Volumetric estimation of an intensive apple orchard with GIS

The Earth’s ecosystems provides to a wide range of benefi ts to humanity known as “ecosystem services.” This has given the present balance of human resource usage increasing attention is being paid in almost every area of science. The UN initiative created by the Millennium Ecosystem Assessment in 2003, which discusses in detail the current and future status of individual ecosystems. In this announcement, the experts have been identifi ed in four different types of services, which are listed below: • the supply provisioning services (such as water, timber, fi ber, etc.) • the regulating services (e.g. climate, rainfall, fl ood control, disease control, etc.) • cultural services (such as the cultural, recreational, and cultural benefi ts, etc.) • supporting services (such as photosynthesis, nutrient cycling, soil formation) The green areas are known to plays a decisive role in the proper functioning of terrestrial ecosystems. The vegetation primary service is the production of oxygen, CO2 absorption and fi xation, thereby contributing to the reduction of the greenhouse effect (Gyeviki et.al., 2011). Trees have got the largest carbon sink potential in the ecosystem types. The reduction of forests at the global level (e.g. deforestation, forest fi res, natural decay, etc.). Nevertheless, international actions were appearance, which reduce greenhouse gases. Thus, it became necessary to elaborate forest carbon sink capacity estimations. In our country, Führer and Járó (1989) and Führer et al., (1991) dealt fi rst with the estimates the magnitude and variance of the forest carbon stock. The role of vegetation is highly depends on some vegetation indices which correlate to the biomass, species and variety composition, spatial location, the leaf surface and volume of tree (Radó, 2001). The estimation of assimilated CO2 fi xation in an orchard is especially necessary to defi ne the volume of trees, because during the assimilation, the amount of absorbed and incorporated carbon dioxide appear in the plant biomass production (Kiss et al., 2011). Although, fruit orchards are artifi cial ecosystems, but they represent a signifi cant biologically active green mass in the environment Volumetric estimation of an intensive apple orchard with GIS


Introduction
The Earth's ecosystems provides to a wide range of benefi ts to humanity known as "ecosystem services." This has given the present balance of human resource usage increasing attention is being paid in almost every area of science. The UN initiative created by the Millennium Ecosystem Assessment in 2003, which discusses in detail the current and future status of individual ecosystems. In this announcement, the experts have been identifi ed in four different types of services, which are listed below: • the supply provisioning services (such as water, timber, fi ber, etc.) • the regulating services (e.g. climate, rainfall, fl ood control, disease control, etc.) • cultural services (such as the cultural, recreational, and cultural benefi ts, etc.) • supporting services (such as photosynthesis, nutrient cycling, soil formation) The green areas are known to plays a decisive role in the proper functioning of terrestrial ecosystems. The vegetation primary service is the production of oxygen, CO 2 absorption and fi xation, thereby contributing to the reduction of the greenhouse effect (Gyeviki et.al., 2011). Trees have got the largest carbon sink potential in the ecosystem types. The reduction of forests at the global level (e.g. deforestation, forest fi res, natural decay, etc.). Nevertheless, international actions were appearance, which reduce greenhouse gases. Thus, it became necessary to elaborate forest carbon sink capacity estimations. In our country, Führer and Járó (1989) and Führer et al., (1991) dealt fi rst with the estimates the magnitude and variance of the forest carbon stock. The role of vegetation is highly depends on some vegetation indices which correlate to the biomass, species and variety composition, spatial location, the leaf surface and volume of tree (Radó, 2001). The estimation of assimilated CO 2 fi xation in an orchard is especially necessary to defi ne the volume of trees, because during the assimilation, the amount of absorbed and incorporated carbon dioxide appear in the plant biomass production (Kiss et al., 2011). Although, fruit orchards are artifi cial ecosystems, but they represent a signifi cant biologically active green mass in the environment ecosystems. Gyeviki et al., (2012) examined and evaluated the active green mass, in the context of CO 2 fi xation and water use of a cherry plantation. They examined the leaf surface temperature, stoma conductivity and photosynthetic activity of fruit trees.
In our investigation, to estimate timber volume, a special 3D remote sensing instrument and GIS software were used. Remote sensing, also called earth observation, refers to obtaining information about objects or areas at the Earth's surface without being in direct physical contact with the object or area (Belényesi et al., 2008). Remote sensing provides to get information from large areas beside/instead of traditional sampling data (Burai, 2007). According to Lóki (1996) the remote sensing means not only a special data collection, but processing and evaluation of these data also. The principle of remote sensing based on interactions and investigations of electromagnetic radiation with material (e.g. earth's surface). The basis of remote sensing is incoming radiation to the object (E I ). When the radiation incident upon the object's surface, is either refl ected (E R ) by the surface, transmitted (E T ) into the surface or absorbed (E A ) and emitted by the surface. These variables are depending by the wavelength (λ). So, it could be created the following equation: It could be determined from the equation that on given wavelength the refl ection, absorption and transmission are equal to the total incoming radiation. The values are always depended on the physical characteristics of the object and the geometric structure. When a remote sensing instrument has a line-of-sight with an object that is refl ecting solar energy, then the instrument collects that refl ected energy and records the observation. Most remote sensing systems are designed to collect refl ected radiation (Short, 2011). Based on the measured values it could be concluded to physical and possibly chemical characteristic of the observed object (Molenaar, 1993). There are two types of remote sensing: passive remote sensing and active remote sensing. Passive remote sensing is detected natural radiation that is refl ected by the object or surrounding area being observed. Refl ected sunlight is the most common source of radiation measured by passive sensors (Belényesi et al., 2008). One of the newest and extremely developed spatial data collection techniques are the LiDAR (Light Detection And Ranging) technique (Heritage and Large 2009). Laser scanner analyze the real world by a laser beam and collect surface information about the object. Based on the refl ected part of the laser beam, a high density so-called point cloud is creating by the scanner, which is a high quality 3D representation of the object's geometry. The fi rst laser scanners were used in forestry, which has a wide range of rather complex 3D modelling tasks in fi eld such as forest management, tree modeling, habitat examinations and carbon sink analysis as well (Vosselman and Hans-Gerd 2010).
There are researches, where positions and typical parameters (such as foliage structure, trunk diameter, tree height, etc.) of fruit trees can be successful determined by the 3D laser scanner remote sensing (Huang and Pretzsch 2010;Seidel et al., 2011). Stephens et al., (2007) report investigations to confi rm the relationship between LiDAR variables and forest carbon, height, basal area and age at plot scale. Carbon in artifi cial orchard by laser scanning technologies are less studied.
We examined in present investigation the software background of point cloud processing and it was defi ned the correlations of some parameters by the laser scanner.

Material and methods
In our research the plot area was an intensive apple orchard with drip irrigation system, protected by hail net in Regional Research Farm of the University of Debrecen near Pallag. The ScanStation C10 by Leica Geosystems based on the time-of-fl ight (TOF) principle for ranging. A short laser pulse is emitted towards the object and is refl ected on its surface and a part of the refl ected radiation comes back to the scanner where it is detected by a sensor.
The high scan rate of up to 50,000 points/sec which provides the fast determination of objects. The laser beam is emitted in the green wavelength range at 532 nm wavelength. The green laser light scans the objects; the defl ection of laser beam is occurred by a Smart X-Mirror™ automatically spins polygon mirror system. Thus the scanner creates a point cloud with high speed. The maximum range-which depends on the albedo-of laser scanner is 300 m. The beam divergence is 0.1 mrad, so it means that the diameter of laser point is 10 mm on 100 m. This value is only 3 cm in case of maximal measurement distance. The horizontal viewing angle of laser scanner is 360° and vertical is 270°. The integrated 4 megapixel (1920x1920 pixel) camera takes photos to color the point cloud. The Field-of-View of the camera is 17°, so the automatically spatially rectifi ed (panoramic) dome was made form 260 images on each scan position.
We have measured two rows (92 apple trees) of study plantation with 8 scan stations in leafl ess condition. The overlapping of scanning areas provided the joining of point clouds, and increased the accuracy of measurement. The scan resolution was 8 mm on 10 m, so the accuracy was below 1 cm on the right side. The processing of raw point cloud was carried out by Leica Cyclone 7.1 software than for post processing, Geomagic Studio 12 Software was used. The Leica Cyclone was ideal for registering the 8 scan stations and export the point clod for further data mining. Cleaning the point cloud form noises (disconnected points and components) the Geomagic Studio 12 software was more effective. After the noise reduction, an automatically triangulation was used to create the volume of trees. This operation is robust, since Delaunay triangulation method determines the number of created triangles. Considering the -Due to the branches, data losing was happened in many cases, which caused errors after the triangulation as well (Figure 1). After this step the volume of this digital model were calculated.

Results
Based on the wrap, the volume was calculated by the software. As a result, a continuous surface created to the volume calculation. The volumetric values were correlated with the diameter of the trunk (Figure 2) .
Trunk and height of trees are complex vegetative indices for identify tree biomass. Based on the physically parameters with the calculated volume of trees, middle poly nomial correlations were detected. Between the perimeter of trunk and volume of trees was higher correlation (r=0.61) than height of tree and volume of tree (r=0.54). The measured parameters are in Table 1.
The values of calculated trees' volume are smaller than existing values. The reason of this difference is the automatically wrapping by the software.

Conclusions
Based on evaluation of 3D laser scanner survey results the detection of certain quantitative and qualitative parameters becomes faster, however it will be necessary to refi ne the fi eld survey results. However, to carry out further investigations (e.g. CO 2 capture estimate) the calculated volume values of the software may also mean approximate baseline. Longer plan to clarify this results use with fi eld survey and estimate the annual CO 2 fi xing in the intensive apple orchard in biomass with GIS.