development of soil information system and its application in korea

DEVELOPMENT OF SOIL INFORMATION SYSTEM AND ITS APPLICATION IN KOREA S. Young Hong, Yong Seon Zhang, Yi Hyun Kim, Kang Ho Jung, Ye Jin Lee, Eunyoung Choe, Sang Keun Ha, Yeon Kyu Sonn, Byung Keun Hyun, Myung Suk Kim, Seung Oh Hur, Kwan Cheol Song, Goo Bok Jung, Gun Yeob Kim, and Ki Yeol Jung National Academy of Agricultural Science (NAAS), Rural Development Administration (RDA) 150 Suinro, Gweonsun-gu, Suwon 441-707, Korea

ABSTRACT An internet-based Soil Information System (SIS) was developed based on the results of all the soil survey projects and related activities carried out by the Rural Development Administration (RDA) in about 40 years of managing soil resources in Korea. Korea’s soil information system provides web-based fertilizer prescription program that recommends the amount and type of fertilizer, and the timing of fertilizer application for optimum plant growth based on soil nutrition diagnosis in the soil. It also contains soil theme maps and crop suitability maps; soil statistics for selected soil themes; a textual information about Korean soils; and soil information for kids. The Agro-environmental Resources Information System has also been developed at the NAAS-RDA. It contains spatial database on natural resources, 3-D soil maps, an image database, statistics, and cyber forum. Among the applications of the soil information are: digital soil mapping of soil carbon storage and water capacity; mapping of hydrologic soil groups; mapping of soil erosion in Gyeongbuk Province; and soil fertility mapping of Korea. Future research direction is towards pedometrics or the application of mathematical and statistical method for the study of soil genesis and distribution. This will be useful for soil mapping and prediction of soil and environment properties. Remote Sensing and Geographic Information System will continue to play important roles in spatial and temporal assessment of agricultural environment. Key words: soil information system, soils, agro-environmental resources, remote sensing, digital soil mapping, Korea

INTRODUCTION Soil map delineates the boundaries of different kinds of soils whose characteristics are markedly different due to the various factors affecting soil formation. These factors include climate, parent material, topography, vegetation, and length of time for the soil formation. Detailed knowledge of soil characteristics is important in soil resources use and conservation. The importance of soils as the basis of Korean farming was recognized long ago. A book titled “Nong-Sa-Jik-Seol”(Instruction for Agriculture) was published in 1429 during the reign of Great King Sejong. The quality of soils was mentioned in detail, touching upon how and when to plow the lands, how to improve the fertility of barren soils, and even

how to test the quality of soil by tasting it. Many books about agriculture were published during the Choson Dynasty, following the example of “Nong-Sa-Jik-Seol.” The first modern soil survey was initiated in 1964, when the Korean Government and UNDP/FAO of the United Nations jointly established the Korea Soil Survey Organization Plant Environment Research Institute (now National Academy of Agricultural Science, NAAS) under the Office of Rural Development (now Rural Development Administration, RDA) in Suwon. Since then, soil survey has made substantial progress that leads to understanding spatial soil distribution, recommending fertilizer prescription, land suitability for crop cultivation, and rationally managing soils in agricultural lands. 1

All the soil maps surveyed and made in RDA were computerized to make digital soil maps. An Internet-based soil information system was developed based on the results of all the Soil Survey Projects carried out by RDA in about 40 years of managing soil resources rationally and providing soil information to the public promptly. This paper introduces and explains agroenvironmental inventory in Korea focusing on soils database, soil information system, and its application and future direction for natural resources management.

AGRO-ENVIRONMENTAL INVENTORY IN KOREA Agro-environmental inventory can be defined as ‘an integration of data and value-added information on natural resources to assess agricultural and environmental sustainability. Agro-environmental resources include soils, land surface, water, crops, microbes, vegetation, management practice, and climate. Agro-environmental resources and soil properties are important for land management, plant production, and environment and ecosystem management. Land quality, soil quality, water quality, land use intensity, agrodiversity, and greenhouse gas emission are important factors indicating the static and dynamic condition of the agro-environment. Agro-environmental indicator (AEI) is defined as a measure of a key environmental condition, risk, or change resulting from

agriculture, or of management practices used by producers (McRae et al. 2000). AEIs are becoming more important and are gaining international interest because of issues on environmental sustainability. Farmers, governments, researchers, environmentalists, and consumers as decision makers link with each other and influence each other in ensuring a sustainable agriculture industry. For instance, farmers’ management practices directly influence environmental sustainability, but agriculture policy influences decision-making activities. Information is one of the common needs of all decision makers concerned with agricultural sustainability. Soils Database The soil survey started with a reconnaissance survey, making use of aerial photographs purchased from USA using the Korea Soil Survey Organization fund between 1964 and 1967 (when the reconnaissance soil survey was carried out). As a result, 1:250,000 and 1:50,000 scales of soil maps of Korea were published. The detailed soil survey was followed thereafter adopting the Soil Taxonomy of the United States Department of Agriculture (USDA) between 1968 and 1990. Now, detailed Soil Maps (1:25,000) for the entire country are available both in hard copies and digital (digitized) soil map files. Super-Detailed Digital Soil Maps (1:5,000) for the entire country will be provided to the public on-line and will be published soon (Fig. 1).

Fig. 1. History of soil survey and database established in Korea (NIAST 1999). 2

During a field study, morphological and physical characteristics of soil such as parent material, soil texture, topography, soil drainage classes, gravel content, slope, effective soil depth, and erosion rates are often used to determine the series, order, and age of soil. Topography, soil drainage classes, soil texture, effective soil depth, and gravel content not only play important roles in determining soil series, but along with vegetation and climate are significant and fundamental factors for both understanding the formation of soils and analyzing the characteristics of farmland. Based on National Soil Survey Projects, two big soil databases were established. One is the spatial database of soil map in a variety of scales (1:250,000, 1:50,000, 1:25,000, and 1:5,000) established between 1998 and 2005. Table 1 summarizes soil survey methods at each scale and its application. All the soil maps were digitized and made into a GIS file format. The other is parcel-based soil fertility (chemical properties) database by soil testing

established using an oracle relational database management system. Soil testing determines both the amount of soil nutrients essential for plant growth and the existence of any harmful ingredients in the soil. It can be used to make recommendation on the amount and type of fertilizer, the timing and the exact location of fertilizer application for optimum plant growth. Furthermore, it can help achieve proper and effective nutrient management of crops. Improper application of fertilizer would be problematic because of the uniqueness of soils and crops. Understanding soil chemistry by measuring the exact content of chemicals in soil is the first step towards maintaining and managing environmentally sound soil. Two big important soil databases are the solid basis of the soil information system of Korea. The purpose of the data collection and utilization, the check items, and the form of data sampled and built of the two databases are significantly different (Table 2). RDA is in charge of National Soil Survey Project and updating of soils

Table 1. Soil survey methods at each scale and its application (NIAST 1999) Division

Reconnaissance Soil Survey

Detailed Soil Survey

Super Detailed Soil Survey

Scale of base map

1:40,000

1:10,000-18,750

1:1,200-5,000

Scale of soil map

1:50,000

1:25,000

1:5,000

Minimum mapping unit

6.25 ha

1.56ha

10a

Accuracy and soil classification

Aerial photogrammetry

Field investigation

Analysis of individual plots

Soil group and subgroup

Soil series and mapping units

Soil series and other mapping units, land use

Distance between sampling points

500-1,000m

Less than 100m

Less than 50m

Application

Soil genesis and distribution of soil group

Agricultural advise at regional and local level

Individual household farm management



Central and regional land use planning

Regional land use planning

Soil management for individual plots



Identification of major crop zone

Choice of crop for individual plots



Soil fertility management

Identification of suitable crops



Basic information for soil management

Selection of specific land management

3

Table 2. Characteristics of databases for soil map and soil fertility data

Soil Map Database

Soil Fertility Database

Characteristics

Morphological, physical, and chemical characteristics from topsoil to subsoil

Chemical properties of top soil

Sampling unit

Soil boundary

Each field with parcel information

Purpose

Soil survey

Soil testing

Utilization

Land use recommendation, crop suitability, land suitability

Fertilizer recommendation

DB

Spatial database

Non-spatial database

Data form

A variety scale of soil map - 1:250,000, 1:50,000, 1:25,000, 1:5,000

Greater than 5 million samples (since 1995)

information. For a soil testing project, RDA makes a plan which the Agricultural Technical Center (ATC) of each city or county carries out. RDA provides the standard of fertilization for 99 crops, checks samples for laboratory analysis, and handles the web-based operating program to print out fertilizer recommendation. Each local Agricultural Technical Center collects and analyzes soil samples for the areas. They also need to upload the analyzed soil fertility data to RDA database. Web-based fertilizer recommendation program developed in 2007 uploads soil testing data from local ATCs automatically to the RDA soil fertility database while they input the soil test data for printing out a fertilizer prescription. Agro-environmental Change Monitoring Database and Others The agro-environmental change monitoring project started in 1999 under the ‘environment-friendly agriculture fostering law’ of the Ministry of Agriculture and Forestry (now the Ministry for Food, Agriculture, Forestry and Fisheries). Data for soil physical and chemical properties, water quality, and soil microorganisms at fixed sampling points evenly distributed in the country were collected and analyzed regularly. The number of samples collected and the sampling frequency differ between sampling items. Sampling was designed to represent regions and topography in relation to soil characteristics. 4

NAAS is in charge of ‘the agroenvironmental change monitoring project’ and controls the quality of the data analyzed for soils and water by providing analysis methods and checks samples of each Provincial Agricultural Research and Extension Services who are implementing the sampling and laboratory analysis. Soils data from topsoil and/or subsoil were collected in paddy fields, dry fields, plastic film houses, and orchards to analyze their chemical properties including pH, organic matter, available phosphate, and exchangeable cations. Soil samples for each cropland type are collected every four years in turn in paddy fields, dry fields, plastic film houses, and orchards from 1999. The number of samples for paddy fields is about 2,000 points, dry fields about 1,600 points, plastic film houses about 1,200 points, and orchards about 1,300 points. Four hundred fifty samples of soil microbes and their biomass C were examined yearly in turn from paddy fields, dry fields, plastic film houses, and orchards using the same soil samples for soil properties (Fig. 2). Surface and groundwater quality of farming areas were examined and elements like T-N, T-P, NH4-N, NO3-N, COD Mn/Cr and exchangeable cations were analyzed thrice and twice a year depending on the farming period and using the standard methods for water quality analysis (Fig. 2). Biological diversity is important in assessing agro-ecosystem and its sustainability. Vegetation and weeds in the agricultural

Fig. 2. Sampling sites for agro-environmental change monitoring.

Fig. 3. Sampling sites for vegetation and climate resources. fields were surveyed to understand the spatial distribution, appearance frequency, and dominance of plant species. Spatial database for synoptic meteorological data observed from 57 weather stations of the Korea Meteorological Administration (KMA) were established as well (Fig. 3). NAAS started to make these data spatially referenced using the parcel information on the sampling sites for soil physical and chemical properties, water quality, and soil microbes to monitor spatial distribution and temporal changes for the resources of the country in 2007 as shown in Figs. 2 and Fig. 3.

The usefulness of an information system will depend on its ability to provide decision makers with the right data at the right time in the proper manner. Internet-based soil information system of Korea was developed to provide soils information for the public in 2006 – anybody, anywhere, and anytime – since the super detailed soil map (1:5,000) and soil testing data started to be computerized from 1998. The development of the National Agroenvironmental Resources Information System (NARIS) has just started.

INFORMATION SYSTEM DEVELOPMENT

A digital soil database based on paper soil maps made in Korea over a period of more than 40-years and at different scales – 1:250,000, 1:50,000, 1:25,000, and 1:5,000 – was established to provide valuable soil information over the whole country. RDA made a soil information system (http://asis.rda. go.kr) of Korea based on super detailed soil map (1:5,000) and opened ‘all about soils’

Information systems are the software and hardware systems that support data-intensive applications. A geographic information system (GIS) is used to integrate, store, edit, analyze, share, and display georeferenced information. GIS plays essential roles in integrating a variety of data layers to express a real world.

Soil Information System

5

information to the public through the web as shown in Fig. 4. It is an Internet system to show soil theme maps for suitable crops and recommend the amount of fertilizer. Korean soils web site consists of five main parts, which are digital soil map services such as: 1) soil theme maps and crop suitability maps; 2) fertilizer recommendation; 3) soil statistics for selected soil themes; 4) Korean soils; and 5) ‘soil love’ for kids. Soil attributes shown graphically in the digital maps are 79 items which are soil texture, gravel content, drainage, available soil depth, slope, topography, parent material, land use at the time of soil survey, soil suitability group for paddy, upland, and orchard, soil classification regimes, and the like (Table 3). Each soil property can be provided in a form of map, showing its spatial distribution with statistics such as sum and average of the attributes. Soil map-based attributes provided at the web site are morphological, physical, land use, soil classification, crop and soil suitability (Fig. 5). Parcel information which is a spatial database provided by Ministry of Land, Transport, and Marine Affairs was used to link the soil testing database to express the parcel-based soil fertility map using the parcel information as a relational key in the web site (Fig. 6).

Soil testing determines both the amount of soil nutrients essential for plant growth and the existence of any harmful ingredients in the soil. It can be used to make a recommendation for the amount and type of fertilizer and the timing and exact location of fertilizer application for optimum plant growth. Furthermore, it can help achieve proper and effective nutrient management of crops. Improper application of fertilizer becomes a problem because of the uniqueness of soils and crops. Understanding soil chemistry by measuring the exact content of chemicals in soil is the first step towards maintaining and managing environmentally sound soil. Soil testing reveals the chemical properties of soil such as pH, soil organic matter, available phosphoric acid, available silicic acid, and exchangeable cations. The soil information system of Korea (http://asis.rda.go.kr) provides web-based fertilizer prescription program that recommens the amount and type of fertilizer, and the timing of fertilizer application for optimum plant growth based on soil nutrition diagnosis in the crop land (Fig. 7). A person in charge of soil testing and diagnosis at ATC has his or her own ID and password to access the system to use the data and functions of program provided by the system. The Program provides fertilization standards for

Fig. 4. An overview of the Korean Soil Information System (http://asis.rda.go.kr).

6

Table 3. Soil theme map and its attributes shown through the information system Soil Theme Maps

No. of Attributes

Soil Attributes

Morphological and 9 physical maps

Soil texture, drainage class, available depth, slope, gravel content, soil color, soil erosion grade, etc.

Land use maps

Land use, land use recommendation(paddy field, dry field, orchard), soil type(paddy field, dry field, forest), etc.

6

Soil classification maps 5

Soil order, suborder, topography, parent materials, deposition

Crop suitability maps 29

Apple, pear, mandarin orange, water melon, grapes, strawberry, tomato, cucumber, cabbage, etc.

Soil suitability maps

Paddy field, dry field, orchard, grass, forest soils

5

Soil chemical maps 25

pH, organic matter, available phosphate, potassium, alcium, available silicate, etc. for each cropland unit

Fig. 5. Soil map-based attributes provided at the website. 99 crops and issues fertilizer and management prescription. Soil test data are automatically uploaded to the oracle database located at NAAS when one inputs the soils data and saves them for the program operation. For the web-based fertilizer prescription program, parcel (Korean Land Information System, KLIS) provided by Ministry of Land, Transportation, and Marine Affairs and farming land information provided by Ministry for

Food, Agriculture, Forestry and Fisheries to offer parcel-based soil fertility map and fertilizer prescription for farming. The statistics of soil attributes queried on the web can be calculated for representative areas in the form of pie charts, bar charts, and tables (Fig. 8). Soil information that can be queried for statistics are all 62 attributes including soil morphological, physical, and chemical properties. 7

Fig. 6. Parcel-based soil chemical properties provided at the web-site.

Fig. 7. Web-based ferilizer prescription program from data input to fertilizer recommendation. Also, the system provides text information on Korean soils. Soil is a mixture of mineral particles and organic matter of varying size and composition. The particles make up about 50% of the soil’s volume. The proportions of sand, silt, and clay in a soil determine its texture. Soil is not a constant entity by itself but in dynamic equilibrium with its environment. The zone at the soil surface, called the pedosphere, is the envelope of the Earth where soils occur and where soil-forming processes are active. Most soils take a long time to form. Soils are formed from various parent materials under different environmental conditions. Their characteristics are very different, depending on the combination of soil forming factors. 8

Factors of soil formation include: climate, biota, source  materials, slope, and time. The influence of these factors is interdependent and, in fact, they exert a combined but by no means equal control on soil formation. Three hundred ninety soil series were recognizable in Korea as of 2007. Fig. 8 also shows soil types of paddy fields and dry fields in Korea. Development of Agro-environmental Resources Information System Based on agro-environmental resources inventory, development of agro-environmental resources system just started in NAAS. The purpose of system development is the establishment of national agro-environmental

a) Soil statistics

b) Korean soils

Fig. 8. Soil statistics in the form of pie charts, bar charts, and tables (a) and Korean soils (b). resources inventory and its rational use. WebGIS system for agro-environmental resources information will provide a spatial database on natural resources, 3-D soil maps, an image database, statistics, and cyber forum as shown in Fig. 9 from early 2009.

APPLICATIONS OF SOIL INFORMATION Digital Soil Mapping of Soil Carbon Storage and Water Capacity Soil carbon storage and available water capacity are important properties for land management, plant production and environment, and ecosystem management. This project tried to apply the digital soil mapping concept for mapping these two properties in South Korea. A Korean soil database was compiled, which includes chemical and physical properties such as particle size, moisture retention, organic matter, cation exchange capacity, and a limited number of bulk density data based on 380 soil series. The first step is to estimate bulk density for estimation of both C storage and available water capacity. Bulk density at different depths of soils was predicted by deriving a pedo transfer function model with sand, depth, and organic matter, based on Adams’ model (1973). Organic C distribution with depth was first derived by converting from mass basis

C(kg/kg) to volume basis C(kg/m3). C storage (kg/m2) was first calculated by multiplying C on the volume basis to the thickness of each soil layer (m), and finally integrated to a depth of 1 m for each soil series. Mapping available water capacity was more challenging as only half of the database contains measurement of water retention at -33 and -1,500 kPa. Furthermore, measurement of water retention is in mass basis and based on disturbed soil samples. Pedotransfer functions were derived for volumetric water content at field capacity (-33 kPa) and wilting point (-1,500 kPa). Further adjustments based on total soil porosity are required as the field capacity values were derived from disturbed soil samples. Field capacity was calculated from clay content and predicted bulk density and adjusted by considering porosity. Wilting point was calculated from clay content and adjusted for any discrepancy with predicted field capacity and porosity. Available water capacity (mm) to a depth of 1 m was estimated by multiplying the amount of water stored between field capacity and wilting point and the thickness of the layer. The carbon storage and available water capacity from surface to a depth of 1 m for the south part of whole Korean peninsula were mapped using the estimated parameters in a soil series map unit (1:25 000). Mean value of carbon storage of Korea is 9

Fig. 9. Main page of agro-environmental resources information system. approximately 7 kg/m 2 and available water capacity is approximately 138 mm as shown in Fig. 10. Mapping Hydrologic Soil Group The soils were classified into four hydrologic soil groups (HSG) on the basis of measured infiltration and percolation rates, and pedotransfer functions for Ks (saturated hydraulic conductivity) of the representative soils (Jung et al. 2006). They are A(7.62-11.43 mm/hr), B(3.81-7.62 mm/hr), C(1.27-3.81 mm/ hr), and D(0-1.27 mm/hr) groups (Mishira and Singh 2003). HSG A is faster in the infiltration and percolation rates than B, C, and D group. HSGs by Jung et al. (1995) and Jung et al. (NIAST 2006) were mapped using the estimated infiltration rate in a soil type map (1:25,000) in Fig. 11 and compared the areas by each group in Table 4. The area of HSG D increased more than four times and that of the other HSG decreased by Jung et al. (NIAST 2006) in comparison with the classification of Jung et al. (1995). Soil Erosion Soil loss is one of the important properties for understanding environmental loads and 10

practicing best management of land and agricultural fields. Soil erosion can be classified as natural soil erosion or accelerated erosion. Under normal climate conditions, and with natural ground cover, soil erosion can often balance out with the rate of soil production. This is called 'natural erosion', or 'geological erosion'. Natural soil erosion includes water erosion caused by rainfall and runoff of water, wind erosion, glacial erosion caused by flowing ice, stream erosion occurring during channelized water flow, coastal erosion caused by waves and tides, and erosion due to transportation. Erosion occurring at a rate that exceeds the rate of natural erosion is called accelerated erosion. Accelerated erosion can result from certain human land use practices such as surface mining, forestry, agriculture, and construction. Additional natural erosion processes, such as water erosion, are responsible for accelerated soil erosion. Soil erosion in Korea is primarily caused by rainfall. This rain-splash erosion detaches earth materials from the surface by the impact of water striking the surface and by the force of surface runoff. Wind erosion is found in coastal regions and highlands. The intensity and amount of rainfall, soil type, the length and steepness of slope determine the degree of soil erosion by rainfall.

a) Soil carbon storage

b) Available water capacity

Fig. 10. Soil carbon storage and available water capacity map of Korea.

Fig. 11. Distribution of hydrologic soil groups based on soil map (1:25,000) of Korea. Soil erosion of Gyeongbuk Province in Korea was mapped, basically using the Universal Soil Loss Equation (USLE), multiplying rainfall-runoff erosivity (R), soil erodibility (K), slope length and steepness (LS), cover management factors (C), and support practice factors (P) (Fig. 12). Revised USLE (RUSLE) was used to calculate for K

and LS factors. Rainfall and runoff erosivity factors (R) of 158 locations of cities and counties were spatially interpolated by the inverse distance weight (IDW) method. K factors and C and P factors were applied to soil map and land use/cover map, respectively, at 1:25,000 scale. Digital elevation model (30 m) was used to derive LS factors using 11

Table 4. Areas of hydrologic soil groups by classification method in soil map (1:25,000) HSG 19951

Area (km2)

Ratio (%)

A B C D Total

41,261 28,679 18,041 9,648 97,629†

42.2 29.4 18.5 9.9 100

HSG 20062 A B C D Total

Area (km2) 34,302 15,342 5,342 42,643 97,629

Ratio (%) 35.1 15.7 5.5 43.7 100

Jung et al. (1995), 2Jung et al. (NIAST 2006); Total area does not include river and reservoir and areas with no soil maps. 1

RUSLE. A grid-based soil loss concept and model was made to map soil erosion using GIS (Fig. 12).   Soil loss is classified into seven grades in Korea and five grades in OECD. It needs to do conservation practice for the classes, which are greater than moderate grade.  Runoff Potential “Curve number” (CN) indicates the runoff potential of an area. The US Soil Conservation Service’s CN method is a simple, widely used, and efficient method for estimating the runoff from a rainfall event in a particular area, and is also used in Korea in disaster assessment. CN is based on an area’s hydrologic soil group, land use, management practices, and hydrologic condition. The two former characters are of greatest importance. Spatial variability in land use, cover, and soil hydrological properties reflecting infiltration rate, at the field level during the crop growing season must be considered when runoff is being estimated. HSG information NIAST(2006) of the soil map (1:25,000) of Korea and land use/cover map(1:25,000) provided by Ministry of Environment were used to calculate CN values, with antecedent soil moisture condition II, of 30 m grid format. CN values, which are ranged from 0 (no runoff) to 100 (all precipitation runs off), of Korea were mapped (Hong et al. 2007) as shown in Fig. 13. Soil Fertility Map Soil fertility map for the country was expressed in terms of soil pH and organic

12

matter (OM) content for paddy and dry fields using a weighted average value of soil chemical property based on soil testing database (Fig. 14 and Fig. 15). Mean OM content of paddy field soils had a tendency to be greater than that of dry field soils. pH of paddy field soils in the mountainous areas was lower than in the rest of the soils. Dry field soils showed higher pH, in general, than paddy field soils. The pH of soil can be an important parameter for crop growth. When a soil’s pH is about neutral (6.5-7.0), availability of soil nutrients is at maximum. As soil becomes more acidic (pH