Indonesian Journal on Geoscience Vol.
12 No.
3 December 2025: 467-480
INDONESIAN JOURNAL ON GEOSCIENCE
Geological Agency Ministry of Energy and Mineral Resources Journal homepage: h ps://ijog.
ISSN 2355-9314, e-ISSN 2355-9306 Spectral-Wise Method Derived by Optical Images For Phosphate Mineral Exploration In Mamuju.
West Sulawesi.
Indonesia Rajab Kurniawan1,2 and Asep Saepuloh1 Geological Engineering Study Program.
Faculty of Earth Sciences and Technology.
Bandung Institute of Technology.
Jl.
Ganesha No.
Bandung.
West Java.
Indonesia Department of Exploration and Mine Development.
PT.
Trimegah Bangun Persada Tbk.
Jakarta.
Indonesia Corresponding author: razens03@gmail.
Manuscript received: November, 08, 2024.
revised: March, 29, 2025.
approved: September, 24, 2025.
available online: November ,20, 2025 Abstract - Phosphates are predominantly used in the manufacture of fertilizers for crop nutrition and the production of animal feed supplements.
Only 10 Ae 15 % of the world phosphate rock production has other uses, such as pharmaceuticals, ceramics, an important alternative source of rare earth elements (REE), and Lithium-Iron-Phosphate (LFP) In Indonesia, various types of rocks are composed of phosphate components, but rocks containing economic value are limited.
They are generally found in sedimentary or biogenic phosphate deposits, whereas phosphate resources derived from igneous rocks are still undiscovered.
Therefore, regionally accurate mapping using spectral-wise methods is proposed in this study.
The five spectral references originated from phosphate minerals from the USGS spectral library were used: fluorapatite, chlorapatite, hydroxylapatite, monazite, and xenotime.
The spectral-wise aims to identify anomalies of phosphate minerals based on the behaviour of spectral absorption at wavelength 2 AAm.
Their spectra showed a similar absorption pattern at a wavelength of 2 AAm originating from orthophosphate (PO.
This spectral-wise formulation was then applied to detect the phosphate occurrences using the Landsat 9 image.
According to the proposed method, three PO4 prospect zones could successfully be detected covering 2,604 Ha with P2O5 content 14 2.
73 %.
It has been verified that the proposed method has an accuracy of about 70 % compared to rock samples containing P2O5 at the field study.
Keywords: Mamuju, phosphates.
Landsat 9 image, spectral-wise method
A IJOG - 2025
How to cite this article:
Kurniawan.
and Saepuloh.
, 2025.
Spectral-Wise Method Derived by Optical Images For Phosphate Mineral Exploration In Mamuju.
West Sulawesi.
Indonesia.
IndoAnesian Journal on GeoAscience, 12 .
, p.
DOI:10.
17014/ ijog.
Introduction Various rocks have phosphate-containing components such as phosphorite, francolite, and carbonatite (Ptacek, 2.
However, the rocks containing phosphate with economic value for mining materials are limited (Steiner et al.
Phosphate is commonly used as an industrial raw material.
There are about 85 % raw phosphate material used for plant fertilizers and animal supplements (Ptacek, 2.
The remaining usability is for industrial purposes such as pharmaceuticals, ceramics, textiles, and an alternative source for rare earth elements (Ptacek.
The phosphate is also used for LithiumIron-Phosphate (LFP) electric vehicle batteries (Xu et al.
, 2.
Phosphate deposits derived from igneous rocks, which are mostly found in carbonatite or other types of alkaline intrusions, are generally low grade, but high quality with Indexed by: SCOPUS
PUBLISHED IN IJOG
Indonesian Journal on Geoscience.
Vol.
12 No.
3 December 2025: 467-480 low content of unwanted contaminants such as Cd.
Pb.
As.
U (Ptacek, 2.
In Indonesia, rock containing phosphate minerals is generally found in the sedimentary or biogenic phosphate deposits, whereas phosphate resources derived from igneous rocks are still undiscovered (Ptacek, 2.
Therefore, phosphate resources derived from igneous rocks become an interesting exploration target to be researched and sought after.
The Apatite [Ca5(PO.
3(OH.
] is the most abundant phosphate mineral of all phosphorus in the earth crust (Ptacek, 2.
The apatite derived from intrusive igneous rocks, one of which is associated with alkaline rocks (Ptacek, 2.
The alkaline rocks are rich in alkali metals and feldspathoids .
nepheline, leucit.
(Abdullah et al.
, 2.
Primary apatite from igneous rock sources originated from chlorapatite, fluorapatite, or hydroxilapatite varieties (Ptacek, 2.
Monazite is a phosphate mineral containing light rare earth elements (LREE) found in a variety of rock types and also radioactive sources due to the presence of thorium and uranium (Ptacek, 2.
Xenotime is a rare earth phosphate mineral with the main component is yttrium orthophosphate (YPO.
known as a heavy rare earth element (HREE).
The association of phosphate elements with rare earth elements and radioactivity was reported by a previous study (Indrastomo et al.
Ciputra et al.
, 2.
Southwest Mamuju District.
West Sulawesi Province.
Indonesia, has been selected with a region of interest (ROI) of 30y30 km as the studied area (Figure.
The studied area is composed of leucite-bearing lapilli tuffs, volcanic breccias, leucite basalt fragments, and leucite basalts of Adang Formation (Ratman and Atmawinata, 1993.
in Shaban et al.
, 2.
The leucite mineral was found in the phenocrysts of basalt rocks indicating feldspathoid lava with a high potassium content (Indrastomo et al.
, 2.
The abundance of feldspathoid minerals is interpreted as an indication of phosphorus-bearing igneous rock.
The purpose of this study is to determine the equation model for the presence of phosphate minerals based on spec- tral absorption in the visible (VIS) - short wave infrared (SWIR) range, then to apply the equation model to detect the presence of phosphate minerals in the Mamuju area, and to analyze geological controls on the distribution of phosphate minerals.
Remote sensing using multispectral images with conventional methods commonly used for mineral identification is the band ratio equation.
This method uses the comparison between peaks and absorb reflectance in the spectral signature of the target mineral with the approach of the wavelength values of the bands in the multispectral image (Landsat.
ASTER, etc.
The drawback of this method is that for some spectral minerals the wavelength position of the multispectral image bands does not always represent the peak or absorb reflectance values of the spectral signature.
So, the ratio equation results can not represent the specific target mineral and cause the resulting anomalous area distribution to be overestimated.
In this study, the application of the band ratio reveals that the target mineral, specifically the phosphate mineral apatite, which indicates the presence of phosphate, can not have its peak reflectance and absorption accurately represented by the wavelengths of the Landsat 9 image bands This is because the absorption pattern falls between these bands.
Therefore, a specialized method is required to calculate the spectral absorption value of the phosphate mineral and convert it into the Landsat 9 image.
The spectral-wise method can be used to identify the phosphate element as a specific target mineral based on its spectral characteristics.
this case, the spectral characteristics of phosphatebearing minerals are compared, so that it can be known that uniform absorption patterns will lead to the same element.
This method can identify the absorption pattern between the wavelengths of the bands on Landsat 9.
Then, the relationship between the reflectance value of the absorption and the two bands between them is illustrated to form an angle that can be calculated and converted into an angular index value that can be adjusted on the Landsat 9 image.
The workflow in this research is divided into three stages, starting from the data preparation PUBLISHED IN IJOG Spectral-Wise Method Derived by Optical Images For Phosphate Mineral Exploration In Mamuju.
West Sulawesi.
Indonesia (R.
Kurniawan and A.
Saepulo.
5 km INDEX MAP
Mamuju
District
LANDSAT 9 IMAGE
SHOWING THE STUDY AREA
WITH NATURAL COLOUR
COMPOSITE
Sulawesi Island Subdistric Boundary 725000 .
Figure 1 .
The studied area is about 30y30 km in The Southwest Mamuju District.
West Sulawesi Province.
Indonesia, on the natural colour composite image of Landsat 9 (R.
B=Band 4,3,.
stage including literature studies, secondary data collection in the form of spectral signature (SS) of five phosphate minerals.
Landsat 9 images, reference field sampling data results from previous studies, digital elevation model (DEM), and regional geological maps.
In this stage, the spectral signature of the five minerals is overlaid on the scatter diagram, resulting in similar absorption at a wavelength of 2 m, which is interpreted as the absorption of orthophosphate (PO.
Atmospheric correction was performed on the Landsat 9 images to obtain surface reflectance from the satellite images, then raster extraction of the corrected images was performed focusing only on the 30y30 km studied area.
The national digital elevation model (DEMNAS) and Mamuju sheet regional map were used to identify primary phosphate mineralization pathways based on lineament analysis and geological structure interpretation.
The analysis stage combines the data from the preparation stage process including the analysis of the absorption angle at a wavelength of 2 m using the formula for calculating the angle in the triangle concept which produces the gamma angle value ().
This value will be plotted on the Landsat 9 images using the equation formula to see the distribution of pixels that are indicative of phosphate minerals.
The pixel density is calculated using the 2y2 km gridding method, and then kriging interpolation is used to determine the distribution From these steps, the zones indicating phosphate mineral anomalies (PO Zone.
were Conformity analysis with vegetation response and geological control of PO zones was conducted to obtain phosphate mineral anomalies.
The final stage is verification and validation, by ground checking or taking several rock or soil samples in the predetermined anomaly zones.
The samples were analyzed at The Intertek Laboratory using X-ray fluorescence (XRF) and Inductively Coupled Plasma-Mass Spectrometry (ICP-MS) The results were validated against the PUBLISHED IN IJOG Indonesian Journal on Geoscience.
Vol.
12 No.
3 December 2025: 467-480 success ratio of the spectral-wise method.
An illustration of the research flow can be seen in the research flowchart (Figure .
Materials and Methods Materials The optical images from Landsat 9 has been used, retrieved from the United States Geological Survey (USGS) satellite data centre .
ttps:// The Landsat 9 was used to identify anomalies of phosphate minerals based on spectral analysis of the phosphate mineral spectral The atmospheric correction was applied using QGIS version 3.
34 Prizren to obtain surface reflectance from satellite images.
Detailed satellite images are listed in Tables 1 and 2.
Spectral reflectance data of the apatite minerals were obtained from the USGS Spectral Library at https://crustal.
gov/speclab/ (Table.
The spectral signature of monazite and xenotime has also been used to identify phosphate minerals.
The lineament analysis was carried out to identify the geological structure that controls primary phosphate mineralization .
By analyzing struc- tural patterns, potential areas can be identified for phosphate exploration and better understand the geological processes that lead to mineralization.
For the lineament analysis, the national digital elevation model (DEMNAS) 8.
3 m resolution was used, obtained from w.
Eleven samples have also been measured from the field consisting of rock and soil to verify the proposed In addition, nine measured rock samples have also been added from a previous study (Sukadana et al.
, 2.
Geochemical measurements were carried out at The Intertek Laboratory .
using X-Ray Fluorescence (XRF) and Inductively Coupled Plasma Ae Mass Spectrometry (ICP-MS) methods.
Phosphate rock content is conventionally expressed as phosphorus pentoxide (P2O.
, with economic minimum levels of about 5 % (Ptacek, 2.
Geochemical measurement result shows the P2O5 from 0.
17 to 1.
14 %.
K2O from 0.
82 to 32 %.
and Na2O from 0.
03 to 7.
5 %.
Reference measured rock samples from the previous study (Sukadana et al.
, 2.
show the P2O5 from 0.
73 %.
K2O from 0.
54 to 9.
88 %.
and Na2O 05 to 6.
99 %.
The result indicates a good prospect of phosphate-bearing igneous rock.
Figure 2.
Flowchart of the research to determine the potential distribution of phosphate minerals in the researched location, using Landsat 9 image analysis by applying the equation model .
pectral-wise metho.
PUBLISHED IN IJOG
Spectral-Wise Method Derived by Optical Images For Phosphate Mineral Exploration In Mamuju.
West Sulawesi.
Indonesia (R.
Kurniawan and A.
Saepulo.
Table 1.
Specification of The Data Used in This Study Image
Criteria
Cloud
Cover
Acquired Date
Source Path/ Frame
Scene ID
Resolution Landsat-9
Level 1
Collection 2
< 10%
July 27th.
United States Geological Survey
(USGS)
LC1150622023208LGN00
2013-141,
2013-142,
2013-123,
Mamuju 2013
Digital Elevation Model
(DEM)
Regional Geology Map
Single Band
Tiff 32 Bit
National Digital Elevation Model
(DEMNAS)
Mamuju Quadrangle.
Sulawesi Centre For Geological Survey Table 2.
An Overview of The Characteristics of The Landsat 9 Images Used in This Study Day/Month/Year Band Name Centre Wavelength (AA.
Resolution .
27/07/2023 B10
B11
Coastal Aerosol Visible .
lue/B) Visible .
reen/G) Visible .
ed/R) Near Infrared (NIR) Shortwave Infrared-1 (SWIR-.
Shortwave Infrared-2 (SWIR-.
Panchromatic Cirrus Thermal Infrared-1 (TIR-.
Thermal Infrared-1 (TIR-.
Table 3.
Specification of The Spectral Signature of Reference Minerals from The USGS Spectral Mineral Spectral Specification Chlorapatite Ca5(PO.
3(C.
Fluorapatite Ca5(PO.
3(F) Hydroxylapatite Ca5(PO.
3(OH) Monazite (Ce,La,T.
PO4 splib07a Chlorapatite WS423 ASDFRc
AREF
splib07a Fluorapatite REE WS418 xtl
ASDFRc AREF
splib07a Hydroxyl-Apatite WS425 BECKb
AREF
splib07a Monazite HS255.
3B ASDFRb
AREF
splib07a Xenotime GDS966 Iveland REE ASDFRa AREF Xenotime (YPO.
Spectral-Wise Method The mineral elements have a spectral signature at a certain wavelength as a basis for identification through their spectral signatures (Gupta et al.
, 2.
The spectral signatures of the five phosphate minerals have been identified in the fluorapatite, chlorapatite, hydroxilapatite, monazite, and xenotime samples (Figure.
Their spectra show a common absorption pattern at a wavelength of 2 AAm due to the absorption of phosphate ion or orthophosphate (PO.
The absorption at 2 AAm was located between band 6 and 7 in the Landsat 9 image (Table .
Sample Origin Sample Condition Snarum.
Norway Solid Otter Lake.
Quebec.
Canada Solid near Cokeville.
Wyoming Miguel County.
New Mexico Setesdal.
Iveland Municipality.
Norway Powder Solid Therefore, to identify phosphate minerals in the Landsat 9 image, the absorption angle was carefully been examined between the reflectance at 61 AAm (Band .
, 2 AAm .
hosphate absorptio.
, 2 AAm (Band .
The gamma angle was then calculated as an intersection of the reflectance line from 1.
6 AAm to 2 AAm and 2.
2 AAm to 2 AAm, termed as 1 to 5 (Figure .
Calculating the , the triangle concept was used as follows (Figure .
The is 180A minus the sum of the tangents and , each multiplied by 180A/A to convert from radians to degrees.
Tangent or equal to tangent A/X is the tangent
PUBLISHED IN IJOG
Indonesian Journal on Geoscience.
Vol.
12 No.
3 December 2025: 467-480 Spectral reflectance of the five phosphate minerals used in this study (Kokaly et al.
, 2.
, showing similar absorption at a wavelength of 2 AAm.
The angle among reflectance values at 1.
61, 2, and 2.
2 AAm was used as a basis to identify phosphate minerals.
X = i between 1.
61 AAm .
and 2 AAm Y = i between 2.
2 AAm .
and 2 AAm To obtain the phosphate index of on the Landsat 9 image, the Band 6 and 7, termed as B6 and B7 of Landsat 9, were applied to equation 2 as follows:
180o tan tan .
Figure 4.
The illustration of the in the triangular equation model to estimate phosphate absorption using the Landsat 9 image.
of the spectral absorption curve at i reflectance A .
6 and 2 AA.
divided by i wavelength X .
6 and 2 AA.
, and tangent or equal to tangent B/Y in the reverse direction .
etween 2 and 2.
2 AA.
Based on the triangle concept, the is expressed as follows:
B .
tan X tan Y Where:
A = iA between 1.
61 AAm .
and 2 AAm B = iA between 2.
2 AAm .
and 2 AAm The dense vegetation covers are a common problem utilizing the remote sensing approach under the Tropics.
The dense canopy vegetations serve as an obstacle to the reflected light received by satellite sensors (Prodomou et al.
, 2022.
Wang et al.
, 2.
To minimize the problem in this study, dense and sparse vegetation were separated using NDVI analysis.
The sparse vegetation area was used for mineral assessment.
To support the lack of data in areas with dense vegetation, mapping data was used from previous studies.
In the same case, if data are available, to minimize this problem leveraging advanced remote sensing techniques were used, such as hyperspectral imaging.
LiDAR, and multitemporal data analysis.
The accuracy and reliability of lithology identification in vegetated areas can be significantly improved.
Furthermore.
PUBLISHED IN IJOG Spectral-Wise Method Derived by Optical Images For Phosphate Mineral Exploration In Mamuju.
West Sulawesi.
Indonesia (R.
Kurniawan and A.
Saepulo.
exploring potential synergies among different data sources and adopting innovative data fusion methods can contribute to better classification results (Chen et al.
, 2.
However, the vegetation can respond to the environment in which it grows, such as soil elements, the amount of nutrients, soil moisture, salinity, surface roughness, slope, direction of light, and seasonal changes, including the type of rock (Saepuloh, 2.
The plant responses to soil contamination include tree density, health, height, species distribution, species replacement, growth phase, and aging (Saepuloh, 2.
The stressed vegetation can be assessed by remote sensing as an indicator of radioactivity soil contamination (Saepuloh et al.
, 2.
, hydrothermal alteration (Wu et al.
, 2.
, nutrient deficiency (Mee et al.
, or water shortage (Le et al.
, 2.
Therefore, the normalized difference vegetation index (NDVI) was applied to quickly delineate vegetation and vegetation stress (Huang et al.
, 2.
The NDVI was used to determine the dense vegetation area (Huang et al.
, 2.
that might be affecting the reflectance of (Chen et al.
, 2.
IDENTIFYING PHOSPHATE OCCURRENCE BASED ON IMAGE SPECTRAL
AND SOIL-VEGETATION RESPONSE
Identifying Phosphate Indices The range of from 160.
41A has been calculated in fluorapatite to 176.
38A in hydroxyapatite samples (Table .
Following equation .
, the calculated of the Landsat 9 image shows an index with an interval from 152A to 200A (Figures 5a, .
For chlorapatite, monazite, and xenotime the interval of is from 184A to 208A.
The is lower than this range, the phosphate mineral threshold is in the 1st quartile (Figures 5c,d,.
Following the calculation of for five selected reference minerals, all images have been superimposed to identify the occurrences of phosphate minerals based on their pixel density.
The density is derived from gridding an area of 2y2 km.
The occurrence of pixels were counted in each grid and perform kriging interpolation to find out the distribution.
To determine the threshold of the phosphate anomaly, the pixel gamma data was plotted on a histogram, resulting in a positive distribution (Figure .
Based on this distribution, the phosphate anomaly threshold was defined at the 4th quartile or number of pixels greater than 70/ grid (Figure7.
There are three main phosphate anomaly zones termed as PO Zones, namely Simboro.
Mamuju, and Tapalang prospect, named by location of the subdistrict, with a total area of 16,920 Ha (Figure 7.
Soil-Vegetation Response The dense vegetation can be a barrier to the reflected light received by satellite sensors (Prodomou et al.
, 2022.
Wang et al.
, 2.
The Normalized Difference Vegetation Index (NDVI), which standardizes indices to produce images that display greenness .
elative biomas.
, intends to distinguish plants with healthy conditions from plants that are experiencing stress (Saepuloh.
Healthy vegetation can grow better and more densely than stressed vegetation.
Thus, the NDVI was used to determine the dense vegetation area that might be affecting the reflectance of .
The spectral analysis results show that areas with high pixel density are generally directly proportional to high P2O5 content, but in some sampling locations, there are opposite results.
Based on reference measured rock samples Table 6.
Calculation Results of The Based on Equivalent .
for Phosphate Minerals Wavelength () Band 6 .
61 AA.
PO4 .
AA.
Band 7 .
2 AA.
i Reflectance (A) i Reflectance (B) i Wavelength (X) i Wavelength (Y) Tangent Tangent Gamma angle () Fluorapatite Chlorapatite Hydroxylapatite Monazite Xenotime PUBLISHED IN IJOG Indonesian Journal on Geoscience.
Vol.
12 No.
3 December 2025: 467-480 Hydroxylapatite Flourapatite Band 1 (Gra.
Band 1 (Gra.
199,69 5 Km 5 Km Chlorapatite Monazite Band 1 (Gra.
Band 1 (Gra.
5 Km Phosphate Xenotime Band 1 (Gra.
Phosphate 5 Km 5 Km Phosphate Figure 5.
The images from Landsat based on equation .
present the interval values of the .
Hydroxylapatite.
Chlorapatite.
Monazite and .
Xenotime index minerals.
On chlorapatite, monazite, and xenotime minerals, the minimum angle value is at 184.
36A .
bove the maximum gamma angle valu.
, and the phosphate mineral threshold is interpreted to be in quartile 1.
areas are covered by dense vegetation (Figure.
that the reflectance in the area was concluded disturbed by the very dense vegetation.
Verification and Discussion Figure 6.
The histogram of pixels from Spectral-wise method analysis shows a positive distribution.
The anomaly of phosphate is defined in the 4th quartile.
from the previous study (Sukadana et al.
, 2.
especially in the upstream area of The Mamuju River (Figure .
, it has a high P2O5 content of 73 % on MJU118 and 2.
67 % on MJU121, but both locations have low pixel densities.
The overlain with the NDVI image shows that Sample analysis results show P2O5 values 14 % in MMJ603 located in the Ampalas area in The Kalukku Subdistrict (Figure .
An outcrop on The Ampalas River is a foid syenite rock that is grey to light brown in colour, porphyritic in texture, and hypocrystalline (Draniswari et al.
, 2.
Phenocrysts are pyroxene, sanidine, leucite, nepheline, alkali feldspar, and plagioclase with aphanitic basement.
This outcrop extends for about 30 m on the banks of The Ampalas River, and is generally altered (Figure 9.
A),
PUBLISHED IN IJOG
Spectral-Wise Method Derived by Optical Images For Phosphate Mineral Exploration In Mamuju.
West Sulawesi.
Indonesia (R.
Kurniawan and A.
Saepulo.
MMJ603
Legend
Kriging PO4 Pixels 0 - 10
10 - 20
20 - 30
30 - 40
40 - 50
50 - 60
60 - 70
70 - 80 Phosphate 80 - 100 Threshold .
th Quartil.
> 100 Phosphate Anomaly (PO Zone.
5 Km PO Zones Subdistrict Boundary River Road 4 Km Figure 7.
The PO Zones are generated from kriging interpolation of pixels from Spectral-wise method analysis.
The threshold is defined on the 4th quartile zone.
The black point is sampling location for verification in Figure 9.
The Superimposed image overlain with PO Zones in black polygons, and density interpreted as the indicative area of phosphate minerals at Simboro prospect.
indicating the presence of secondary minerals in the form of sulfide minerals such as pyrite and chalcopyrite (Figure 9.
B).
Referring to the classification of volcanic rocks and based on the content of alkaline minerals and silica plotted on the Total Alkali-Silica (TAS) diagram, it is a tephri-phonolite rock type (Le Bas et al.
, 1.
In reference to samples from previous research (Sukadana et al.
, 2.
P2O5 reached 2.
in MJU118, which is in the upstream area of The Mamuju River in The Mamuju Subdistrict (Figure .
The plot results on the TAS diagram show the type of phono-tephrite rock.
Both tephri-phonolite and phono-tephrite are volcanic rocks with high potassium content, with total alkali of up to 12.
53 %.
PO Zones are mostly found in the western to northern part of the studied area, which is the Simboro and Mamuju prospect, and some appear in the middle part, which is the Tapalang prospect.
Following the concentration of potassium minerals in West Sulawesi (Sukadana et al.
, 2.
, about 15 % of PO Zones are in the high potassium zones (Figure .
This part is interpreted as the main concentration of phosphate minerals.
Analysis of the lineaments which is carried out by interpreting the straightness in the morphology of ridges and valleys which aims to determine its relation with the developing geological structure as well as its relation with indications of certain mineralization processes (Indrastomo et al.
, 2.
This analysis shows that the main lineaments is southwest - northeast, in line with the main structural pattern in the regional geology (Ratman and Atmawinata, 2.
and the PO Zone distributions (Figure .
PUBLISHED IN IJOG
Indonesian Journal on Geoscience.
Vol.
12 No.
3 December 2025: 467-480
MJU327
MMJ604
MMJ202
5 Km MJU37
MMJ603
MMJ203
MMJ402
MJU118
MJU24
MJU121
MJU51
MJU4
MMJ510
MMJ507
MJU27
Dense Vegetation NDVI_New MMJ Band 1 (Gra.
MMJ504
MMJ503
MJU38
MMJ502
MMJ505
Content of P2O5 (%) 0 - 0.
25 - 0.
5 - 1
1 - 1.
> 1.
PO Zones Dense vegetation area Figure 8.
The NDVI image shows the dense vegetation presented by dark red.
The light green polygonal line is the boundary.
Figure 9.
Outcrop of .
altered foid syenite at the northern part of the studied area for sampling location MMJ603.
The presence of secondary minerals in the form of sulfide minerals such as pyrite and chalcopyrite, in dark grey sections.
hand specimen of altered foid syenite in The Ampalas River.
Kalukku Subdistrict.
West Sulawesi.
The sampling locations are depicted in Figure 7a.
PUBLISHED IN IJOG
Spectral-Wise Method Derived by Optical Images For Phosphate Mineral Exploration In Mamuju.
West Sulawesi.
Indonesia (R.
Kurniawan and A.
Saepulo.
Content of P2O5 (%) 0 - 0.
25 - 0.
5 - 1
1 - 1.
> 1.
5 Km Figure 10.
The PO Zones obtained from the Spectral-wise method in black polygons overlain on the potassium (K) concentration map .
odification from Sukadana et al.
, 2.
Red polygons show areas of high potassium, and the blackshaded polygons are intersecting areas with PO Zones.
5 Km PO Zones Content of P2O5 (%) 0 - 0.
Intersection PO Zones and High Potassic 25 - 0.
Lineaments 1 - 1.
Fault Interpretation > 1.
5 - 1
Figure 11.
The identified geological lineaments developed in the studied area based on visual observation using the shaded map of The National Digital Elevation Model (DEMNAS) produce the rose diagram showing the main direction of the southwest northeast .
Overlaid with the regional geological map of Mamuju (Ratman and Atmawinata, 2.
, the lineaments show in accordance with the direction of the main fault .
PUBLISHED IN IJOG
Indonesian Journal on Geoscience.
Vol.
12 No.
3 December 2025: 467-480 Conclusion Acknowledgement The spectral absorption of fluorapatite, chlorapatite, hydroxylapatite, monazite, and xenotime minerals at a wavelength of 2 AAm is the absorption of orthophosphate (PO.
The Spectralwise method has been applied successfully to determine the orthophosphate in the Landsat 9 The phosphate mineral indication area (PO Zone.
was determined by the pixel density of the orthophosphate absorption.
The Spectral-wise method detected three prospect areas at Simboro.
Mamuju, and Tapalang about 16,920 Ha.
Based on the concentration of potassium minerals in West Sulawesi, about 2,604 Ha or 15 % of the PO Zone is located in the high potassium zone.
Ground check sampling verified that the eight samples were located precisely at The PO Zones and the twelve samples around the PO Zones.
Those samples had a higher P2O5 concentration than those outside The PO Zones.
The three samples in the upstream area of The Mamuju River were excluded due to dense vegetation covers presented by the normalized difference vegetation index (NDVI) with a pixel value index of more Based on the P2O5 sampling location, the correctness of the Spectral-wise method have been calculated, that is about 70 %.
The problem was primarily due to vegetation cover, the dense canopy vegetation served as an obstacle affecting the reflectance of .
To find the representative data, dense and sparse vegetation were separated and used only the sparse vegetation area for mineral Sapping data was used from previous studies to support the lack of data in areas with dense vegetation.
In the same case, to minimize this problem one can also explore potential synergies among different data sources and adopt innovative data fusion methods.
Based on the lineament analysis.
The PO Zones at the western part were in concordance with the SW NE strike-slip fault on the regional geology of Mamuju.
The PO Zones at the southwest to northeast are also in concordance with those strike slip fault.
Therefore.
The PO Zones in the studied area occurs at an intensive geological structures in SW NE direction.
The author would like to express sincere gratitude to Mr.
Robby Rafianto.
Head of the Exploration and Mine Development Department at PT Trimegah Bangun Persada Tbk.
, for his invaluable support and insightful advice throughout this Appreciation is also extended to Mr.
Tri Hadi Nugraha and Mr.
Gunawan Dwijatmiko for their assistance in collecting field data.
Additionally, the use of Landsat-9 satellite imagery, courtesy of the U.
Geological Survey, greatly contributed to the success of this study.
Reference