TIRS is band 10 brightness temperature value.

Normalized Difference Bareness Index (NDBaI) estimation
Zhao and Chen (2005) introduced NDBaI to detect different kinds of bare area using Landsat data. The ability of bare soil to strongly reflect the radiation of TIR and the almost total absorption of middle infrared (MIR) wavelengths is the basis of NDBaI. (Chen, et al. 2006). The NDBaI distinguishes bare land from other land use classes. The NDBaI is generated according to Zhao and Chen (2005) using Eq. 14.
where ρ SWIR is band 5 reflectance value and TIRS is band 6 brightness temperature value in Landsat 5 TM and Landsat 7 ETM + . ρ SWIR is band 6 reflectance value and TIRS is band 10 brightness temperature value in Landsat 8 OLI.

Results and discussion
Urbanization and modernization influence land cover. The spatial distribution of temperature, vegetation, water, built-up and bare land in the study area, and their peculiar transformation are presented and discussed to identify the past and present implications of mining operations on the environment. Figure 3a, b, and c depict the spatial distribution of surface temperature for the 3 years studied in the Jos area.

Spatial distribution of vegetation
The spatial distribution of different vegetation types (grassland or forest) on the Jos Plateau is a result of the interplay of a number of environmental and ecological factors, including precipitation, temperature, topography, and soil conditions, that have been weakened by mining exploitation in the Tin City. A dense vegetation distribution improves ecosystem services and the overall economy of the mining communities. The denuded Jos Plateau's greenery was estimated to assess its peculiarity and present condition. It can be deduced from the NDVI values that vegetation has increased gradually for almost 40 years in the study area, but the resulting map depicts a moderate vegetation representing shrubs and grasslands (values between 0.2 and 0.5) confirming the findings by Orewere et al. (2019) who found out that the greenery of the Jos area is mainly shrubs and grass, with very few trees. Loss of canopy trees has been occurring for hundreds of years, possibly since the Jos Plateau began to be encroached for mining in the twentieth century. The loss of forest which is a primary ecosystem service occurs for a variety of factors other than mining (e.g., firewood collection, intense agriculture), because mining is now and primarily taking place in abandoned mines. Though efforts to reafforest the denuded Jos Plateau began in 1924, with the encouragement of natural woodland covers (Buckley 1987), the NDVI and OSAVI values show that the study area remain populated with grasslands, shrubs, and crops. Massive cultivation of trees and woody shrubs is suggested to achieve maximum vegetation growth on the Plateau. Figure 4d, e, and f show the visual description of the spatial and temporal water variations on the Jos Plateau using MNDWI. The highest peak of the MNDWI values from 1984 to 2020 (0.664, 0.923. and 0.971 respectively) is not compensated by the evident fall in the minimum value of MNDWI from 1984 to 2020 (− 0.491, − 0.621, and − 0.805 respectively), resulting in a consistent declining trend in the mean values (− 0.349, − 0.391, and − 0.411) in the last 36 years. Mineral recovery, mineral processing, and seasonal variations are some of the reasons for the drastic low waterbodies in the study area, thus making waterbodies one of the most stressed natural resources in the study area. In rural communities near mining regions, water harvesting technologies especially rainwater harvesting on rooftops can attenuate extreme water concerns during dry seasons.

Spatial variation of built-up and bare land
Generally, built-up and bare land are directly linked to industrialization, especially mining activities, since they necessitate land cover removal, building structures, and the development of mining host communities.

NDBI
The temporal built-up variations in the Jos area are depicted in Fig. 5a- The increase in built-ups and bare lands in 2020 corroborates the reason why vegetation (from NDVI and OSAVI values) declined in 2020 and confirms that the increase in built-ups and bare lands is responsible for the devegetation from the resulting maps.

DBSI
The variability of bare land regions is depicted in Fig. 6a-

NDBaI
The informative visual representation of the pattern of distribution of bare land regions using the NDBaI is depicted in Fig. 6d- The values of bare land mapped using NDBaI also correspond with that of DBSI. The negative mean values obtained from the NDBaI and DBSI revealed that bare land has reduced on the Jos Plateau due to improved vegetation growth. The decrease in bare soil is undoubtedly due to the rise of greenery.

The relationship between LST, NDVI, and OSAVI
Generally, vegetation and water bodies are inversely or negatively related to surface temperature since both  OSAVI in 1984, 2002. (d-f) MNDWI in 1984, 2002 tend to absorb more heat energy compared to other land covers. The correlations between NDVI and LST are depicted in Figs. 7a-c and 8a-c. The study area portrays a contrary exception in 1984 (Figs. 7a and 8a) as there is a significant weak positive relationship between LST and NDVI, r (1570) = 0.11, p < 0.001; and, between LST and OSAVI, r (1570) = 0.14, p < 0.001. Figures 7b and 8b show that a significant weak negative relationship exists between LST andNDVI in 2002, r (1570) = − 0.33, p < 0.001;andbetween LST andOSAVI in 2002, r (1570) = − 0.30, p < 0.001. In 2020, there was a significant weak negative relationship between LST and NDVI, r (1570) = − 0.25, p < 0.001; and between LST and OSAVI, r (1570) = − 0.23, p < 0.001 as shown in Figs. 7c and 8c. The negative relationships between LST and NDVI for 2002 and 2020 are consistent with the findings of Mohanta and Nandi (2017) and Fabeku et al. (2018). Vegetation has not increased enough to bring about a decrease in surface temperature. Hence, areas with least vegetation are experiencing higher land surface temperatures.
Forests can both absorb and release carbon. They are the second-largest carbon sinks after the oceans. Forests in good condition can trap more carbon dioxide from the atmosphere than they emit, making them significant carbon sinks. The loss of trees translates to higher release of carbon dioxide to the atmosphere. Deforested areas bring about more carbon release, thus becoming carbon sources. Higher levels of greenhouse gases (GHGs) in the atmosphere, climate change, soil erosion, flooding, and less crop production can all result from the decline of vegetative cover. Depletion of the ozone layer causes global warming which is primarily due to the increase in GHGs from human activities including mining. Forest loss leads to the extinction of animal species owing to habitat loss. Forests provide the canopy that regulates surface temperature. Low greenery results in drastic temperature variations. A surface temperature that continues to rise above plausible range can cause  (a-c) NDBI in 1984, 2002. (d-f) EBBI in 1984, 2002 heat waves and other health effects. Without forest canopy, soil erodes and washes away, giving rise to poor crop yield and dangerous gully erosion. Eroded soil lacks the ability to offset carbon due to soil organic carbon displacement.  LST and NDVI in 1984, 2002 Relationship between LST and MNDWI Figure 9a, b, and c show the linear regression between LST and MNDWI. The correlation coefficients (r) are − 0. 56, − 0.52, and − 0.47 in 1984, 2002, and 2020 respectively portraying a significant moderate inverse or negative correlation relationship between LST and MNDWI. This correlates with the findings of Kushawaha and Sharma (2019) that surface temperature has a negative relationship with water bodies, implying that decrease in water bodies leads to increase in surface temperature. Reduced surface water with the ever-increasing population of people in the world is consequential to an increasing energy production to pump, treat, transport, and heat water, thereby increasing pollution, greenhouse gases (GHGs) emissions, and global warming. Excellent management of surface water is essential to correctly respond to climate change's harshest effects and limit GHGs emissions.
Both dense vegetation and water are good indicators of a healthy climate and weather pattern. The forest assists to regulate the hydrological cycle, which helps to control the amount of water in the atmosphere. There is limited water in the air to be returned to the land in deforested areas, resulting in soil dryness and, consequentially, crop failure.

Relationships between NDBI, EBBI, and LST
In Figs. 10a and 11a, there is a significant moderate positive association between LST and NDBI in 1984, r (1570) = 0.51, p < 0.001; and between LST and EBBI, r (1570) = 0.58, p < 0.001. From Figs. 10b and 11b, a significant strong positive correlation exists between LST and NDBI in 2002, r (1570) = 0.75, p < 0.001 and between LST and EBBI in 2002, r (1570) = 0.74. In 2020, Figs. 10c and 11c also reveal a significant strong positive relationship between LST and NDBI, r (1570) = 0.69, p < 0.001 and between LST and EBBI, r (1570) 0.66, p < 0.001. For the three time nodes, this study is congruent with the discovery of Kushawaha and Sharma (2019) and Sarif and Gupta (2019) indicating that surface temperature has a negative relationship with built-up regions.  LST and MNDWI in 1984, 2002 Thus, the increase in built-ups due to urbanization and industrialization leads to a corresponding increase in the surface temperature of the study area.  (2017) and Imen et al. (2021). This implies that a reduction in bare land areas will result in a lower surface temperature due to increased vegetation growth, since vegetation cools the environment by transpiration and shading. The cultivation of native trees, shade trees, and soil-improving trees in bare land areas are sustainable mitigation measures to reduce bare land regions in the investigated area.

Conclusions
In this study, the potential of geospatial techniques to study the surface temperature of land, waterbodies, vegetation, and developed areas within the Jos area was assessed by estimating land surface temperature (LST) and seven geospatial indices which act as land cover indicators. Vegetation and waterbodies are natural resources  LST and EBBI in 1984, 2002 which are basic for sustainability of an environment. Sun's energy that reaches the surface of the earth is closely linked to vegetation and waterbodies within the study area by evapotranspiration, condensation, precipitation, and photosynthesis. The findings revealed that the variation of surface temperature in the researched area is determined by greenery, surface waters, bare lands, and developed regions. Furthermore, it was established that when there is deficient greenery and available bodies of water, the surface temperature will tend to increase. The results show that the Jos area is not densely vegetated; the number of waterbodies are also dwindling. Additionally, the increase in bare land and builtup corresponds with the Jos Plateau's increased surface temperature. The findings in this study show how artisanal mining activities have affected the dense vegetation and waterbodies in the Jos area. Afforestation and reclamation are not heightened in the Jos area as the average values from the study revealed moderate vegetation representing shrubs and grasslands. That woodland (forest) is not dominant in the Jos area for the past 36 years cannot be unrelated to artisanal mining and other anthropogenic activities.
Sequel to the findings of this study, profits should not be prioritized over the environment despite the apparent and prevailing global economic challenges, rather a palpable balance should be maintained. Secondly, mitigation measures should be taken to avoid further degradation and devitalization of the environment to ensure long-term environmental quality and viability. Thirdly, further work could be undertaken to better understand the peculiar factors responsible for the stress of vegetation and waterbodies using satellite data of different spatial resolutions such as Sentinel, IKONOS, and MODIS. Finally, revegetation in the Jos area should be enhanced as the study revealed that the area has moderate vegetation.
Acknowledgements The authors are indebted to US Geological Survey (USGS) for making the Landsat data publicly available.
Ayodele Olumuyiwa Owolabi: conceptualization, monitoring, and evaluation, writing-editing Author contribution Olumuyiwa Temidayo Ogunro: acquisition and analysis of satellite images, writing-first draft of a manuscript.
Data availability On reasonable request, the corresponding author will make the datasets and materials utilized during this research available.  LST and NDBaI in 1984, 2002