https://orcid.org/0000-0003-0532-7689
Abstract
Climate variability and its impact have become of serious interest to environmentalists worldwide, but reports from many sub-Saharan African countries are still relatively more uncertain than many parts, mostly due to challenges with datasets and methodology. In this study, the daily rainfall and temperature records for 34 years (1984–2018; for which data were available at the time of study) were examined alongside changes in the land cover at a natural forest reserve in the Ife area of southwestern Nigeria. The specific objective was to examine climate variability with changes in the land cover of the nature reserve. Data used included archival records of rainfall and temperature and open-access Landsat satellite imageries of the area. Results showed that area experienced rainfall fluctuations, significant monthly decrease and temperature rise in many months increase. The Normalized Difference Vegetation Index and supervised classification of land cover also revealed a decline in vegetation health and loss of forested land to non-forest uses such as farmlands and built-up lands over time. The study could not establish a direct link between forest loss and climate change in the study area but detected a complex implication of urban pressure through human activities and urbanization, and thus concluded that the impact of climate change on the forest environment has been exacerbated by pressure for urban growth and probably a poorly monitored quest for land resource.
Introduction
The forests cover over 42 million km2 in tropical, temperate, and boreal lands, and about 30% of the entire land surface where they provide ecological, economic, social, and aesthetic services to natural systems and humankind [1, 2]. These forests provide refuges for biodiversity, food, medicinal, and forest products supplies to humankind. The forest ecosystem regulates the hydrologic cycle, protects soil resources, and influences climate through exchanges of energy, CO2, water, and atmosphere constituents. Globally, forests are known to store up to 45% of terrestrial carbon and contribute about 50% of terrestrial net primary production with the capacity to sequester large amounts of carbon annually [3]. Forestry and livelihood activities are extensively vulnerable to change and variability in climatic elements, especially temperature and rainfall [4, 5]. Studies have argued that change, variability as well as extreme climatic conditions may have, along with poor forest management and unsustainable land uses aggravated forest degradation and associated consequences in many parts of the world [6, 7]. Climate variability may span the short term but it can impact both environmental and agricultural processes significantly, in either term [8].
As typical of many rainforest belts, unsustainable exploitation of forest resources by man through deforestation, urbanization, bush burning and conversion to agricultural lands have remarkably affected the Nigerian's primary forests [9–12]. Land use and land cover changes with climate variability are key drivers of global environmental change, and these drivers have implications for national and international policy issues [13–15]. Wright [16] noted that land use and forest cover are changing throughout the tropics and that analysis of multidate Landsat imageries across Africa revealed an annual value of 85000 ha per year—as the rate of degradation over the region. Also, studies [17, 18] argued that many reserved forest areas have increasingly been endangered by anthropogenic and natural causes and that Nigeria ranked high on the list of countries with significant cases of deforestation [19, 20]. In general, conservation of the forest resources is a major aim for establishing forest reserves, and the reserves are expected to be adequately managed to ensure the fulfilment of the purposes for which they are created [21], and as such as to perform essential ecosystem functions and maintain the micro-climatic condition of the area.
Furthermore, studies on Nigerian forests have emphasized rich but threatened ecosystems [22–25] with significant consequences on human and environmental compositions. The forest ecosystems are largely threatened by deforestation and unsound management strategies initiated by inadequate enforcement of existing laws and corrupt practices [26]. Many parts of forest reserves have been lost to unmonitored urban growth while others have been severely poached for their animals, and timber resources are often unsustainably harvested [27–29]. Natural factors of extreme climatic events such as drought and anthropogenic factors, including mining, agriculture, urbanization, transportation, and construction have also been implicated in a significant level of forest degradation [30]. In Africa, the impact of climatic variability (change, extreme conditions, and fluctuations) may have been exacerbated by poor technological development, low economic strength or high poverty level, climate awareness and education as well as the financial and political will to reverse the vulnerability [31]. Devastations due to climatic variabilities have also been exacerbated by anthropogenic activities, including unsustainable agricultural activities and urbanization [32]. Both climate and human activities are however linked in feedback interacting systems, such that climatic variabilities affect human livelihoods, and vice versa. Planton et al. [33] attributed an increase in the frequency and number of extreme conditions to greenhouse gases and aerosol anthropogenic emissions while food security has become an important issue of concern in discussions of climate change [34]. In all, the subject matters of climate change, variability, and extreme conditions have become important issues for discussion among agricultural industries, and governments, especially as the climate becomes an important subject of the Sustainable Development Goals (SDG 13). SDG 13 focuses on urgently combating climate change and its effects.
Like forestry, agricultural production in sub-Saharan Africa mainly depends on rainfall for wetness and may be rigorously compromised by fluctuations and extreme climatic conditions [35]. Climatic variability events such as extreme climate conditions, spells, fluctuations, and change will cause loss of agricultural land, shorten growing seasons, lower yields, and consequently cause food shortages [36]. The impact of climate change and extreme climatic conditions on agriculture in Nigeria has generated interest and discussion among researchers at many local and international forums [34, 36–38]. Adele & Todd [38] noted that agriculture and forestry engagements in West Africa is particularly sensitive to environmental variability, particularly rainfall variability, with significant effects on crop productivity and farming livelihoods in the region. Climate change is known to modify the frequency and severity of extreme climate conditions, giving rise to upsets in agricultural conditions, such as disruption of water balance and air temperature balance [39]. In Nigeria, Apata [40] argued that the impact of climate change on agriculture in Nigeria varied regionally; whereas rainfall-associated grain production declined by about 178.4% between 1971 and 2000 in the northwestern region, it varied by 20% in the southwestern, and 281% in the south-southern regions of Nigeria. In another set of studies [41–43], findings showed a prevalence of cold stress and drought incidence in the northern part of Nigeria, and a prevalence of flooding and attendant fatalities [44, 45] since 1951, exacerbating stress on people and ecosystems, and consequently reducing productivity. According to Mustapha and Zineddine [41], drought occurrences have resulted in retarded plant growth and eventual death while flooding causes erosion runoff that washes away farmlands. Also, elevated temperature increases the incidences of pest and disease infestation, thus limiting overall yield and ecosystem support services [34, 42]. Other studies [46, 47] also indicated that crop yields and farm productivity (net revenue per hectare) were sensitive to a marginal change in climate variables (temperature and precipitation). The studies indicated that the degree of the impact of climate change is however influenced by available responses, coping, and adaptation strategies. In general, Table 1 shows the generalized summary of the impact of climate change in Nigeria [48], revealing that the impact of climate change is predicted to significantly worsen in various ways, including the increased severity and frequency of extreme events such as floods, droughts, and heat waves.
Projected trends in some climate variables over the ecological zones in Nigeria [49]
| Ecological zones | ||||
|---|---|---|---|---|
| Climate variable | Mangrove | Rainforest | Guinea and Sudan Savanna | Sahel savanna |
| Temperature | Increasing | Increasing | Increasing | Increasing |
| Rainfall amount | Increasing | Increasing | Decreasing | Decreasing |
| Rainfall variability | Increasing | Increasing | Increasing | Increasing |
| Drought | Likely to intensify | Likely to intensify | Increasing | Increasing |
| Storms and floods | Increasing | Increasing | Likely to intensify | Likely to intensify |
| Sea level rise | Increasing | No available record | ||
| Ecological zones | ||||
|---|---|---|---|---|
| Climate variable | Mangrove | Rainforest | Guinea and Sudan Savanna | Sahel savanna |
| Temperature | Increasing | Increasing | Increasing | Increasing |
| Rainfall amount | Increasing | Increasing | Decreasing | Decreasing |
| Rainfall variability | Increasing | Increasing | Increasing | Increasing |
| Drought | Likely to intensify | Likely to intensify | Increasing | Increasing |
| Storms and floods | Increasing | Increasing | Likely to intensify | Likely to intensify |
| Sea level rise | Increasing | No available record | ||
Projected trends in some climate variables over the ecological zones in Nigeria [49]
| Ecological zones | ||||
|---|---|---|---|---|
| Climate variable | Mangrove | Rainforest | Guinea and Sudan Savanna | Sahel savanna |
| Temperature | Increasing | Increasing | Increasing | Increasing |
| Rainfall amount | Increasing | Increasing | Decreasing | Decreasing |
| Rainfall variability | Increasing | Increasing | Increasing | Increasing |
| Drought | Likely to intensify | Likely to intensify | Increasing | Increasing |
| Storms and floods | Increasing | Increasing | Likely to intensify | Likely to intensify |
| Sea level rise | Increasing | No available record | ||
| Ecological zones | ||||
|---|---|---|---|---|
| Climate variable | Mangrove | Rainforest | Guinea and Sudan Savanna | Sahel savanna |
| Temperature | Increasing | Increasing | Increasing | Increasing |
| Rainfall amount | Increasing | Increasing | Decreasing | Decreasing |
| Rainfall variability | Increasing | Increasing | Increasing | Increasing |
| Drought | Likely to intensify | Likely to intensify | Increasing | Increasing |
| Storms and floods | Increasing | Increasing | Likely to intensify | Likely to intensify |
| Sea level rise | Increasing | No available record | ||
A report of Nigeria’s Federal Ministry of Environment [49] noted that the monetary implication of climate change impacts for Nigeria will be worth 6%–30% of the national Gross Domestic Productivity by 2050, amounting between US $100 billion and US $460 billion. Furthermore, studies have linked many reported cases of climate change-induced forced migrations (specifically, from the more vulnerable northern Sudano-Sahelian ecosystems to the southwards Guinean ecosystem), with the increasing wave of farmers-herders and between-herdsmen conflicts in Nigeria [50–52]. For example, Fabiyi and Otunuga [50] and Onwuamanam [51] documented notable changes in the ecological landscape of Northern Nigeria that have resulted from environmental degradation, resource scarcity, population growth, climate change, and conflicts between sedentary farmers and pastoralists. Examples of such changes include the disappearance of the foraging grounds for cattle and, the shrinking of Lake Chad, whose 95% of water volume has reportedly been lost within the past 50 years, as being associated with desertification and other impacts of climate change.
Given the recognition of the nature of the impact of climate in Nigeria, efforts were geared towards improving adaptation, coping strategies, and resilience capacities at local, regional, and national scales. Such actions include the previously cited National Adaptation Plan Framework [49], which outlines the adaptation goals and activities at the national level, drawn in line with the Cancun Adaptation Framework, whose objective is to enhance action in adaptation, including through international cooperation and coherent consideration of matters relating to adaptation under the United Nations Framework Convention on Climate Change. Other efforts include Nigeria's updated Nationally Determined Contributions, 2021 Climate Change Act, National Climate Change Policy, Long-Term Vision, Medium-Term National Development Plan, and the Biennial Update Report, among other plans. Furthermore, the Nigerian Meteorological Agency has since 2016 embarked on massive improvement in climate data collection across the different ecological zones—mangrove, rainforest, and savanna—Guinea, Sudan, and Sahel—by setting up digital meteorological/synoptic stations in Nigeria to complement the existing ones. There are about 123 synoptic stations in Nigeria against the 55 stations as of 2009 [42].
Butu et al. [53] described the Nigerian government's efforts as belatedly emerging, suggesting that the actions have been delayed beyond the usual time, and signalling that such actions may have been expectedly ineffective. Specifically, they noted that climate change adaptation was yet to feature in the mainstream national discourse as of 2022, and as such requisite frameworks for adaptation practice are yet to substantially emerge. Butu et al. [53] also criticized the limited political will, and consequently, limited financing to drive adaptation and climate emergency endeavours. Other factors that have been identified in literature include the absence of citizens demanding improved environmental governance weak collaboration between ministries and government agencies with different roles in climate change administration. With regards to the synoptic stations, climate data have, rather become cheaper, and grown far more expensive for an average researcher with increased bureaucratic bottlenecks that have forced many climate researchers to adopt remotely sensed data from the archives of the Prediction of Worldwide Energy Resources (POWER) of United States of America’s National Aeronautics and Space Administration’s Langley Research Centre instead of the national records of in situ data. Besides, except for the well-known stations (Fig. 1), whose data have been adjudged to be spatially deficient (due to the size of Nigeria, see [42]), climate data from recently fixed stations have not been proven to be well-managed, due to lack of climate records from most, if not all of them.
![Distribution of existing meteorological stations in Nigeria based on accessible in situ climate records [42].](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/oocc/5/1/10.1093_oxfclm_kgaf008/3/m_kgaf008f1.jpeg?Expires=1747897074&Signature=n3nLmzQJQmlfUzFeskLGjvam6Q6ZXDM1kUw01s5ZW~r8m07tfGvmxQSp-sXG0~UqkSG0spZZ7EtAOTp6qPiHRtagTOj-t7zHJneCUqGN9PcxouUmk1H8hG96UfOJLkCe3cI1audzRnkCnm871I4KvrgN7qJly0lPZWNUkgA5NLiqOpi98gTpOpdGPeElI2t3dEdM40OSLXmSg-WsNHJYWYzrZl3135sFyvb3pzATbbq1ENaRauMSovjqpQd9pv-UfjghvLODukaHRGfC7az0n8jTxZPM32v-ZKx8K6MboT0gKC9lzgRfPZO6LcEVaxyq0H48lbHE4Ydc9Szz5Sd5ng__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Distribution of existing meteorological stations in Nigeria based on accessible in situ climate records [42].
In this study, the impact of climate variability on forestland and agricultural activities in a part of the Nigerian forest belt is investigated using a multi-perspective approach of remote sensing, fieldwork, and social surveys. Specific objectives were to examine climate variability and its impact on forest ecosystems and agricultural activities in a part of southwestern Nigeria. The study is expected to represent occurrences in many forest reserve ecosystems and farmlands in the region.
Materials and methods
Study area
The study area is southwestern Nigeria where forest reserve and agricultural activities are well replicated. Southwest Nigeria hosts at least 13 forest reserves, including those in Oluwa, Akure, and Idanre forest reserves in Ondo State, Ise in Ekiti State, and Oba Hills in Osun State, among others. The Ife Nature Forest Reserve is one of the forest reserves in the region which has experienced a significant level of climate and human-induced degradation in recent times [26]. The forest reserve extends up to 100.1 km2 of the landmass in Ife South Local Government area of Osun State (7°4ʹN, 4°19ʹ E–7°15ʹ N, 4°28ʹ E) (Fig. 2).
Ife Nature Reserve in southwestern Nigeria (Google Earth maps).
The study is situated within the tropical rainforest climate (Af) based on Koppen’s climate classification. The region in this class is characterized by an average precipitation of at least 60 mm, a small temperature range, and convectional storms [42]. Average climate in the area varies as 82–200 cm mean annual rainfall, with the wet season occurring between March/April and October; and the dry season occurring between November and March [54]. The relative humidity is about 70% while the mean monthly temperature is about 27°C [55]. The study area is generally undulating, lying at an altitude of about 160 m above sea level with soils formed from rocks of pre-Cambrian basement complex formation, particularly granites, gneisses, quartz-schist, biotite gneisses, and schist, and largely belonging to the ferruginous tropical soil. The natural vegetation of the area is the tropical rainforest characterized by multiple canopies and lianas [56]. Commonly found trees are Milicia excelsa, Afzelia bipindensis, Antiaris africana, Brachystegia nigerica, Lophira alata, Lovoa trichiliodes, Terminalia ivorensis, T. superba, Triplochiton scleroxylon [57]. The farm settlements around the reservoir are occupied by a population consisting mainly (85%) of small-scale farmers. Their farm products include tree (Theobroma cacao, Cola acuminata), tuber/root (Dioscorea rotundata, Manihot esculenta), cereal (Zea mays) crops, and leafy vegetables (Abelmoschus esculentus, Talinum triangulare, Amaranthus hybridus, Basella alba).
Materials and methods
Data used in the study included multi-date Landsat (Thematic Mapper, TM) of 1986 (Landsat 5 of 17th December) and 1999 (Landsat 7 of 13th December), Enhanced Thematic Mapper plus (ETM+, of 1st March) of 2002 and Operational Land Imager (Landsat 8, OLI/TIRS of 4th March) version of 2014 and 2021. Data also included results of daily records of temperature and rainfall for 38 (1984–1988) years from the database of the POWER program of the USA’s NASA, based on availability.
The climate records were first checked for consistency and suspects were removed before the data were subjected to descriptive and inferential analysis including measures of central tendency, non-linear (wavelet analysis), and linear (regression). Wavelet analysis of signals in time–frequency space into small waves or bit signals [58], was used in this study to characterize the variability in each of the selected parameters for study. The simple linear regression analysis, on the other hand, was used to test the magnitude of the relationship and influence among dependent variables (year/time) and independent variables (climatic variables), as well as the rate of change for each land cover. The Landsat imageries, which were preferred based on their availability, were first corrected for geometric and radiometric errors using basic Geographic Information System software before they were classified, following a modified Anderson [59] landcover/use classification schema. The classification schema was adapted to represent the primary land use types: (i) non-forested areas, which encompass built-up regions, roads, bare land, and other intensively cultivated zones, (ii) secondary regrowth, which includes arable lands, permanent crops, pastures, and degraded areas with some savanna characteristics, (iii) forestland, characterized by a tree-crown area density of at least 10%, is populated with trees that can produce timber or other wood products, and (iv). waterbodies, including streams, rivers, and other aquatic features. The areas covered by a given land cover type for the different times were subsequently compared. Wavelet analysis was achieved using Paleontological Software (PAST).
Results
Climate variability
Table 2 shows the general variability in temperature and rainfall over the study area. Average daily rainfall was 0.46 cm with 100%–136.2% annual variability. Rainfall variability peaked in November–April and August while rainfall occurred at maximum reach in July-August-September (8.1–10.02 cm). The temperature in the area varies from 23.6°C to 26.17°C.
Temporal variability in rainfall and temperature
| Rainfall | Temperature (°C) | ||||||
|---|---|---|---|---|---|---|---|
| Month | Mean (cm) | CV (%) | Peak (cm) | Trend (a + bx) | Mean | CV (%) | Trend (a + bx) |
| January | 0.04 | 323.9 | 2.0 | −27.7 − 0.01x* | 24.2 (19.1 - 27.4) | 6.8 | 28.9 − 0.002x |
| February | 0.09 | 201.1 | 1.6 | −33.7 − 0.02x* | 25.7 (20.7–29.0) | 4.8 | 10.52 + 0.008x* |
| March | 0.25 | 141.3 | 3.4 | 55.2 − 0.03x* | 26.2 (22.5–28.9) | 3.2 | 10.28 + 0.008 x* |
| April | 0.44 | 107.9 | 4.8 | 86.5 − 0.04x* | 26.0 (23.7–29.1) | 2.7 | 11.08 + 0.007 x* |
| May | 0.57 | 84.7 | 4.2 | 14.2 − 0.004x | 25.4 (23.2–28.2) | 2.7 | −0.56 + 0.013 x* |
| June | 0.79 | 87.6 | 5.6 | 23.9 − 0.001x | 24.6 (22.1–26.3) | 2.6 | −2.87 + 0.014 x* |
| July | 0.82 | 97.9 | 8.1 | 12.1 − 0.002x | 23.8 (21.4–26.0) | 2.9 | −32.87 + 0.03 x* |
| August | 0.72 | 114.9 | 9.6 | 79.5 − 0.004x | 23.6 (21.0–25.2) | 2.5 | −19.75 + 0.02 x* |
| September | 0.92 | 84.7 | 10.0 | −25.3 + 0.02x | 24.1 (21.8–26.0) | 2.7 | −19.34 + 0.02 x* |
| October | 0.64 | 92.0 | 5.5 | 32.8 − 0.01x | 24.7 (22.5–26.5) | 2.6 | −17.30 + 0.02 x* |
| November | 0.15 | 145.6 | 1.5 | −102.3 + 0.05x* | 24.9 (20.9–26.8) | 3.2 | −22.30 + 0.02 x* |
| December | 0.05 | 350.9 | 2.5 | 8.2 + 0.004x | 24.1 (19.6–26.7) | 5.2 | −27.42 + 0.03 x* |
| Overall (Annual) | 0.46 | 136.2 | 10.0 | −27.7 + 0.01x* | 24.8 (19.1–29.1) | 5.1 | 28.94 + 0.002 x |
| Rainfall | Temperature (°C) | ||||||
|---|---|---|---|---|---|---|---|
| Month | Mean (cm) | CV (%) | Peak (cm) | Trend (a + bx) | Mean | CV (%) | Trend (a + bx) |
| January | 0.04 | 323.9 | 2.0 | −27.7 − 0.01x* | 24.2 (19.1 - 27.4) | 6.8 | 28.9 − 0.002x |
| February | 0.09 | 201.1 | 1.6 | −33.7 − 0.02x* | 25.7 (20.7–29.0) | 4.8 | 10.52 + 0.008x* |
| March | 0.25 | 141.3 | 3.4 | 55.2 − 0.03x* | 26.2 (22.5–28.9) | 3.2 | 10.28 + 0.008 x* |
| April | 0.44 | 107.9 | 4.8 | 86.5 − 0.04x* | 26.0 (23.7–29.1) | 2.7 | 11.08 + 0.007 x* |
| May | 0.57 | 84.7 | 4.2 | 14.2 − 0.004x | 25.4 (23.2–28.2) | 2.7 | −0.56 + 0.013 x* |
| June | 0.79 | 87.6 | 5.6 | 23.9 − 0.001x | 24.6 (22.1–26.3) | 2.6 | −2.87 + 0.014 x* |
| July | 0.82 | 97.9 | 8.1 | 12.1 − 0.002x | 23.8 (21.4–26.0) | 2.9 | −32.87 + 0.03 x* |
| August | 0.72 | 114.9 | 9.6 | 79.5 − 0.004x | 23.6 (21.0–25.2) | 2.5 | −19.75 + 0.02 x* |
| September | 0.92 | 84.7 | 10.0 | −25.3 + 0.02x | 24.1 (21.8–26.0) | 2.7 | −19.34 + 0.02 x* |
| October | 0.64 | 92.0 | 5.5 | 32.8 − 0.01x | 24.7 (22.5–26.5) | 2.6 | −17.30 + 0.02 x* |
| November | 0.15 | 145.6 | 1.5 | −102.3 + 0.05x* | 24.9 (20.9–26.8) | 3.2 | −22.30 + 0.02 x* |
| December | 0.05 | 350.9 | 2.5 | 8.2 + 0.004x | 24.1 (19.6–26.7) | 5.2 | −27.42 + 0.03 x* |
| Overall (Annual) | 0.46 | 136.2 | 10.0 | −27.7 + 0.01x* | 24.8 (19.1–29.1) | 5.1 | 28.94 + 0.002 x |
Note: x = year (where 1984 ……. = 1st………nth), a significant trend of change at a 95% confidence level is asterisked in the trend column.
Temporal variability in rainfall and temperature
| Rainfall | Temperature (°C) | ||||||
|---|---|---|---|---|---|---|---|
| Month | Mean (cm) | CV (%) | Peak (cm) | Trend (a + bx) | Mean | CV (%) | Trend (a + bx) |
| January | 0.04 | 323.9 | 2.0 | −27.7 − 0.01x* | 24.2 (19.1 - 27.4) | 6.8 | 28.9 − 0.002x |
| February | 0.09 | 201.1 | 1.6 | −33.7 − 0.02x* | 25.7 (20.7–29.0) | 4.8 | 10.52 + 0.008x* |
| March | 0.25 | 141.3 | 3.4 | 55.2 − 0.03x* | 26.2 (22.5–28.9) | 3.2 | 10.28 + 0.008 x* |
| April | 0.44 | 107.9 | 4.8 | 86.5 − 0.04x* | 26.0 (23.7–29.1) | 2.7 | 11.08 + 0.007 x* |
| May | 0.57 | 84.7 | 4.2 | 14.2 − 0.004x | 25.4 (23.2–28.2) | 2.7 | −0.56 + 0.013 x* |
| June | 0.79 | 87.6 | 5.6 | 23.9 − 0.001x | 24.6 (22.1–26.3) | 2.6 | −2.87 + 0.014 x* |
| July | 0.82 | 97.9 | 8.1 | 12.1 − 0.002x | 23.8 (21.4–26.0) | 2.9 | −32.87 + 0.03 x* |
| August | 0.72 | 114.9 | 9.6 | 79.5 − 0.004x | 23.6 (21.0–25.2) | 2.5 | −19.75 + 0.02 x* |
| September | 0.92 | 84.7 | 10.0 | −25.3 + 0.02x | 24.1 (21.8–26.0) | 2.7 | −19.34 + 0.02 x* |
| October | 0.64 | 92.0 | 5.5 | 32.8 − 0.01x | 24.7 (22.5–26.5) | 2.6 | −17.30 + 0.02 x* |
| November | 0.15 | 145.6 | 1.5 | −102.3 + 0.05x* | 24.9 (20.9–26.8) | 3.2 | −22.30 + 0.02 x* |
| December | 0.05 | 350.9 | 2.5 | 8.2 + 0.004x | 24.1 (19.6–26.7) | 5.2 | −27.42 + 0.03 x* |
| Overall (Annual) | 0.46 | 136.2 | 10.0 | −27.7 + 0.01x* | 24.8 (19.1–29.1) | 5.1 | 28.94 + 0.002 x |
| Rainfall | Temperature (°C) | ||||||
|---|---|---|---|---|---|---|---|
| Month | Mean (cm) | CV (%) | Peak (cm) | Trend (a + bx) | Mean | CV (%) | Trend (a + bx) |
| January | 0.04 | 323.9 | 2.0 | −27.7 − 0.01x* | 24.2 (19.1 - 27.4) | 6.8 | 28.9 − 0.002x |
| February | 0.09 | 201.1 | 1.6 | −33.7 − 0.02x* | 25.7 (20.7–29.0) | 4.8 | 10.52 + 0.008x* |
| March | 0.25 | 141.3 | 3.4 | 55.2 − 0.03x* | 26.2 (22.5–28.9) | 3.2 | 10.28 + 0.008 x* |
| April | 0.44 | 107.9 | 4.8 | 86.5 − 0.04x* | 26.0 (23.7–29.1) | 2.7 | 11.08 + 0.007 x* |
| May | 0.57 | 84.7 | 4.2 | 14.2 − 0.004x | 25.4 (23.2–28.2) | 2.7 | −0.56 + 0.013 x* |
| June | 0.79 | 87.6 | 5.6 | 23.9 − 0.001x | 24.6 (22.1–26.3) | 2.6 | −2.87 + 0.014 x* |
| July | 0.82 | 97.9 | 8.1 | 12.1 − 0.002x | 23.8 (21.4–26.0) | 2.9 | −32.87 + 0.03 x* |
| August | 0.72 | 114.9 | 9.6 | 79.5 − 0.004x | 23.6 (21.0–25.2) | 2.5 | −19.75 + 0.02 x* |
| September | 0.92 | 84.7 | 10.0 | −25.3 + 0.02x | 24.1 (21.8–26.0) | 2.7 | −19.34 + 0.02 x* |
| October | 0.64 | 92.0 | 5.5 | 32.8 − 0.01x | 24.7 (22.5–26.5) | 2.6 | −17.30 + 0.02 x* |
| November | 0.15 | 145.6 | 1.5 | −102.3 + 0.05x* | 24.9 (20.9–26.8) | 3.2 | −22.30 + 0.02 x* |
| December | 0.05 | 350.9 | 2.5 | 8.2 + 0.004x | 24.1 (19.6–26.7) | 5.2 | −27.42 + 0.03 x* |
| Overall (Annual) | 0.46 | 136.2 | 10.0 | −27.7 + 0.01x* | 24.8 (19.1–29.1) | 5.1 | 28.94 + 0.002 x |
Note: x = year (where 1984 ……. = 1st………nth), a significant trend of change at a 95% confidence level is asterisked in the trend column.
Both rainfall and temperature fluctuated temporally over the study period. The trend analysis used to examine the rate of change in rainfall and temperature over time indicates that rainfall exhibited a complex trend; fluctuating and appearing to increase in some months while it reduced in others. While the rainfall exhibited a decreasing trend in March–August and October, it was increasing (positive) for most of the dry months (November–February) and September. For air temperature, the months, except January as well as the overall annual mean value showed a significant rate of change since the 1980s (Fig. 3a and b). In general, rainfall has exhibited a significant decrease in the early four months (January-April) of a year while most months have exhibited an increasing pattern; both indicating significantly warmer climates (Table 2).
(a) Temporal variability in mean daily rainfall in Ife area, southwestern Nigeria. (b) Temporal variability in mean daily air temperature in Ife area, southwestern Nigeria.
This result is also important because the imageries used for forest analysis in tropical areas, including the study area are often the ones acquired in the dry season, because of the limitation of the impact of cloud cover on image quality at this period.
In addition, the results of the boxplot description showed outliers that indicated some extreme levels of rainfall and temperature in the area (Fig. 4ai–bii). Specifically, July, August, and September showed outliers above rainfall peaks throughout the year, suggesting that the study area may be vulnerable to the effects of excess wetness in these periods. The years 1988, 1997, 2006, 2008, and 2018 were the periods of the July-September outliers (Fig. 4ai). Furthermore, analysis of the wavelet transform plot (Fig. 4aii) shows an unclear band of consistent change in low signal (blue colour) at the upper limit, which is likely to be associated with noise; missing data is not suggested as there were none. The red band in the middle indicates a high disturbance in signals, correlating with relative uniformity in rainfall in the mid-period of the year. Furthermore, the different shapes in the patterns observed in the wavelet result suggest irregular temporal patterns in rainfall across the study period. These irregular patterns are the consequence of temporal variability in rainfall which signals significant variability in both the average and extreme rainfall events in the study area.
Periods of extreme conditions in rainfall (ai), air temperature (aii), and variability signals in rainfall (bi) and air temperature (baii), within the study period.
In terms of air temperature, variability occurred more in December and January than in the rest of the months, probably due to the prevalence of Harmattan, which typically dominates the dry season in these months. The Harmattan is a cold dust-laden wind from the Sahara Desert, brought by the Tropical Continental (cT) airmass, that accompanies the hot dry season between December and March [42]. The air temperature relatively declined from April before rising in September. Extreme air temperature conditions in the study area are expectedly low (lower than that of the rainfall), as low-temperature variability is an important characteristic of areas in the tropics [6] (Fig. 4bi). The wavelet transforms plot shows two clear distributions of the signals; the upper with low signal disturbance (blue) and the lower part with high disturbance (red band), suggesting relative uniformity in temperature, most of the time (Fig. 4bii). Unlike rainfall, the patterns shown in the wavelet results were more regular and predictable, indicating lesser temporal variability in temperature than rainfall.
Change in forest land and other land uses/cover
It is important to recast that the dry season months (when Landsat imageries are preferred to capture more quality data, being of the optical remote sensing group that is vulnerable to atmospheric distortions) experienced decreasing rainfall but increasing temperature trends, indicating warmer climates since 1984. Similarly, the results of the Normalised Difference Vegetation Index (NDVI) revealed that the forest reserve was healthier in 1986 than all the subsequent years, suggesting that vegetation health has decreased over time (Fig. 5).
Changes in vegetation health (Normalized Difference Vegetation Index) over the study area.
By 2004 and 2018, the vegetation index has become coarser across the study area and vegetation degradation has become obvious. What is however not obvious is whether the vegetation degradation is a result of climate change/variability or anthropogenically motivated. To ascribe the vegetational changes to a jurisdiction, the landuse/cover change was evaluated, and the results indicated that the forested vegetation was more pronounced in 1999 than in the subsequent years (Fig. 6). Areas classified as forest increased by about 16% between 1986 and 1999 but later declined by a similar proportion between 1999 and 2004, and by 17% between 2004 and 2018.
Changes in land cover over the study area.
Secondary regrowth and areas classified as non-forest occupied more land area in 1999 but were generally reduced (from 40.6 to 74.7 km2, and from 2.8 to 5.0 km2, respectively) between 1986 and 2018. The area covered by forest vegetation thus declined by 62.9% (from the initial value of 56.1 km2 in 1986 to 20.8 km2 in 2018). A specific proportion of land cover gains and losses is provided in Fig. 7.
Proportion of landcover/use by year of study.
Figure 7 shows the proportions of the different land cover types, and this provides evidence of land cover changes over time, some of which may be associated with climate change or human activities. First, the significant decrease in forests from 1986 to 2014, particularly between 1986 and 1999 may be attributed to a combination of factors, including deforestation for agriculture, logging, and urbanization. On the other hand, the proportion of non-forest areas which has increased over time, particularly between 1986 and 1999 can be attributed to deforestation and land conversion for agriculture and other human activities. In terms of secondary regrowth, it has fluctuated over time, a situation that may be due to natural factors, such as forest fires and drought, as well as human activities, such as deforestation and land management practices while the proportion of waterbody areas has remained relatively stable over time, albeit with some fluctuations, probably due to natural factors, such as rainfall patterns and evaporation, as well as human activities, such as water use. The predicted values for the land cover in the years 2025 and 2030 as well as the trend indicated that the non-forest land cover will increase significantly (b = −7.6, R2 = 0.72) while more water will be exposed to the surface (b = −3.9, R2 = 0.7). The forest will however significantly decrease (b = −12.5, R2 = 0.8) (Table 3).
Actual and predicted landcover change in the study area
| Landcover (area in km2) | Actual | Projected | Linear trend | R2 | ||||
|---|---|---|---|---|---|---|---|---|
| 1986 | 1999 | 2002 | 2014 | 2025 | 2030 | y = a ± bx | ||
| Forest | 56.08 | 72.34 | 55.64 | 20.82 | 15.0 | 10.0 | 82.04 − 12.5x | 0.80 |
| Secondary regrowth | 40.59 | 20.02 | 39.45 | 20.82 | 25.0 | 30.0 | 34.98 − 1.6x | 0.12 |
| Non forest | 2.76 | 7.93 | 4.95 | 4.99 | 35.0 | 40.0 | −10.81 + 7.6x | 0.72 |
| Waterbody | 1.40 | 0.35 | 0.79 | 0.31 | 15.0 | 20.0 | −7.34 + 3.9x | 0.70 |
| Landcover (area in km2) | Actual | Projected | Linear trend | R2 | ||||
|---|---|---|---|---|---|---|---|---|
| 1986 | 1999 | 2002 | 2014 | 2025 | 2030 | y = a ± bx | ||
| Forest | 56.08 | 72.34 | 55.64 | 20.82 | 15.0 | 10.0 | 82.04 − 12.5x | 0.80 |
| Secondary regrowth | 40.59 | 20.02 | 39.45 | 20.82 | 25.0 | 30.0 | 34.98 − 1.6x | 0.12 |
| Non forest | 2.76 | 7.93 | 4.95 | 4.99 | 35.0 | 40.0 | −10.81 + 7.6x | 0.72 |
| Waterbody | 1.40 | 0.35 | 0.79 | 0.31 | 15.0 | 20.0 | −7.34 + 3.9x | 0.70 |
Actual and predicted landcover change in the study area
| Landcover (area in km2) | Actual | Projected | Linear trend | R2 | ||||
|---|---|---|---|---|---|---|---|---|
| 1986 | 1999 | 2002 | 2014 | 2025 | 2030 | y = a ± bx | ||
| Forest | 56.08 | 72.34 | 55.64 | 20.82 | 15.0 | 10.0 | 82.04 − 12.5x | 0.80 |
| Secondary regrowth | 40.59 | 20.02 | 39.45 | 20.82 | 25.0 | 30.0 | 34.98 − 1.6x | 0.12 |
| Non forest | 2.76 | 7.93 | 4.95 | 4.99 | 35.0 | 40.0 | −10.81 + 7.6x | 0.72 |
| Waterbody | 1.40 | 0.35 | 0.79 | 0.31 | 15.0 | 20.0 | −7.34 + 3.9x | 0.70 |
| Landcover (area in km2) | Actual | Projected | Linear trend | R2 | ||||
|---|---|---|---|---|---|---|---|---|
| 1986 | 1999 | 2002 | 2014 | 2025 | 2030 | y = a ± bx | ||
| Forest | 56.08 | 72.34 | 55.64 | 20.82 | 15.0 | 10.0 | 82.04 − 12.5x | 0.80 |
| Secondary regrowth | 40.59 | 20.02 | 39.45 | 20.82 | 25.0 | 30.0 | 34.98 − 1.6x | 0.12 |
| Non forest | 2.76 | 7.93 | 4.95 | 4.99 | 35.0 | 40.0 | −10.81 + 7.6x | 0.72 |
| Waterbody | 1.40 | 0.35 | 0.79 | 0.31 | 15.0 | 20.0 | −7.34 + 3.9x | 0.70 |
Furthermore, a linear relationship between change in air temperature and the area occupied by each of the land use/cover types (Equations 1–3, P < .05) suggests that climate is an important factor in the degradation of the forested areas.
Where , , = change in the area (in hectares) occupied by forest, secondary regrowth/farmland, and non-forest features, respectively; x refers to the mean temperature (°C).
Discussion
The study area experienced significant decreasing rainfall and increasing temperature; including scenarios of rainfall fluctuations, including flood events, suggesting that apart from warmer climate, the study area has been vulnerable to the effects of excess wetness at certain periods. Occurrences of rainfall and temperature extremes have also been recorded in many parts of Nigeria [60, 61]. While studies [62, 63] argued that forests can reduce floods in developing nations, the condition in the study area indicated an inverse relationship between the size of different land uses/land cover (forest, secondary regrowth/farmland, and non-forest features) and mean air temperature in the area. Evidence from Adewoyin [64] also indicated that the study area has experienced extreme climate conditions, especially floods in the study period but there is yet no clear understanding of the ecosystem conditions in the forest reserve during the period of floods. In many parts of this region, the effects of climate variability on forest loss in protected areas is primarily obvious through changes in precipitation patterns, temperature fluctuations, and extreme weather events, leading to increased tree mortality and altered growth rates. However, anthropogenic activities such as agricultural expansion, farming, logging, and infrastructure development directly contribute to deforestation by reducing vegetation cover and habitat fragmentation. While a clear distinction between these impacts is critical for developing targeted conservation strategies to address natural and human-induced threats to forest ecosystems, it is important to note that the changes observed in this study may be influenced by a combination of factors, including climate change, human activities, and natural variability. Determination of the relative contribution of each factor will require further analysis, such as comparing the observed changes to climate data and historical land use information, but information about the latter was not sufficiently available at the time of this study. Nonetheless, future studies will also consider the specific drivers of deforestation and land conversion in Nigeria, such as population growth, economic development, and policy frameworks.
Van Dijk et al. [63] considered the relationship between flooding and forest ecosystems as an actively topical issue. They supported the hypothesis that changes in population density may explain about 83% of the variation in reported flood occurrences, significantly higher than the 10% reported in earlier studies. In this study, it is more likely that the observed reduction in the forested area has been largely caused by urbanization and lumbering activities which have been reported in a previous study [26]. The study revealed that the management strategy to keep the forest reserve as a nature reserve, which it was dedicated to, was threatened by poorly implemented strategies that probably became compromised by corruption. For example, the level of degradation is more obvious with the results of NDVI than that of the classified image; the NDVI highlights vegetation health through coarseness or smoothness, while classification would aggregate typically interlocked forests of tropical areas and therefore hide information about undergrowth in most medium to low-resolution imageries like Landsat [65, 66]. In addition, cases of herdsmen-farmer conflicts are common in southwestern Nigeria, but the present study area mainly witnessed conflicts arising from agroforestry practices, where farmers request to clear some parts of the forest for farming activities or otherwise plant their crops in between the tree plantations. The impact of this is the significant, and arguably underreported, losses in the vegetal component and a reduction in animal diversity of the forest reserve, especially in the face of population pressure and urbanization through land grabbing for built-up purposes and activities of loggers who often engage in illegal removal of economic trees. Consequently, it may be hypothesized that climate change and variability will impact drivers of urbanization, lifestyle/socio-economic issues, and lax management strategies or/and poor implementation in the region as they do in many other parts of West Africa.
Issues relating to lifestyle/socio-economic consideration are replete in the literature. Many studies have linked the extensive use of firewood for domestic purposes among many rural dwellers [19, 65–68] and its commercialization for industrial uses and domestic purposes (especially with recent increases in prices of petroleum products) as well as belief systems on the use of fuelwood in some urban communities to increased threats on the forests [69, 70]. Another lifestyle-related issue is that of purposeful (with the intent of animal gaming and reducing the cost of clearing farmlands) bush burning by ranchers and farmers in many rural areas, including in settlements close to forest reserves [71, 72]. All these influences often create an immediate impact and exacerbate the impact of climate variability on forest resources.
In the course of this investigation, we observed that many countries will not be able to adequately account for the impact of climate change due to limitations in climatic data. Nigeria as an example has a very coarsely distributed network of meteorological stations with significantly problematic access to climatic data, which are equally expensive, especially in a country with very limited access to research funds. Since the study area was not covered in the network of the existing meteorological network, the study essentially made use of satellite datasets, which also proved useful and replicable, especially for areas outside the conventional meteorological network as the present study area. The results of this study do not however suggest a lack of contingency plans to mediate the effects of climate change or extreme weather conditions. Many programmes exist at local, state, and federal government levels that are targeted at mitigating the effects of climate change in Nigeria but their degrees of implementation often vary with locations and government in power. For instance, of all the states in Nigeria, Lagos (which is outside the present study area) is the only state in Nigeria that has demonstrated visible progress with non-structural measures and emergency planning [73]. Given the government structure of Nigeria, the Federal Government is tasked with setting infrastructure standards, identifying high-risk infrastructure, building climate adaptation into a national masterplan, and developing policies to enhance capacity to mitigate the impact of climate change and extreme conditions. State (36 and the federal capital territory, Abuja) governments are expected to work in partnership with the federal government to develop infrastructure standards and develop risk assessment and response plans in collaboration with federal agencies and private sector providers while the local governments (there are 774 local government areas in Nigeria) are expected to link communities to households’ level with mandates from the federal government.
Despite the structured line of events to enhance climate adaptation, resilience, and response among citizens, studies have shown that the programmes have failed in most cases due to poor implementation and corrupt systems [74, 75]. Nonetheless, concerns have been raised among relevant research groups and institutions on tree death and disease infestation that have necessitated reforestation efforts, especially the reforestation efforts by the Forestry Research Institute of Nigeria and the States Ministry of Agriculture and Forestry to repopulate degraded areas of forest reserves in Nigeria [76, 77].
Conclusions
The study demonstrated the capacity of remote sensing-based datasets and geographic information systems in the assessment of variability in climate and land resources data to determine the impact of climate variability on forest change in a data-scarce environment. Given the pattern exhibited by the selected climatic data (i.e. the declining rainfall and increasing temperature trends) over the study period, as well as the degrading nature of the forest reserve, it is safe to conclude that the impact of climate change on the forest environment has been significantly exacerbated by pressure for urban growth and probably a poorly monitored quest for socio-economic improvement of the society.
Acknowledgements
The authors acknowledge the insightful comments of the anonymous reviewers that have improved the manuscript. We also appreciate the journal for the waiver
Author contributions
Damilola Grace Oanipon (Conceptualization [supporting], Data curation [supporting], Formal analysis [supporting], Investigation [supporting], Methodology [supporting], Project administration [supporting], Resources [supporting], Software [supporting], Supervision [supporting]), Adelowo Adefisayo Adewoyin (Conceptualization [supporting], Data curation [supporting], Formal analysis [supporting], Investigation [supporting], Methodology [supporting], Resources [supporting], Software [supporting]), and Adebayo Oluwole Eludoyin (Conceptualization [lead], Data curation [lead], Formal analysis [lead], Investigation [lead], Methodology [lead], Resources [lead], Software [equal], Supervision [lead], Validation [lead], Visualization [lead])
Conflict of interest: None declared.
Funding
This research received no external funding.
APC funding
There authors cannot afford the APC for open-access publication on their own but they rely on the magnanimity of the Developing Countries Initiative for full waiver, as all the authors are covered by this; APC waiver policy | Oxford Academic.
Data availability
The data underlying this article are available in public domains; Landsat Data Access | U.S. Geological Survey, and Find Data | NASA Earthdata.
Institutional review board statement
The study was conducted according to the guidelines of the Declaration of Obafemi Awolowo University, Ile-Ife, Nigeria, and approved by the Institutional Review Board (or Ethics Committee) of Department of Geography, Faculty of Social Sciences and Postgraduate College, Obafemi Awolowo University. The study did not directly involve handling humans or animals.
