The normalized difference vegetation index value of volcanic paddy soils of Mt. Talang using Landsat 8 image

Accurate and up-to date information on the spatial distribution of paddy fields in volcanic soils is important. The volcanic soil in West Sumatra covers an area about 602.500 ha. Soils of volcanic origin are considered fertile and productive soils. They are widely used to grow annual and perennial crops. Paddy cultivation in volcanic soils is commonly practiced in West Sumatra. Remote sensing data can be used to observe the condition of rice plants in volcanic soils from by calculating the vegetation index value through the NDVI (Normalized Difference Vegetation Index). This study aims to determine the vegetation index of paddy grows in Mt. Talang by using Landsat 8 satellite imagery and correlate the index value with the total carbon and fractions content. Soil samples were collected using the 1x1 km interval grid method from 5099,78 ha of volcanic paddy fields. Soil samples were collected undisturbed and disturbed condition. The undisturbed soil samples were used to analyze bulk density (BD). The disturbed soil samples were dried and sieved with a 2-mm sieve and then analyzed for soil pH (H20 and KCl), Organic C, Labile C, and Very Labile C. All of soil samples have low bulk density (0.76-0.89 Mg/m3). Soil pH is considered neutral (6.50 to 6.93). The average vegetation index value was high, with a maximum value of 0.60 and a minimum value of 0.39. The results of the regression analysis showed that there was no relationship between NDVI and soil chemical properties. Determination of the vegetation index during the growing period of paddy is very useful to improve the accuracy of paddy identification sites at the regional scale and to predict the potential paddy production in due time.


Introduction
Rice is the staple and essential foods in West Sumatra Indonesia as well as in Indonesia.In West Sumatra, it is usual to cultivate paddy (rice) on volcanic soils.The volcanic soil in West Sumatra covered an area about 602,500 ha.In this region, there are four active volcanoes (considered type A), one of them is Mt.Talang [1].Soils of volcanic origin are considered fertile and productive soils [2].They are widely used to grow annual and perennial crops.Paddy cultivation in volcanic soils is commonly practiced in West Sumatra.Paddy cultivation in Indonesia is often found in areas near volcanoes [3].Soil made from the volcanic parent along the Barisan mountain ranges of Sumatra, Indonesia, is widely used for rice cultivation [4].
The production of paddy is crucial to maintain local and national food security in West Sumatra and Indonesia, so the need for spatial data is increasing.One example of an important spatial database is the vegetation index data which is needed by many parties, such as the government and agricultural sector, especially in Indonesia [5].Over the last three decades, remote sensing (RS) has been routinely employed to map paddy fields worldwide.Spectral indices are the most commonly used in RS to monitor the growth of rice plants.The timing of flooding stages was used to identify paddy fields from natural wetland areas.
Observation of the condition of rice plants in volcanic areas can be done from remote sensing by calculating the vegetation index value through NDVI (Normalized Difference Vegetation Index).However, this method rarely used to map paddy fields condition in volcanic soils of Solok District in West Sumatra.This research work has used both remote sensing data and applied geostatistical of Geographical Information System (GPS) to detect, estimate and predict the extent of paddy fields in the study area.This study aims to determine the vegetation index of paddy grows in Mt.Talang by using Landsat 8 satellite imagery and correlate the index value with the total C content.

Studied Area
The research area is a volcanic rice field area of Mount Talang, Solok Regency, West Sumatra Province.Geographically, it is located at 0° 49' 10.582"S -1° 4' 14.350"S and 100° 37' 50,060" E -100° 49' 25.250" E with the peak at an altitude of 2400 m asl.To the north, it is bordered by Kubung Subdistrict and Bukit Sundi Subdistrict, Solok Regency, West Sumatra Province.To the south, it is bordered by the IV Nagari Bayang Utara Subdistrict, Pesisir Selatan Regency; to the west by Lubuk Kilangan and Pauh Subdistricts, Padang City; and to the east by Lembah Gumanti Subdistrict and Tigo Lurah Subdistrict, Solok Regency.Administratively, the research area covers several sub-districts in Solok Regency, including Mount Talang, Danau Kembar, and Lembang Jaya (Figure 1).

Research methods 2.2.1 Vegetation Index Processing Research Method
The method used to obtain the vegetation index value was the NDVI (Normalized Difference Vegetation Index).In this process, the radiometric correction was carried out in Lapan software which consisted of the following: 1) lpni_tiff2ers_l1t_ldcm, used to convert standard Landsat-8 data from ground stations to ers format while performing radiometric corrections.The output of this software was in the form of multispectral and panchromatic files that were as follows: 1. L8uts1210065m_120913_nutm49_30_toa (multispectral) 2. L8uts1210065p_120913_nutm49_15_toa (panchromatic) 3. L8uts1210065m_120913_nutm49_30_toa_mask (cloudmask) 2) _L8_PanSharp, used to create color images of Landsat-8 data with a resolution of 15 meters by combining a panchromatic channel with a resolution of 15 meters with a multispectral channel with a resolution of 30 meters.This pan sharpening process produced a Landsat-8 image consisting of 5 multispectral bands, namely: band 1 (blue), band 2 (green), band 3 (green), band 4 (NIR), and band 5 (SWIR).
3. Processing in GIS softwares The radiometrically corrected image was processed in GIS software.After the preprocessing was completed, it continued processing the Landsat 8 image using the NDVI algorithm.Then, it was processed using ArcGIS 10.8 software.After processing the vegetation index with the NDVI algorithm, the NDVI map was cut with the research object, then overlaid with sampling points obtained from the survey results in the field to know the vegetation index distribution value.

Soil Sampling Analysis
Soil sampling was carried out using the 1x1 km grid interval method.Soil samples were taken from 76 sites with two depths, namely (0-20)cm and (20-40)cm, extending over an area 5099.78 ha of paddy fields in the Mt.Talang (Figure 2).Undisturbed and disturbed soils were collected with soil auger.Soil analyses were performed to determine soil pH, organic C, labile C, and very labile C. The undisturbed soil samples were taken using a sample ring at a depth of 0-20 cm and 20-40 cm for bulk density (BD) data.The chemical results of soil properties of the laboratory analysis were correlated with the vegetation index value from satellite image processing.

Bulk Density
Bulk density in topsoil is lower (0.74 ± 0.02 Mg/m 3 ) than subsoils (0.82 ± 0.02 Mg/m 3 ) and they are different to each other.Figure 3 shows the spatial distribution of soil bulk densities of volcanic paddy soils of Mt.Talang.The bulk density value less than 0.90 Mg/m 3 cover an area of 3093.40 ha, while the bulk density with a value range of (0.90-0.96)Mg/m 3 has an area of 1128.77ha.The bulk density value with a range of values (0.96-1.05)Mg/m 3 has the least area, about 871.12 ha.The bulk density of volcanic soil is low because it has a high porosity, making it easier for roots to penetrate the soils and to plow, increasing water holding capacity.

Chemical Soil Properties 3.2.1 pH
Figure 4 shows the spatial distribution of pH of H2O dan KCl in the studied soil.The pH (H2O) of volcanic paddy soils in Mount Talang is classified as neutral.The lowest average pH of H2O has a range of values 6.50-6.61with an area of 224.081 ha.The highest pH value of H2O has a range of values 6.85-6.93 with an area of 486.90 ha.The neutral condition of the soil pH is related to the organic matter content of the soil.The higher the C-Organic content of the soil, the more OH-will be contributed, which will neutralize the concentration of H+ ions.The highest average pH of KCl has a range of values 5.42-5.60 with an area of 1194.52 ha, while the lowest average pH of KCl has a range of values 6.22-6.52 with an area of 1306.63 ha.

Labile-C
The average value of the labile-C was classified as very low, low, and medium with a range of values of 0,687-2,12% (Figure 5).Higher content of labile-C found in the topsoil than subsoils and they are different to each other.The labile-C with a range of values 9.570-11.710ppm has the most area (1777.81ha), while labile C with a range of values 16.670-21.200ppm has the least area (169.89ha).The existence of the labile C fraction is very dynamic, so it is easy to lose due to soil culture.Zou [6] explains that the labile-C fraction is the most active C fraction of soil organic carbon with a rate transition that is changed and fast substantially after soil management or soil degradation so that it affects soil carbon availability.

Very Labile Carbon
The average value of very labile C was very low, with a range of values 0.0214-0.0584ppm (Figure 6).Very labile C with a range of values of 275-327 ppm has the most area, which is 1830.65 ha, while very labile C with a range of values of 327-393 ppm has the least area, which is 523.19 ha.The very labile C value is very low because this fraction is easy to be decomposed by microorganisms in the soil.Soil treatment, done continuously, can also cause very labile C values to be low, as very labile C has very small molecules, and water is easily carried, especially in the 0-20 depth of the top layer.Ros et al. [7] state that continuous soil treatment could damage the soil's overview, which will affect the existence of very labile C.

Vegetation Index
The results showed that the average value of the vegetation index was high to medium, with a maximum value of 0.60 and a minimum value of 0.39.The lowest NDVI value with a range of 0.39-0.44 has an area of 971.64 ha.The highest NDVI value with a range of 0.54-0.60 has an area of 616.73 ha (Figure 7).The Normalized Difference Vegetation Index (NDVI) is considered to be a good indicator of vegetation greenness and activity [8].Due to its favorable correlations with vegetation coverage, photosynthetic capacity, leaf area, biomass, etc. [9].

Correlation Between NDVI and Chemical Soil Properties
Correlation analysis demonstrated that there was no relationship between NDVI and four elements of chemical soil properties.The results of the regression analysis showed that the relationship between NDVI with pH, labile C, very labile C, and organic C was weak (r<4) (Figure 8).Labile and very labile carbons showed low regression and correlation with NDVI values.This is because carbon is fast to be lost and easily degraded, especially in areas with high rainfall, such as the volcanic area of Mount Talang.

Conclusion
In this study, we evaluated the paddy fields in Solok District West Sumatra.All of soil samples have low bulk density (0.76-0.89Mg/m3).Soil pH is considered neutral (6.50 to 6.93) and We found that the average vegetation index value was high, with a maximum value of 0.67 and a minimum value of 0.34.However, the relationship between NDVI with soil pH, soil organic C, soil labile C and soil very labile C fractions were lower.Our findings suggest that the NDVI value is not good predictor to estimate the soil pH, total soil carbon content and carbon fractions.

Figure 1 .
Figure 1.Rice Fields In The Volcanic Area of Mount Talang.
The equation of the vegetation index was: NDVI =NIR -R NIR + R Information : NIR : Near-Infrared R : Red The process to get the vegetation index value through the NDVI method was: 1. Image Download Landsat 8 images of Solok Regency (path 127, row 61) recorded on July 2, 2021, were downloaded in the Landsat catalog.The Landsat-8 data directory contained images of 11 bands and metadata.2. Radiometric Correction IOP Publishing doi:10.1088/1755-1315/1306/1/0120233

Figure 2 .
Figure 2. Soil Sampling Sites in the Volcanic Area of Mount Talang.

Figure 3 .
Figure 3. Spatial distribution of soil bulk densities in studied area.

Figure 4 .
Figure 4. Spatial distribution of various pH H2O dan KCl in topsoil volcanic paddy soil.

Figure 5 .
Figure 5. Spatial distribution of C-labile content in volcanic paddy soils of Mt.Talang.

Figure 6 .
Figure 6.Spatial distribution of C-very labile content in volcanic paddy soils of Mt.Talang.

Figure 7 .
Figure 7. Spatial distribution of Normalized Difference Vegetation Index (NDVI) value in volcanic paddy soils of Mt.Talang,

Figure 8 .
Figure 8. Scatterplots of four chemical soil properties with NDVI.