48  

Geospatial modelling of Forest Canopy Density and Landscape 

Assessment in Omo Biosphere Reserve, South-western Nigeria 
 

Z.H. Mshelia1,2, A.I. Bamgboye3, M.A. Onilude4, O.J. Taiwo5 
 

1Institute of Life and Earth Sciences (including Health and Agriculture), Pan African University, 

University of Ibadan, Nigeria 
2Department of Environmental Modelling and Biometrics, Forestry Research institute of Nigeria, 

Ibadan 
3Department of Agricultural and Environmental Engineering, University of Ibadan, Nigeria 

4Department of Wood Product Engineering, University of Ibadan, Nigeria 
5Department of Geography, University of Ibadan, Nigeria 

 

Date Received: 23-04-2021   Date Accepted: 10-12-2021 

 

Abstract 

Forest has an important role in the global carbon cycle that covers over one-fourth of the 

world’s geographical area. It is one of the major natural resources and magnificent terrestrial 

ecosystems of the world. Forest Canopy Density (FCD) is imperative in the assessment of forest status 

and is a primary indicator of potential management interventions. Landsat images of 1990 and 2018 

were used in this study. Remote Sensing has demonstrated to be very cost-effective in mapping and 

monitoring changes in forests, and other environmental issues. Forest cover change and fragmentation 

were analysed using FCD and Landscape metrics. The FCD was obtained from the combination of 

data from the Advance Vegetation Density Index (AVI), Bare Soil Index (BI), and Forest Shadow 

Index (FSI). Four categories of change were identified in the reserve, no change, growth, degradation 

and deforestation. There was no change in 222.57 ha (52.98%), growth had 81.54 ha (0.69%), 

degradation with 116.01 ha (27.61%) and deforestation with the least change with 0.81 ha (0.19%). 

Degradation with a change rate of 0.97% contributed more in terms of change. There is a slight increase 

in the values of the three diversity indices (SHDI, SHEI, SIDI) while a high degree of homogeneity is 

recorded in the no forest class and the three others classes were fragmented. Understanding the 

dynamics of the forests is important in mitigating climate change and support for biological resources. 

 

Keywords: FCD, landscape metrics, deforestation, degradation, fragmentation 

 

1. Introduction 

The management of forests as carbon reservoirs could support the protection of biological 

resources such as water, soil, habitat, and raw materials, etc. (Thornley et al., 2000). Forest 

conservation and sustainable forest management are key in mitigating climate change at all scales. 

Farming, industrialisation, urbanisation and mining activities have caused the loss of large forest areas 

resulting in a high rate of deforestation and forest degradation. Forest areas, forest density and 

greenness of an area are major issues for the ecosystem, biodiversity and so on (Banerjee et al., 2014). 

Forest maps are an effective tool for identifying the state of forest resources and monitoring 

ongoing spatial processes in forested landscapes. One of the most important forest properties is the 

canopy cover and it provides habitats for many animal species (Akike et al., 2016). 

  

 
*Correspondence: zackmshelia@gmail.com 
© University of Sri Jayewardenepura 
 



Mshelia et al. /Journal of Tropical Forestry and Environment Vol. 11 No. 1 (2022) 48-66 

49 

Conventional remote sensing methodology is based on qualitative analysis of information 

derived from “training areas” (i.e. ground-truth). This has certain disadvantages in terms of the time 

and cost required for training area establishment, and the accuracy of the results obtained. In response 

to these problems, a new methodology was developed during ITTO Project PD 32/93 Rev. 2 (F), 

“Rehabilitation of Logged-over Forests in Asia-Pacific Region, Sub-project III” (Rikimaru et al., 

2002). The Forest Canopy Density (FCD) Mapping and monitoring Model utilizes forest canopy 

density as an essential parameter for the characterization of forest conditions. FCD data indicates the 

degree of degradation, thereby also indicating the intensity of rehabilitation treatment that may be 

required (Rikimaru, et al., 2002). Forest cover analysis is the first step in assessing forest 

fragmentation, as forest cover modifies the fragmentation pattern. There is link between forest 

fragmentation with forest cover changes (Gupta et al., 2018). 

Landscapes are spatial entities of the earth’s surface explicitly defined by their structure, 

function and composition. They are dynamic geographical units composed of various structured 

elements that interact at different scales and ranges (Rajendran et al., 2015). Unlike traditional the 

ecosystem concept, the landscape concept focus on spatial heterogeneity and its impact on the 

ecological processes. The ecological processes that maintain complex landscapes at one scale can be 

different from other scales. Understanding the dynamism of landscape characteristics is vital for 

ecological stability and biodiversity conservation (Rajendran et al., 2015). Remote Sensing has 

demonstrated to be very cost-effective in mapping and monitoring changes in forests, and other 

environmental issues (Wang et al., 2009). 

The focus of this study is to access the forest canopy density and landscape pattern of Omo 

biosphere reserve with specific objectives: (i) to examine the forest cover change between 1990 and 

2018 using the forest canopy density model; (ii) to examine the rate of forest cover change; (iii) to 

analyse the forest landscape characterisation using landscape metric model within the study area. 

 

2. Methodology 

2.1 The study area 

Omo Forest Reserve, which derives its name from River Omo that traverses it, is located between 

latitudes 6o 42' to 7o 05' N and longitude 4o 12' to 4o 35' E (Figure 1) Ogun state South-western Nigeria. 

Omo covers about 130 500 ha, which includes a 460 ha Strict Nature Reserve established in 1977 

known as Omo Biosphere (Okali and Ola-Adams, 1987). 

 



50 

 
Figure.01. Location of Omo Biosphere Reserve in Omo Forest Reserve South-western Nigeria 

 

The climate is tropical and it is characterized by wet (February to November) and dry (December and 

January) seasons. The temperature ranges between 21-34 °C while the annual rainfall ranges between 

150 and 3000 mm (Larinde et al., 2011; Adedeji et al., 2015). 

2.2 Data acquisition and analysis 

Landsat satellite images of 5th January 1990 (Landsat TM) and 19th January 2018 (Landsat 8 

OLI) in path 190 and row 55, were acquired from the official website of the United States Geological 

Survey (USGS). The satellite images obtained were subjected to radiometric calibration to adjust the 

data for use in quantitative analysis (Agbor et al., 2017). The images used in this study were first 

converted to Top of Atmosphere (TOA) radiance using equation 1 (Giannini et al., 2015). 
 



Mshelia et al. /Journal of Tropical Forestry and Environment Vol. 11 No. 1 (2022) 48-66 

51 

Lλ= (
(LMAXλ-LMINλ)

Q
CAL

λ
) Q

CAL
+LMINλ          (1) 

 

The above expression does not consider the atmospheric effects, therefore there is a need to 

convert images from radiance to reflectance measures, using equation 2 ((Giannini et al, 2015). 
 

2

Esun

* r*d

E *Cos
sz

TOA



 





            (2) 
 

 

 

Figure 02. Flow Chart of the Methodology 

  

Landsat ETM 1990 (30m)  and 

Landsat OLI 2018 (30m) 

Conversion from DN to Radiance and Reflectance 

Advance Vegetation Index 

(AVI) 

Bare Soil Index (BI) Canopy Shadow Index 

(SI) 

Principal Component Analysis 

(PCA) 

Vegetation Density (VD) 

Scale Vegetation Index (SVI) Scale Shadow Index (SSI) 

Forest Canopy Density (%) Land cover Classification 

map based on density  

Change Analysis and Landscape 

metrics computation  

Accuracy Assessment Projection of Forest cover 

classes 



52 

2.2.1. Forest Canopy Density Model 

The Forest Canopy Density model utilizes forest canopy density as an important parameter for 

the assessment of forest conditions. This model involves bio-spectral phenomenon modelling and 

analysis utilizing data derived from four indices (Azizia, 2008 and Akike, 2016): Advanced Vegetation 

Index (AVI), Bare Soil Index (BI), Shadow Index or Scaled Shadow Index (SI, SSI) and Thermal Index 

(TI) (Azizia, 2008 and Akike and Samanta, 2016). The four indices were calculated using equations 3 

to 8. 

2.2.2. Advanced Vegetation Index (AVI) 

This index was calculated using Equations 3 and 4 (Azizia, 2008 and Akike et al, 2016). 
 

3 B4)}-B4)(B5-1)(65536+ {(B6 = AVI for OLI (Landsat 8)       (3) 
 

Or 
 

3AVI =  {(B5 +1)(256-B3)(B5-B3)}
 for ETM (Landsat 5 or 7)       (4) 

 

2.2.3. Bare Soil Index (BI) 

BI was calculated using equations 5 and 6 (Akike et al., 2016 and Saei et al., 2000). 
 

( 6 4) ( 5 2)
*100 100

( 6 4) ( 5 2)

B B B B
BI

B B B B

  
 

    for OLI (Landsat 8)        (5) 
 

Or 
 

( 5 3) ( 4 1)
*100 100

( 5 3) ( 4 1)

B B B B
BI

B B B B

  
 

    for ETM (Landsat 5 or 7)       (6) 
 

2.2.4. Shadow Index (SI) 

SI was calculated using equations 7 and 8  
 

3SI =  ((65536 +B2) * (65536-B3) * (65536-B4))
for OLI        (7) 

 

Or 
 

3SI =  ((65536 +B1) * (65536-B2) * (65536-B3))
for ETM        (8) 

 

The source of thermal information is the infrared band of Landsat data (bands 6 and 10). Land 

Surface Temperature (LST) retrieval was carried out through three phases (Giannini et al., 2015). All 

the image bands are quantized as 8-bit data except Landsat 8 which is 16 bit, thus; all information is 

stored in DN which were then converted to radiance with a linear equation (9) given as: 

 

Y= mx + b              (9) 
 

Where: 

Y=TOAr (Top of Atmosphere) radiance-the radiance measured by the sensor 

m=Radiance multiplicative value 

x=Raw band 

b=Radiance additive value 

By applying the inverse of the Planck function, thermal bands radiance values were converted 

to a brightness temperature value using equation 10 (Giannini et al., 2015). This is satellite temperature 

in Kelvin. 
 



Mshelia et al. /Journal of Tropical Forestry and Environment Vol. 11 No. 1 (2022) 48-66 

53 

2

1
ln 1

k
BT

k

TOAr


  

  
             (10) 

 

Where: 

BT=º Kelvin 

TOAr=Top of Atmosphere radiance 

K1=calibration constant 1 (607.76 for TM), (666.09 for ETM+) and (774.89 for OLI band 10) 

K2=calibration constant 2 (1260.56 for TM), (1282.71 for ETM+) and (1321.08 for OLI band 10) 

Surface temperature=BT–273.15 

2.2.5. Vegetation density (VD) 

The principal component analysis was used to calculate the vegetation density (VD) by 

synthesizing Advanced Vegetation Index with the Bare Soil Index. The value was scaled from 0 to 

100%. The 100% shows the area of the high forest while the 0% indicate the areas of no vegetation 

(Rikimaru, 1996; Saei and Abkar, 2004). 

2.2.6. Scaled shadow index (SSI) 

SSI is was calculated from the Canopy shadow index (SI) by using a linear transformation. The 

value of SSI was also scaled from 0 to 100%. SSI by 100% responds with the highest possible shadow 

whereas 0% responds the opposite. SI is important in forestry and crop monitoring because the canopy 

shadow provides some information on tree and plants arrangement. 

2.2.7. Integration process to achieve FCD model 

Integration of VD and SSI means transformation for forest canopy density value. Both 

parameters have dimensions and have percentage scale units of density. It is possible to synthesize 

both indices safely by employing the corresponding scale and unit of each. FCD was calculated using 

equation 11. 
. 

( * 1) 1FCD SVD SSI  
                     (11) 

 

2.3. Landscape Metrics and Diversity Analysis  

Several studies in landscape ecology emphasized the use of spatio-temporal satellite data along 

with landscape metrics in landscape evaluation and policymaking. Remote Sensing data will be 

primarily utilized to create the necessary database for two time periods, 1990 and 2018. Landscape 

Metrics and Diversity Analysis. 

The LecoS plugin in Quantum GIS (QGIS) was used for the land metrics and diversity analysis. 

The result of the Forest Canopy Density model of 1990 and 2018 was the input images for the analysis. 

Shannon Diversity Index expresses, Simpson Diversity index and Shannon Evenness Index 

(Equitability) was used to determining the level of diversity and evenness in the Omo biosphere and 

the entire reserve. The degree of fragmentation and dominance or homogeneity was examined using 

the following indices Land Cover, Landscape proportion, Edge length, Number of Patches, Patch 

Density, Greatest patch area, Landscape division, Effective mesh size and Splitting index.  Calculated 

coefficients can be classified according to the type of evaluated characteristic into categories of indices: 

of shape, size, diversity, edges and proximity (Stejskalova et al., 2012). Statistically, many of the 

metrics are correlated and can be depicted in concise form according to the structural characteristics 

(Rajendran et al, 2015). Table 1 shows the indices, acronyms used and a short description of each 

indicator. 

  



54 

 

Table 01. Landscape metrics used in the study 

Metric Abbreviation Description 

Land Cover LC Equals the number of cells for each class based on a 

classified land cover matrix. The resulting values were 

multiplied by the cell’s value; (ha) 

Landscape 

proportion 

LP Landscape proportion (LP) quantifies proportional 

abundance of a certain class  in the total  landscape 

area(0<LP≤100); % 

Edge Length EL Equals the total length of all patches from a specific 

class.  The resulting values were, of course, multiplied 

with the cell’s value; (m). 

Edge Density ED Edge  Density  equals  the  sum  of  the  lengths  of  all  

edge segments involving the corresponding  patch type, 

divided by the total landscape area ; (m/ha) 

Number of 

Patches 

NP Express the number of patches identified for each class; 

(no.). 

Patch density PD Equals the number of patches of the corresponding 

patch type divided by total landscape area; (no. /100 

ha). 

Greatest patch 

area 

GPA Greatest Patch Area identifies the area under the single 

largest patch in a given landscape. It is a measure of 

dominance i.e. degree of homogeneity 

Mean Patch 

area 

MPA The mean Patch area serves as a fragmentation index. 

A landscape with a smaller mean patch area for the 

target patch type than another landscape might be 

considered more fragmented. 

Over all Core 

area 

OC Total core area (ha) or the percentage of the landscape 

comprised of core area at the class or landscape level. 

The core area is a compound measure of shape, area 

and edge depth 

Landscape 

division 

LD Landscape Division is defined as the probability that 

two randomly chosen places in the landscape to be 

found in the same patch. 

Effective 

mesh size 

m The probability that two randomly chosen cells are 

connected (to be included into the same patch); (ha). 

Splitting 

index 

S The number of patches one gets when dividing the total 

region into parts of equal size in such a way that this 

new configuration leads to the same degree of 

landscape division desired; (nr.). 

Shannon’s 

Diversity 

Index 

SHDI Based on information theory; represents the amount of 

"information" per individual (or patch type, in this 

case); larger values indicate a greater number of patch 

types and/or greater evenness among patch types. 

Shannon 

Equitability 

Index 

SHEI Shannon Equitability (Evenness) Index expresses the 

dominance of patches within the total area. 



Mshelia et al. /Journal of Tropical Forestry and Environment Vol. 11 No. 1 (2022) 48-66 

55 

Simpson 

Diversity 

Index 

SIDI Simpson Diversity Index represents the probability that 

any two pixels selected at random would be different 

patch types. The larger the value the greater the 

likelihood that any 2 randomly drawn cells would be 

different patch types 

 

3. Results 

3.1 Forest Cover Change Analysis 

The results of the Advance Vegetation Index (AVI), Bare Surface Index (BSI), Shadow Index 

(SI), Thermal Index (TI) and the Forest Canopy Density (FCD) for 1990 and 2018 are presented in 

Figures 03, 04, 05, 06 and 07 respectively. The Advance Vegetation Index (AVI) shows a positive 

relationship with the quantity of vegetation, which means the number of vegetation increases as the 

value of the AVI increases (Figure 03). The colours green and yellow are areas covered by vegetation 

while the colour red shows areas of non-forest which was mosaic of water body, rock outcrop and open 

space. As shown in Figure 4. Bare Surface Index (BSI) increases as the percentage of bare soil exposure 

of ground increases. The BSI showed the colour red as non-forest to determine the effect of soil 

exposure in the analysis. This index helps in separating the vegetation with a different background. 

This result showed that there were no remarkable changes in soil percentage over the years. The SI 

was utilised for spectral information on the forest shadow itself and thermal information on the forest 

influenced by the shadow. The SI takes care of the cooling effect inside the forest and evaporation 

from the leaf structure. The tree arrangement has a low shadow casting than matured tree arrangement. 

The result in Figure 5 showed that there were no remarkable changes in shadow percentage created by 

tree canopy over the years. Inside the forest stand, the canopy cover blocked the incoming solar 

radiation and it is the reason for the cool temperature inside the forest. The soil area is characterised 

by high temperatures. The combination of the Thermal Index (TI) with Shadow Index (SI) was helpful 

in black soil detection. The result shows that changes in the land surface temperature over the years in 

the Omo biosphere was insignificant (Figure 06). The Forest Canopy Density model integrates these 

indices to take into account all factors responsible for deficiency in using a single vegetation index to 

analyse forest cover. 

The AVI and BSI both have a negative relationship with each other, a high AVI value shows 

high vegetation vigour, and similarly, high BSI shows soil exposure. Utilising the various spectral 

indices, vegetation density and scale shadow index, the forest canopy density map was produced for 

the years 1990 and 2018. It was thereafter utilised for the classification of the forest cover and its 

change detection. 

Based on the percentage, each pixel was classified into four classes of forest canopy density: 

high forest density, mid forest density, low forest density and no forest. High forest density is an area 

having a value from 71% to 100%. In the same manner, 41-70%, 5-40% and below 5% were areas 

with mid forest density, low forest density, and no forest respectively (Figure 07). The maps described 

the distribution of forest resources in Omo Biosphere through the FCD model. 

The statistics in Table 02 revealed that from 1990 to 2018, the no forest area decreases by 2.7 

ha at the rate of 0.023% in Omo Biosphere being a strict nature reserve, which could be a result of 

forest regeneration. The changes in low forest areas are insignificant as the difference between 1990 

and 2018 was 0.36 ha. The mid forest density had a significant increase of 42.66 ha at 0.36% change 

rate from 39.2% to 49.3% while the high-density area was the victim of the great increase in the mid 

forest density. It reduces from 221.58 ha to 181.98 ha at 0.34 change rate from 52.58% to 43.32%. 

 



56 

 
Figure 03. Advance Vegetation Index of 1990 (a) and 2018 (b) of Omo Biosphere 

 

 
Figure 04. Bare Soil Index of 1990 (a) and 2018 (b) Omo Biosphere 

 



Mshelia et al. /Journal of Tropical Forestry and Environment Vol. 11 No. 1 (2022) 48-66 

57 

 
Figure 05. Shadow Index of 1990 (a) and 2018 (b) of Omo Biosphere 

 

 
Figure 06. Thermal Index of 1990 (a) and 2018 (b) of Omo Biosphere 



58 

 
Figure 07. Forest Canopy Density of 1990 (a) and 2018 (b) of Omo Biosphere 

 

Table 02. Area of Forest Cover Classes of Omo Biosphere Reserve 

Class 1990 

Area 

(ha) 

% 2018 Area 

(ha) 

% Area Diff. 

(ha) 

Change Rate 

(%) 

No Forest 8.64 2.057 5.94 1.414 -2.7 0.023 

Low Forest Density 25.56 6.084 25.2 5.998 -0.36 0.003 

Mid Forest Density 164.34 39.117 207 49.271 42.66 0.362 

High Forest Density 221.58 52.742 181.98 43.316 -39.6 0.336 

Total 420.12 100 420.12 100 _ _ 

 

 
Figure 08. Area of Forest Density Classes of Omo Biosphere 

0

50

100

150

200

250

No Forest Low Forest

Density

Mid Forest

Density

High Forest

Density

A
r
e
a

 (
h

a
)

1990 Area (ha) 2018 Area (ha)



Mshelia et al. /Journal of Tropical Forestry and Environment Vol. 11 No. 1 (2022) 48-66 

59 

3.2. Accuracy Assessment 

Accuracy assessment of the classified image has been verified from field verification data Table 

03. The overall accuracy for classification was 90% and 92.8% and Kappa statistics of 86.2% and 

90.2% for 1990 and 2018 respectively. 

Table 03. Kappa Statistics of the classified images 

 1990 2018 

Class Name Pa Ua Pa Ua 

No Forest 100 100 81.8 90 

Low Forest Density 86.67 81.25 93.3 93.3 

Mid Forest Density 88 91.67 95.8 92 

High Forest Density 90 90 95 95 

Overall Accuracy 90 92.8 

Kappa Statistics (k) 86.2 90.2 

*Pa=Producer’s accuracy 

*Ua=User’s Accuracy 

 

3.3. Forest Cover Change Detection of Omo Biosphere Reserve 

Classified forest canopy density images in 1990 and 2018 were cross-classified and a tabular 

matrix (Table 4) shows the number of pixels that correspond to each combination of categories in the 

two (column-row) images was compared. The tabular matrix is expressed in terms of the proportion of 

the total number of pixels. From Table 5, four categories of change were identified in the Omo 

biosphere; no change, growth, degradation and deforestation. There was no change in 222.57 ha 

(52.98%) of the biosphere, no forest to low forest density, low forest density to mid forest density and 

mid forest density to high forest density growth occurred in 81.54 ha at a change rate of 0.69. 116.01 

ha of the forest were degraded from high-density forest to mid-density and from mid-density to low 

density, which is 27.61% of the reserve. Figure 10a showed the change map of the biosphere from 

1990 to 2018. The colour blue on the map represent areas that were no forest, low forest, mid forest 

and high forest densities in 1990 and are still the same in 2018 while the colour black on the map 

represent areas that were changed from one class to another. Figures 10b, c and d were the different 

changes that the forest went through for 28 years. Figure 10b showed the portion of the reserve that 

grows from no forest area to low forest, mid forest and high forest while Figure 10c showed the areas 

that went through degradation from high forest to mid forest and low forest. Figure 10d showed a little 

portion of the reserve toward the eastern part of the biosphere that was deforested, that is, moving from 

low forest to no forest class. 



60 

 
Figure 09. Forest Canopy Density Change from 1990-2018 in Omo Biosphere 

 

 
Figure 10a-d. Spatial Distribution of Changes of the four Classes (No Change (a), Growth (b), 

Degradation (c) and Deforestation (d) from 1990-2018 in Omo Biosphere 

  



Mshelia et al. /Journal of Tropical Forestry and Environment Vol. 11 No. 1 (2022) 48-66 

61 

Table 04. Cross-Classification Change of 1990 and 2018 in Omo Biosphere 

 

Table 05. Summary of the Area Cross-Classification Change of 1990 and 2018 in Omo Biosphere 

 

3.4. Landscape Metrics and Diversity Analysis 

Table 06 shows the values resulting from the calculation of global indices performed for the 

two periods. In terms of diversity, there is a slight increase and decrease of the values of the three 

indices (SHDI, SHEI, SIDI) in the biosphere and the entire reserve, as a result of the increase of some 

classes are reported to the general distribution of the landscape. The slight decrease of the diversity 

values can be explained by an increase of the mid forest density which accounts for 49.3% of the 

biosphere (Table 02). However, the values of diversity and evenness remain relatively high, suggesting 

that the study area, which has favourable physical and geographical conditions, has a complex 

landscape with certain dominant species. 

In terms of landscape configuration, features and functionality, some other landscape indices 

were calculated (Table 07). Unlike diversity indices, these were applied particularly to each class. 

Table 01 presents the indicators, the abbreviation used and a short description for each type. 

Land Cover (LC) and Landscape Proportion (LP)-Significant changes were observed in the 

four classes both in the biosphere especially in mid forest density and high forest density classes. The 

edge length (EL) and edge density (ED)-The result showed a decrease of the two indices values for 

most classes, suggesting a tendency of landscape homogenization in the biosphere and an increase of 

the two indices values for most classes tending towards heterogeneity. 

Number of Patches (NP) and Patch Density (PD)-A significant decrease was observed in mid 

forest density (MFD) of the biosphere from 120 to 47. This explained why MFD occupied 49% of the 

biosphere. All the classes in the Omo forest reserve had a decrease in PD except the no forest. Patch 

Density reflects the extent of landscape fragmentation and therefore crucial for landscape structure 

assessment. Comparison of classes with varying sizes showed decreasing PD in most of the classes, 

especially in the entire reserve. However, the rate of the decrease is moderate, thereby making the level 

Category Area (ha) % Change Rate 

No Change 222.57 52.978  

Growth from NF- LF-MD-HF 81.54 19.409 0.693 

Degradation from HF-MD-LF 116.01 27.614 0.986 

Deforestation from LF-NF 0.81 0.193 0.007 

Total 420.12 100  

Class 1900 Class 2018 Changes Number of Pixels Area (ha) % 

No Forest No Forest NF-NF 57 5.13 1.2211 

No Forest Low Forest Density NF-LF 24 2.16 0.5141 

No Forest Mid Forest Density NF-MF 10 0.9 0.2142 

No Forest High Forest Density NF-HF 5 0.45 0.1071 

Low Forest Density No Forest LF-NF 9 0.81 0.1928 

Low Forest Density Low Forest Density LF-LF 51 4.59 1.0925 

Low Forest Density Mid Forest Density LF-MF 169 15.21 3.6204 

Low Forest Density High Forest Density LF-HF 55 4.95 1.1782 

Mid Forest Density Low Forest Density MF-LF 137 12.33 2.9349 

Mid Forest Density Mid Forest Density MF-MF 1046 94.14 22.408 

Mid Forest Density High Forest Density MF-HF 643 57.87 13.775 

High Forest Density Low Forest Density HF-LF 68 6.12 1.4567 

High Forest Density Mid Forest Density HF-MF 1075 96.75 23.029 

High Forest Density High Forest Density HF-HF 1319 118.71 28.256 



62 

of fragmentation insignificant for now. It was further explained following also the values of edge 

density. Edge density, with patch number and patch density, are representative for establishing the 

fragmentation degree of the landscape. The values obtained for the fragmentation (NP and, 

consequently, PD and ED) reveal a decrease in the study area’s fragmentation degree, inducing a 

clustering tendency. 

The greatest patch area (GPA) is related to the degree of homogeneity or dominance of the 

landscape. The results in Table 8 show that the high forest density (HFD) class with 1,244,700 m has 

the highest GPA in 1990 while mid-density forest (MFD) has the highest for 2018 with 1,748,700 m 

in the biosphere. The mean patch area is also higher for the categories mentioned above. 

Landscape division, Effective mesh size and Splitting index (LD, m and s) are interconnected 

and measure the fragmentation degree of the landscape. These indicators were introduced by (Jaeger, 

2000), as a result of the criticism over the simple measurements such as patch number or patch density, 

which presents some limitations for certain phases of the fragmentation process. They have the 

advantage, unlike other conventional indicators, that any omissions or additions of other small-sized 

patches do not influence the final result. 

In this study, values of LD for all classes are high (above 0.9), reflecting a high degree of 

fragmentation of class types. Although landscape division and Mesh are perfectly correlated, inversely, 

both metrics are included because of the differences in units and interpretation. Split (s) is based on 

the cumulative patch area distribution and is interpreted as the effective mesh number, or several 

patches with a constant patch size when the corresponding patch type is subdivided into S patches, 

where S is the value of the splitting index (McGarigal et al., 2002). Jaeger (2000), defines the splitting 

index as the number of patches that resulted after dividing the total area into equal size parts so that 

this new configuration leads to the same degree of landscape division (LD). When its value is 1, the 

landscape is represented by a single patch, the value increasing as the landscape is divided into several 

patches. Considering these aspects, the interpretation of the result must take into account the 

correlation of these three complementary indicators. 

The resulting values of the three indicators suggest different degrees of fragmentation for each 

class. Thus, the areas with a high degree of homogeneity are represented by MFD with a change record 

of 75,130.67 m in 1990 to 730,376.51 m in the Omo biosphere. In the entire reserve, a high degree of 

homogeneity is recorded in the no forest class and the three others classes were fragmented. 

 

Table 06. Landscape Diversity Indices between 1990 and 2018 of Omo Biosphere 

Metric  1990 2018 

Shannon’s Diversity Index 1.249 0.941 

Shannon Equitability Index 0.776 0.678 

Simpson Diversity Index 0.685 0.566 

 

  



 

61  

Table 07. Landscape Metrics Computed for Class Types for 1990 and 2018 in Omo Biosphere 

Class LC (m) LP (%) EL (m) ED NP PD GPA (m) MPA (m) LD m (m) s 

NF 1990 86,400 0.012 5,940 0.0009 15 2.17E-06 34,200 5760 1 229.24 30,210.73 

NF 2018 59,400 0.014 4,800 0.0011 10 2.38E-06 26,100 5,940 1 211.31 19,881.59 

LFD 1990 25,5600 0.037 26,160 0.0038 123 1.78E-05 19,800 2,078 1 197.19 3,512,0.42 

LFD 2018 252,000 0.06 24,120 0.0057 103 2.45E-05 27,000 2,447 1 450.39 9328.01 

MFD 1990 1643,400 0.237 111,360 0.0161 120 1.73E-05 691,200 13,695 0.99 75130.76 92.18 

MFD 2018 2,070,000 0.493 106,500 0.0253 47 1.12E-05 1,748,700 44,043 0.83 730376.6 5.75 

HFD 1990 2,215,800 0.32 99,300 0.0143 66 9.53E-06 1,244,700 33,573 0.96 300081.4 23.08 

HFD 2018 1,819,800 0.433 91,200 0.0217 85 2.02E-05 722,700 21,409 0.95 223219.3 18.82 

 

63 



 

62  

4. Discussion 

The knowledge of land use and land cover is important for many planning and management 

activities as it is considered an essential element for modelling and understanding the earth feature 

system (Akike et al., 2016). The result above reveals the capacity of the biosphere to sequestered 

carbon because the high and mid forest densities have the highest area occupied by forest over the 

years. Although, the result showed the biosphere suffered forest degradation rather than deforestation 

which can be attributed to logging or physical factors of rainstorm uprooting aged trees as noticed 

during the fieldwork. After logging, natural regeneration of a secondary forest takes place and can 

revert carbon (C) losses to recovering C stocks at a certain point in time (Henry et al., 2011). Very few 

studies reported the natural capacity of African ecosystems to regenerate after perturbations. Kotto‐

Same et al., (1997) reported that about 74 % of aboveground C was regenerated after a period of fallow 

of 18 years. According to the regression between the time of fallow and C stocks, the total system C 

would be equal to the original natural forest after 24 years (Kotto‐Same et al., 1997). In a semi‐

deciduous forest of Ghana that was logged in the 1950s, inventory of forest with three types of forest 

management revealed that there was no significant difference in aboveground biomass among a natural 

protected forest, a forest that was logged in the 1950s and forest that was under tree shelter‐wood 

system in the 1950s (Bombelli et al., 2009). Henry et al., (2011) reported that the increasing pressure 

of the timber companies and the farming activities often does not allow the forest to regenerate. But it 

is very difficult to estimate this. Extensive human encroachment, fragmentation and conversion of 

natural forest habitat quickly transformed the natural landscape into a cultural landscape at the foot of 

Bhutan and Bengal (Chamling et al., 2020).  

The degree of landscape fragmentation is an important environmental indicator in the fields of 

biodiversity and sustainable development. In addition, information on the degree of landscape 

fragmentation is relevant in regional planning and for decisions about infrastructure placement or 

removal. Its analysis on different time series shows how strong the current trends are and what their 

direction is (Jaeger et al., 2006). This fundamental research will definitely help to make policy framing 

holistic management approach. 

 

5. Conclusion 

Forest conservation and sustainable forest management is key in mitigating climate change at 

all scales. The FCD model and landscape metrics analysis using the Remote Sensing and GIS 

technologies to estimate forest loss and degradation in this study has been noted to be a time and cost-

effective method. 

The result showed that the rate of forest degradation is more than deforestation in the reserve. 

A high degree of homogeneity is recorded in the no forest class and the three others classes were 

fragmented while values of diversity and evenness remain relatively high, suggesting that the study 

area, which has favourable physical and geographical conditions, has a complex landscape with certain 

dominant types. 

The study determined that forest canopy density and diversity and fragmentation model are 

important for analysing the Spatio-temporal condition of the forest. This will help forest managers in 

decision making, monitoring biodiversity and conversation planning for sustainable forest 

management. Given the importance of natural forested areas and the pursuit to maintain sustainable 

forest management in the midst of growing natural resources extraction and other anthropogenic 

activities, the use of the FCD model and landscape ecology analysis to assess, estimate and quantify 

forest is recommended for use in forest management activities as it has proven to be a time and cost-

effective method. 

  

64 
64 



Mshelia et al. /Journal of Tropical Forestry and Environment Vol. 11 No. 1 (2022) 48-66 

63 

 

Acknowledgements 

This work is supported financially by the African Union under the Pan African University scholarship. 

 

References 

Alexiades, M.N. 1996. Collecting ethnobotanical data: an introduction to basic concepts and 

techniques. Advances in Economic Botany, 10:53-94. 

Anderson, T. 1860. Acanthaceae. In: Thwaites, G.H.K. (Ed.) Enumeratio Plantarum Zeylaniae 3. 

Dulau & Co., London: 161-240.  

Ashton, P.S. & Gunatilleke, C.V.S. 1987. New light on the plant geography of Ceylon. 1. Historical 

plant geography.  Journal of Biogeography, 14:249-285. 

Blume, C.L. 1826. Bijdragen tot de Flora van Nederlandsch Indie 14. Batavia. Ter Lands Drukkerij, 

Batavia:781. 

Bremekamp, C.E.B. 1944. Materials for a monograph of the Strobilanthinae. Verhandelingen der 

Nederlandsche Akademie van Wettenschappen Afdeeling Natuurkunde, Tweede Sectie, 

41(1):1-305. 

Carine, M.A., Jayasekara, P. and Scotland, R.W. 2000. A New Species of Strobilanthes Blume 

(Acanthaceae) from Sri Lanka. Kew Bulletin, 55(4):971-976. 

Clarke, C. B. 1884. Acanthaceae. In: Hooker, J.D. Flora of British India, 4. L. Reeve & Co., London: 

429-477. 

Deng, Y. F. 2019. Transfer of the Philippine species of Hemigraphis Nees to Strobilanthes Blume 

(Acanthaceae). Phytotaxa, 404 (5):1-3. 

Hu, J.Q., Deng, Y.F. & Wood, J.R.I. 2011. Strobilanthes. In: Wu, Z.Y., Raven, P.H. & Hong, D.Y. 

(Eds.) Flora of China 19. Science Press,Beijing & Missouri Botanical Garden Press, St. 

Louis:381-429. 

Janzen, D.H. 1976. Why bamboos wait so long to flower. Annual Review of Ecology and Systematics, 

7:347-391. 

Josekutty, E.J., P. Biju, J. R.I. Wood and J. Augustine. 2018 Strobilanthes kannanii–a new species of 

Acanthaceae from the Western Ghats, India. Nordic Journal of Botany, 36:01689. 

Mabberley D.J. 2017. Mabberley’s Plant-book: a portable dictionary of plants, their classification and 

uses. 4th edition. Cambridge University Press, Cambridge:1102. 

Mascarenhas, M.E. and M.K. Janarthanam. 2013. Taxonomic status of Strobilanthes warreensis 

(Acanthaceae), an endemic species of Western Ghats, India. Rheedea, 23:1-6. 

Nees von Esenbeck, C.G. 1832. Acanthaceae Indiae orientalis. In: Wallich, N. (Ed.) Plantae Asiaticae 

rariores 3. Treuttel, Würtz & Ritter, London:70-117. 

Nilanthi, R.M.R., Samarakoon, H., Jayawardana, N., Wijesundara, S. and Bandaranayake, P.C.G.  

2021 Strobilanthes medahinnensis (Acanthaceae) a new species, based on morphological and 

molecular data, from the Peak Wilderness Nature Reserve, Sri Lanka. Phytotaxa, 514(1):26-

38. 

Patil, S. 2021. Strobilanthes pushpagiriensis (Acanthaceae): a new species from Western Ghats, India. 

Nordic Journal of Botany, 39(5). 

POWO 2019. Plants of the World Online. Royal Botanic Gardens, Kew 

<www.plantsoftheworldonline.org/>. 

Somasekaram, T. 1988. National Atlas for Sri Lanka. Survey Department Colombo, pp. 1-141. 

Vahl, M. 1794. Symbolae Botanicae. 3. Moller, Copenhagen: 147. 

Venu, P. 2006. Strobilanthes Blume (Acanthaceae) in Peninsular India. Botanical Survey of India, 

Kolkata:216. 

Wood, J.R.I. and Scotland, R.W. 2009 New and little-known species of Strobilanthes (Acanthaceae) 

from India and South East Asia. Kew Bulletin, 64:3-47. 

Wood, J.R.I. 1994. Notes relating to the flora of Bhutan: XXIX. Acanthaceae, with special reference 

to Strobilanthes. Edinburgh Journal of Botany, 51(2):175-273. 

65 



64 

Wood, J.R.I. 1995. Notes on Strobilanthes for the flora of Ceylon. Kew Bulletin, 50(1):1-24. 

Wood, J.R.I. 1998. Strobilanthes. In: Dassanayake, M.D. (Ed.), A revised handbook to the flora of 

Ceylon. A.A Balkema, Rotterdam:30-74. 

Wood, J.R.I. 2014. New Names and Combinations in Indian Acanthaceae Novon: A Journal for 

Botanical Nomenclature, 23(3):385-395. 

Wood, J.R.I., Borah, D. and Taram, M. 2021. The rediscovery of Strobilanthes tubiflos (Acanthaceae) 

in north east India. Kew Bulletin, 76:333-338. 

66