Important Value of Economic Potency Mangrove Using NDVI Satellite High Resolution Image To Support Eco Tourism Of Pamurbaya Area (Case Study: East Cost of Surabaya)

Indonesia is a vast maritime country; many mangrove conservations is found around coastal areas of Indonesia. Mangroves support the life of a large number of animal species by providing breeding, spawning and feeding. Mangrove forests as one of the unique ecosystems are potential natural resources, supporting the diversity of flora and fauna of terrestrial aquatic communities that directly or indirectly play an important role for human life in economic, social and environmental terms. East Coast Surabaya is an area with the most extensive and diverse mangrove ecosystems along the coast of Surabaya. Currently Pamurbaya used as a recreational object or nature tourism called eco tours. Utilization of mangrove ecosystem as a place of this eco tour bring positive impact on economic potency around pamurbaya area. So, to know the value of the economic potential of mangrove ecosystems for support of nature tourism Pamurbaya region needs to study mapping mangrove ecosystem conditions in the East Coast area of Surabaya. Mapping of mangrove conditions can use remote sensing technology by utilizing satellite image data with high resolution. Data used for mapping mangrove ecosystem conditions on the east coast of Surabaya are high resolution satellite image data of Pleiades 1A and field observation data such as Ground Control Point, soil spectral parameters and water quality. From satellite image data will be classification of mangrove vegetation canopy classification using NDVI vegetation index method using algorithm formula which then will be tested correlation with field observation data on reflectant value of field and water quality parameter. The purpose of this research is to know the condition of mangrove ecosystem to know the economic potential of mangrove ecosystem in supporting Pamurbaya nature tourism. The expected result of this research is the existence of basic geospatial information in the form of mangrove ecosystem condition map. So that can be used as decision makers to find out how big economic potential of mangrove ecosystem in supporting nature tourism Pamurbaya beside that can be used as reference research related mapping, conservation and development of mangrove ecosystem.


Introduction
Indonesia became the country with the most extensive mangrove forest in the world. According to [13] in 2005 Indonesia has 3 million ha of mangrove forest. Mangrove is a complex ecosystem consisting of flora and fauna in coastal areas or sea tidal boundaries. Mangrove is an influential vegetation in coastal ecosystems. Because Indonesia is a vast maritime country, so much found mangrove conservation around the coastal areas of Indonesia.
Mangrove is a plant community or an individual plant species that forms a community in tidal areas. Mangrove forests are a type of forest that is naturally affected by tidal seawater, flooded during high tides and free from puddles during low tides. The mangrove ecosystem is a system consisting of biotic and abiotic environments interacting within a mangrove habitat. The mangrove forest ecosystem is complex and dynamic, but labile. Complex because of its ecosystem in addition there is mangrove vegetation, is also a habitat of various animals and aquatic biota. Dynamic because the mangrove forest can grow and develop and continue to succeed in accordance with changes in its natural place of growth [16].
Mangrove forest management as a form of conservation of natural resources and coastal waters resources is necessary considering the mangrove forest ecosystem functions physically to keep the coastline stable, protecting the beach from erosion (abrasion), storm and wave dampers, sediment catchers, Rahmawaty in Suci, (2011), while the biological mangrove functions according to [3] as a spawning or shrimp area for shrimp, crabs, shellfish and others, as the area for shelter, nesting and breeding.In addition to ecologically functioning mangroves also function socioeconomically, Rahmawaty in Suci (2011) mangroves as a source of livelihood, production of various forest products such as wood, charcoal, medicine, building materials and handicrafts, natural attractions.
Mangrove ecosystems are found throughout the coastal city of Surabaya, one of which is the East Coast of Surabaya (Pamurbaya) which is a protected area. Pamurbaya is known as the remaining green open space and a fortress to protect Surabaya from the threat of abrasion, sea water intrusion, and land subsidence. Besides, Pamurbaya is also used as a recreational object of nature tourism or so-called eco tour. Utilization of mangrove ecosystem as a place of eco this tour brings a positive impact on the economic potential around the pamurbaya region. To know the economic potential value of mangrove ecosystem in supporting eco-tourism Pamurbaya area can be done by mapping mangrove ecosystem condition in East Coast area of Surabaya by using remote sensing technology that is by utilizing satellite image data of high resolution.
Mangrove ecosystem is one of the objects that can be identified by using remote sensing technology. The geographical location of the mangrove ecosystem located in the land and sea transition areas provides a distinctive recording effect when compared to other land vegetation objects. The recording effect is closely related to the spectral characteristics of the mangrove ecosystem, to the extent that identification requires a separate transformation. In general, for vegetation detection transformation used vegetation index [8].
In this study will map the condition of mangrove ecosystem by studying the condition of mangrove ecosystem vegetation based on Normalized Difference Vegetation Index (NDVI) value using remote sensing technique using satellite imagery. Also in this study the NDVI image value will be validated with field data. From the results of this study is expected to provide a spatial information about the condition of mangrove ecosystems used to determine the economic potential value of mangrove ecosystems in supporting eco-tourism Pamurbaya region. In addition, this research is expected to be used as reference material for research related to the field of mapping, conservation and development of mangrove ecosystem.

Radiometric Correction
Radiometric corrections include radiometric calibration and atmospheric correction. The radiometric calibration converts the DN (Digital Number) to Reflectance i.e. the process of changing the Pixel value to an image originally worth DN in the change to Reflectance. Then the atmospheric correction by the method of Gain of Set.

Cropping
Cropping is done to facilitate the image processing, and only covers the research area. So, minimize the image area to be processed. Cutting by area or area coverage of coastal mangrove ecosystem of North Surabaya.

Vegetation Index
Vegetation index indices vegetation index is done by entering the vegetation index algorithm. In this study using the algorithm Normalized Difference Vegetation Index (NDVI). NDVI algorithm as follows: Information : NIR = Near infrared spectral band value RED = The value of the red spectral band the process of incorporating this vegetation index algorithm will produce an image spectral value between -1 to 1.

The Claasification of Terelia
Furthermore, the classification of terelia by entering the value range of each type of vegetation health. Field observations included taking Parameter Samples (water and soil), Samples of parameters taken are soil and water physically and chemically. Chemically sample of water and soil parameters is done by laboratory test to know PH and salinity, then result of correlation test between index value of vegetation (reflectan) with value of parameter test result. From the tabulation data will be calculated correlation value.

Correlation Test
Correlation test is done on parameter sample and Spectral Ground. Correlation test is done by connecting the value of test result of parameter sample and Spectral Ground value with vegetation index value in the image so it can know the relation form. If the PH value is high, Salinity is low, with a high vegetation index value then it has a positive relationship and vice versa.

Analysis
At this stage we analysed the correlation value between vegetation index value with parameter sample test in the field. Furthermore, carried out the final stage of research, namely the preparation and writing of the report and its map information. Map obtained in this research is map of mangrove ecosystem scale 1: 5.000.

Image Improvement by Image Fusion Data Method (Pansharpening) on Pleiades image
The incorporation of low resolution multispectral images and high resolution panchromatic images is of great importance for remote sensing and mapping applications. Usually the panchromatic image has a wide wavelength ranging from the visible spectrum and near infrared, whereas the multispectral image only includes a narrow spectrum. However, multispectral images have more than three spectral bands, while the panchromatic image has only one spectral band. So by combining the two images are obtained multispectral image that has a high spectral resolution and high spatial information as well. The image integration in this research is done by using Gram-Schmidt transformation method, where each image is multispectral and georefenced choreographic image, so firstly done the co-registration of the images.

Determination of the Number and Distribution of Soil Control Points
The Soil Control Point (GCP) should be planned on a specific object, clearly visible in the image and easily recognizable in the field or object on the surface of the earth which changes relatively slowly / permanently, for example: roads, street corners, areas of striking colors, buildings / monuments that are easily identifiable or recognizable.

Figure 3. Location Selection at Ujung Jembatan
In the process of image rectification process, it is needed the right number of ground control points, so that the implementation of measurement in the field is not too long (Yudha, 2015). The number of ground control points is cultivated according to need, with considerations such as area size, mapping area relief condition, cost, and tools to be used for the measurement to be carried out efficiently.
The determination of the GCP point is also attempted to spread to the outermost position of the image to be performed koreksigeometrik. In this study, the number of GCP used is 8 points. Here is the GCP point spread:  Based on Table 2. the distance between GCP is 0.873-2.069 km, can be seen if the points used are in accordance with the standard distance between points used by the BIG, where 3-4 km for a relatively flat area and 2-3 km for hilly areas, and study areas in the area of the village of Wonorejo, Surabaya, which includes a relatively flat area.  Figure) is a geometric power (form) of a triangular circuit that determines the spread of errors in network flattening. The geometric strength of a good triangular web is mirrored by the small price of the SoF and will ensure thoroughness of the whole jarring. The smaller the power factorof the net, the better the net configuration will be, and vice versa. Here is a net design and SoF calculation on Pleiades 1B image.

Ground Control Point (GCP) Measurement Results
GCP was measured in the field using the Topcon Hyper Pro Geodetic GPS tool with a static method with 40 minutes of observation. Here are the coordinate results obtained from field measurements. The measurement data in Table 3. is the result of retrieving coordinate data which has been done post processing using Topcon Tools software v.7.5.1.

Independent Check Point (ICP) Measurement Results
The ICP point is used to test the accuracy of the geometric accuracy of the corrected image. The number of ICP points dimabil by 13 points, whose distribution is adjusted to the standard distribution of ICP point by BIG (Geospatial Information Agency) on Spatial Plan Validation module. Here is a picture of the sample point depth taken:

Results of Image Rectification
To find out how big the value of GCP error, it must be done image rectification process. Things to consider when performing the image rectification process is at the time of determining the GCP point of the image, try the dots to be determined on the image in accordance with the points measured in the field so that will get a minimum error value. The rectification process uses order 1 and 2nd order polynomials. This is applied equally to both imagery, so that the comparison can be known. At least 1st order polynomial required 3 GCP and for polynomial 2nd order requires 6 GCP minimum, while for 3rd order polynomial required minimum 10 GCP. Therefore, because the number of GCP used in this study is 8 GCP, so the most appropriate is the order 1 and 2nd order polynomial methods. In addition to the number of control points available, there are also other factors that influence the selection of order, such as the topographic state of the region as well as errors or distortions on the image to be rectified (Yudha, 2015). To know the accuracy of the image rectification, it can be seen from RMSE (Root Mean Square Error) value per pixel unit in image [24]. RMSE shows the degree of error by comparing GCP coordinates of transformation with actual coordinates in this case ie the coordinates of GPS measurement results. The following is the result of the Pleiades 1B image rectification process using the 1st order polynomial method and the 2nd order: In Table 6. The average RMSE value for Pleiades 1B with 8 GCP images is 0.322 pixels for 1st order polynomial method and 0.188 pixel for 2nd order polynomial method. Thus, the Pleiades 1B image rectification process is acceptable because it meets the given tolerance of ≤ 1 as suggested.

Geometric Accuracy Test
The geometric accuracy test is obtained by finding the residual measurement data in the field which is the coordinate point of the Geodetic GPS measurement and the coordinate point of the rectified satellite image. The geometric accuracy test is performed on Pleiades 1B satellite images rectified in each method, ie 1st order and 1st order polynomial methods. Here is a list of ICP coordinates obtained from the image interpretation: After obtained image image rectification image, then next is done RMSE calculation, to know how big shift of coordinate of point (residue) in that image. In RMSE calculations for geometric precision tests, 12 points are used because ICP 11 points can not be used because the results do not meet the tolerance. At the point of ICP 11 there is an error in the measurement, where there is no epoch with receiver 1, so the measurement of ICP 11 is absolute, which according to [1] absolute method is not in high accuracy (precision position up to several meters), while required for accuracy testing This geometric is less than 0.5 meters accurate, since the spatial resolution of the Pleiades 1B satellite image is 0.5 meters. The following is the result of the processing of the geometric accuracy test of Pleiades 1B satellite images.  Based on the calculation of RMSE geometric accuracy test of satellite image of Pleiades 1B result of rectification in Table 8. The value of RMSE 0,806 on the image of the 1st order polynomial rectification and on the image of the 2nd order polynomial rectification is 0,647 meter.

Image Feasibility Analysis
Based on the Technical Specification of 2014 Country Map Map by BIG, village maps can be made on a scale of 1: 2,500, 1: 5,000 and 1: 10,000. For the requirements of horizontal geometry accuracy that must be met in the process of making a village map that meets the standard accuracy of the map in accordance with Perka BIG No. 15 Year 2014.  Therefore, based on the multiplication of RMSE value with the coefficient of accuracy (1,5175), the horizontal accuracy of the 1B poliomial image is 1,222 meters. The 2nd horizontal polynomial method is 0.982 meters. Based on Table 10. the image feasibility as the basis for making the village map, Pleiades 1B satellite image meets the requirements of making a 1: 2.500, 1: 5.000 and 1: 10.000 village map.

Correlation Test
After processing NDVI and soil and water parameters test, the correlation test is done.

Figure 11. NDVI Relationship with Salinity
The results of the correlation test between NDVI with Nitrate, Phosphorus, Potassium, pH, High Salinity with value (R) of 0.99, 0.99, 0.99, 0.99.0.99 this shows that there is a strong relationship between NDVI value and NPK nutrient, pH , salinity. Correlation coefficient marked positive mean relation of NDVI value with one way nutrient parameter, if NDVI value is higher then NPK, pH, salinity is also higher because mangrove ecosystem requires NPK nutrient for photosynthesis process. Plants containing N, P, K, pH, enough salinity leaves will be greener. This greener leaf causes a higher reflectant value in the image.

Figure 13. NDVI Relationship with Salinity
The NDVI correlations with water parameters include correlations with very strong relationships with a value (R) of 0.99, 0.99, respectively. The correlation coefficient of NDVI with pH parameter and positive salinity means the higher the NDVI value, the higher the pH value and the water salinity. So the parameters are closely related to the mangrove ecosystem is pH and Salinity.

RMSE based on The Geometrical Accuracy
Based on the geometric accuracy test using ICP points, the RMSE value in Pleiades 1B image is 0.806 meters on the rectification image of 1st order polynomial method. While the rectification image of the 2nd order polynomial method is obtained RMSE in Pleiades image 1B 0,647 meter.

Result of Rectification
For the feasibility of high-resolution imagery for making village maps, Pleiades 1B satellite images meet the requirements of making a village map with a scale of 1: 2.500, 1: 5.000 and 1: 10.000.

The Correlation Test
The results of the correlation test between NDVI with Nitrate, Phosphorus, Potassium, pH, High Salinity with value (R) of 0.99, 0.99, 0.99, 0.99.0.99 this indicates that there is a strong relationship between NDVI value and NPK , pH, salinity. Correlation coefficient marked positive mean relation of NDVI value with one way nutrient parameter, if NDVI value is higher than NPK, pH, salinity is also higher because mangrove ecosystem requires NPK nutrient for photosynthesis process. Plants containing N, P, K, pH, enough salinity leaves will be greener. This greener leaf causes a higher reflectance value in the image.

The NDVI Correlation with Parameters
The NDVI correlation with water parameters includes correlation with very strong relation with value (R) of 0.99, 0.99, respectively. The correlation coefficient of NDVI with pH parameter and positive salinity means the higher the NDVI value, the higher the pH value and the water salinity. So the parameters are closely related to the mangrove ecosystem is pH and Salinity.

Conclusion
Based on geometric (area) and radiometric (classification) image processing for Pleiades 1B image with scale 1: 5000 obtained mangrove species dominated by Rhizophora sp. With an area of 966,45 Ha. The value of economic potency of Rhizopora sp Rp. 773.160.000/year with assuming a 25-year cutting cycle.