Comparison of Land Suitability Evaluation Methods for Development Plan of Orange and Tea Trees in Western Nghe An, Vietnam45230

Comparison of Land Suitability Evaluation Methods for Development Plan of Orange and Tea Trees in Western Nghe An, Vietnam Ha Nguyen Manh 1* , Tuan Nguyen Thanh 2 , Ngoc Hoang Thi Huyen

Comparison of Land Suitability Evaluation Methods for Development Plan of Orange and Tea Trees in Western Nghe An, Vietnam

Ha Nguyen Manh 1* , Tuan Nguyen Thanh 2 , Ngoc Hoang Thi Huyen 1 , Dung Bui Quang 1 , Huong

Vu Thi Thu 1

(1) VAST Institute of Geography, Hanoi, Vietnam

(2) VAST Vietnam National Museum of Nature, Hanoi, Vietnam

* Coresspondence: havdl72@gmail.com

Abstract: Conflicts of agricultural land use and the problem of stabilizing specialized cropland areas for sustainable development of the mountainous economy can be improved through appropriate land assessment to identify potentials and limitations A number of land assessment methods have been applied in Vietnam, each has its certain advantages and disadvantages when applied to each specific case This study was conducted to compare and verify the results of two ALES and LSE methods in determining the appropriate area for orange trees and tea trees in Western Nghe An, Vietnam 15 ecological criteria were used in assessing soil for orange cultivation and 10 criteria for evaluating land for tea cultivation, divided into 3 parameters: parameters of climate, parameters of topography and parameters of soil The study results show that: if the evaluation process of ALES is robust in terms of the limiting factors, the evaluation process of LSE by GIS-MCA platform proved

to be superior in determining define larger areas of relevance Finally, the comprehensive adaptation assessment is a combination of 3 criteria: ecological criteria, environmental criteria and socio-economic criteria The results indicate that 64,505.9 ha (4.69%) in Western Nghe An area a suitable for orange cultivation and 69,388.8 ha (5.05%) are very suitable for tea cultivation Thus, the intentions

of planning and expanding the orange-specialized and tea-specialized areas can be fully met on the very appropriate land areas

Keywords: Land suitability; LSE software; Western Nghe An; Vietnam

1 Introduction

Mountain agriculture has always played an important role in enhancing food security and environmental sustainability in many countries around the world (Roozitalab

et al., 2013) The mountainous area is rich in biodiversity, however also a region with a high

concentration of poor people with the most unfavored living conditions and livelihoods in the world Therefore, sustainable mountainous development has been receiving more attention than ever Mountainous terrain occupies three quaters of Vietnam's land where a number of large rivers are originated (Quy, 1995) The mountainous region of Vietnam has many advantages to develop agricultural productions such as forestry, perennial crops, fruit trees, medicinal plants and livestock Despite of the great potentials, agricultural development are facing substantial constraints Therefore, the Government of Vietnam has made efforts towards policies, schemes and programs for socio-economic development of mountainous areas (VG, 2016, 2017) One of the most disadvantaged mountainous areas of Vietnam is Western Nghe An

The economy in the Western Nghe An predominantly depends on agricultural The land use structure is predominantly agricultural land 177,699 ha of agricultural production

land account for 12.9% of the whole area and 1,073,521 ha of forestry land accounts for 78.18% While, the total value of agricultural production is only approximately 16,647,682 million VND in 2017 (NASO, 2018) In 2017, the average product value obtained per 1 hectare of the cultivated land of Western Nghe An is low, 53.34 million VND The average number of Nghe An is 71.04 million VND, the average of Vietnam is 85.4 million dong Particularly, the value of products obtained on 1 ha of very low cultivated land in some districts such as Ky Son (16.26 million VND), Tuong Duong (21.11 million VND), and Que Phong (40.99 million VND) (NASO, 2018) The exploitation and use of agricultural land in the region is significantly not effective Currently, 99/217 communes in the region (about 45.6%) are benefited under Program 135 The large area of hills and mountains of the Western Nghe An region provides development opportunities to local agriculture, especially trees growing such as fruit trees and perennial crops To date, land are barely used for this purpose, most of the agricultural land used for low-value crops, which easily grow in lowland areas Approximately 107,641 ha (60.6%) of agricultural land is occupied

by food crops and annual crops such as rice, maize, cassava, sweet potatoes, peanuts, sesame and sugarcane Perennial crops such as rubber, tea, coffee and fruit trees including orange, mango, pineapple and longan, are grown on approximately 70,058 ha (39.4% of the area) (NASO, 2018)

Orange and Tea are two of the 10 major agricultural products, which have been focused on development in order to establish a large-scale production area, concentrating high-yield orange, specialty tea varieties of the province Nghe An as well as the Western Nghe An (PCNAP, 2016b) Orange trees were first planted in the late 19th century, then it has been developed to the famous orange areas such as Xa Doai and Bai Phu Oranges have become local specialties with geographical indications "Cam Vinh" established in 2007 The orange growing area in 2017 of the Western Nghe An is 4,872 ha (accounting for 87.2% of the whole orange growing area of Nghe An) Orange trees are grown mainly in Quy Hop (2,661 ha), Nghia Dan (1,035 ha), Thanh Chuong (388 ha), Con Cuong (361 ha), mainly 2 traditional orange varieties Xa Doai and Van Du The growing area for commerical products

is 2,060 ha with a total output of 32,087 tons (NASO, 2018) The Western Nghe An is also the oldest fresh tea area of Vietnam (Quy, 2003) It is an andvantage to be invested in intensive farming, expanding the area, developing into an commercial products for export Tea plays

an important role in poverty reduction and livelihood improvement to mountainous residents in the Western Nghe An The commercial growing area is 7,134 ha, of which the productive area is 6,054 ha, with a total output of 69,062 tons (NASO, 2018) The districts with large tea growing areas are: Thanh Chuong (4,317 ha), Anh Son (1,834 ha), Con Cuong (380 ha)

Orange and tea cultivation in Western Nghe An is facing a number of challenges Farming areas are scattered; many of them are not under controlled that results in low production efficiency, outbreak of pests and diseases, and unstable product quality It leads

to conflicts between different types of land use in the region Land us is neccessarily planned

in accordance with the potential and natural advantages of the region as well as farming practices, the farmers' production demand is a premise for agricultural development and stability It can barely take up references from studies on land evaluation in the Western

Nghe An and district level with the assessment objects being grouped into agricultural and forestry crops; for example (Khang, 1999; Hong, 2017) These results do not seem sufficient

to provide a scientific basis for the potentials or limitations of land or to support land use plans As a result, local people and local government have been confused in allocating land for high-value crops production for many years

Land evaluation is always considered a core component of land use planning (FAO,

1993; Baja et al., 2007; Singha and Swain, 2016) Physical land suitability is a prerequisite for

the development of land use planning, as it indicates decisions on land use catagory to take the most advantage, and supports to land management better, reduce soil degradation and

design effective land use models (Van Ranst et al., 1996; Nahusenay and Kibebew, 2015) On

other words, a part of land-use measures is land evaluation to support land use planning and rational use of natural and human resources (Rossiter, 1996) Different techniques and softwares have been built in order to further improve the accuracy of land evaluation These include Land Evaluation Computer System (LECS), Automated Land Evaluation System (ALES), Intelligent System for Land Evaluation (ISLE), and integrated expert system with multimedia (LIMEX), etc The evaluation used to be quite simple, the criteria in land evaluation were equally important Currently, the need for sustainable development makes analysis for appropriate land use increasingly complex due to the consideration of more requirements/criteria The weight value is applied to indicate the importance of each criteria for a specific object or study area Land evaluation conducted by Geographic Information Systems (GIS) and multi-criteria analysis has been widely used in recent years (Malczewski,

2006, Baja et al., 2007; Jafari and Zaredar, 2010; Mendas and Delali, 2012; Nguyen et al., 2015; Ennaji et al., 2018) Analytical Hierarchy Process (AHP) is a tool to assess the relative

importance of each criteria, the weights in AHP can be estimated by or geometric averages

or Fuzzy logic (Prakarsh, 2003; Kurterer et al., 2008; Akıncı et al., 2013; Kamkar et al., 2014; Zang et al., 2015; Widiatmaka, 2016; Maddahi et al., 2017; Jamil et al., 2018)

Each approach of land evaluation has different inputs and quality (Nahusenay and Kibebew, 2015) There are no rules indicating that a certain approach is adequate, or it need a more complex level of analysis In the context of Vietnam as a newly poverty-escaped country, the database of land resources is limited and asynchronous, especially large-scale data; limited funding makes detailed data analysis far to reach FAO’s land evaluation method applies automatic evaluation software such as ALES in combination with GIS to analyze the space

and map display The results are widely used such as D’haeze et al.,2005; Dinh, 2007, Ngoc et

al, 2013, etc Macro-level land evaluations have contributed in the first step of completing the

FAO’s land evaluation process in Vietnam and provided useful results as a basis for national land planning strategies and large ecological areas planning To date, land research and evaluation at the micro level, with the detailed evaluation object and study area is an urgent requirement It is in order to specify national land planning strategies and develop land use orientations for each locality Therefore, more complex quantitative land evaluation methods

have been implemented (Giap et al., 2005; Son and Shrestha, 2008; Khoi and Murayama, 2010; Nguyen et al., 2015)

This paper was conducted to evaluate the suitable land areas for orange and tea trees

in the Western Nghe An by two evaluation methods Firstly, a traditional and popular method in Vietnam: ALES (Automated Land Evaluation System) software Secondly, a new method which established on the GIS-MCA platform: LSE software (Land Suitability Evaluation) The study selected the same input data of soil, climate and topography for the two methods, thereby comparing two results and comparing with actual farming situation

to verify Then, assessing the overall land suitability, taking into account the socio-economic and environmental criteria to determine the most suitable areas for orange and tea development for the the Western Nghe An

2 Methodology

2.1 Case study

Nghe An is a province in the central position of the North Central region of Vietnam

It belongs to the Northeast of Truong Son Mountain Range, the topography is diverse, complex and divided by mountainous systems with three distinct ecological regions: mountainous, midland, coastal plain This study focused on the Western Nghe An area including 11 mountainous districts and towns of Nghe An province namely Thanh Chuong, Anh Son, Con Cuong, Tuong Duong, Ky Son, Que Phong, Quy Chau, Quy Hop and Tan Ky, Nghia Dan and Thai Hoa Town (

Figure 8 Study area in Vietnam territory

) The region has an area of 1,375,104 ha, about 84% of Nghe An province The Western Nghe An is characterized by two soil groups such as: Acrisols and Ferralsols, about 1,070,650 ha, accounting for 74.3% of the area Especially Rhodic Ferralsols built on basalt with soil structure and nutritions are very favorable for perennial crops and fruit trees

The Western Nghe An has a monsoon tropical climate with a cold winter The topography of the Truong Son mountain range substantially influences the monsoon circulation and creates the regional climatic differentiation There are two distinct seasons: hot, humid, heavy rain in summer and less rain in winter The mean annual temperature is about 23-24ºC at an altitude of less than 200m, corresponding to the total heat of 8,700ºC The maximum temperatures can reach extreme values up to 42-43ºC in April and May The temperature decreases with height, reaches 20ºC at an altitude of 700m, and about 15-18ºC at

an altitude of 1,100-1,700m The precipitation varies in a quite wide range from 1,000 mm to over 2,100 mm with 120-150 rainy days Rain regime in the area is divided into two distinct seasons: rainy season and less rainy season The rainy season coincides with the summer

monsoon season, lasting 6 months, from May to October (Ngoc et al., 2019)

Figure 8 Study area in Vietnam territory

2.2 Methods

2.2.1 Land suitability evalution in ALES

ALES (Automated Land Evaluation System) was developed in 1986-1996 by Rossiter and Wambeke (1997) ALES is an abundant expert system environment in land evaluation toolkit in the world FAO’s land suitability evaluation has started in Vietnam in the late 80s Based on the FAO land evaluation framework (1976), ALES is a land suitability evaluation software which firsly used in 2000 to support land use planning, especially for agricultural and forestry land To date, there are more evaluation methods but ALES is still very common Because ALES has a number of advantages that are relevant to the actual conditions in Vietnam such as:

The order of land evaluation according to FAO guidelines is applied with the integration of geographic information system GIS, ALES software and computer support ALES 4.65 software can be installed and run smoothly on Window 7-32bit With this program, users completely take the initiative in inputting data into the computer as required

by the program and it provides prompt results including the following parameters: Natural adaptation, yield estimation, analysis economic parameters

Input data for simple natural adaptation evaluation: are Land Use Requirements (LUR), the base unit for evaluation is Land Unit Map (LUM), including a set of Land units (LU) with attribute data After creating a Decision tree, importing LUM attribute data into ALES can promply provide appropriate results This is particularly advantageous when simultaneously evaluating multiple Land Utilization Types (LUT) or initiative attribute data including multiple complex LURs or large number of LUs Land suitability evaluation

results not only indicate the suitability of S1, S2, S3, N but also justify for ranking at that level in the form of Physical suitablitity subclass Steps on ALES are simple No complex interpolation or analysis is required

However, ALES itself does not have the functions in map display and spatial analysis Users are required to export ALES evaluation results to GIS to perform spatial analysis The classification of ALES seems too strict Only one attribute element of the LU does not meet the LUR in that class, then LU is immediately ranked lower This may be true

in the case that LURs is limited However, for normal or remedial LURs, this decentralization is too thorough This problem can be solved if ALES has a weighting tool, assigning weight to LURs to provide final result

The appropriate land evaluation steps in ALES is shown in Error! Reference source not found and summarized as follows: Develop the LUM by overlaying the selected input

data Select LUR of each LUT On ALES, set LUR of LUT with attributes corresponding to the attributes of LU Develop a Decision tree and selecting maximum limitation for the suitability classification Import LU definition from xBase to ALES Perform suitability calculations, ALES allows the selection of LU to evaluate, the evaluation results are displayed in the form of Physical suitabilitiy subclass Export and save evaluation results from ALES into data field in LUM database, then set up suitability evalutaion map and calculations in GIS

2.2.2 Land suitability evalution in LSE

LSE is the first software in Vietnam to be developed for land suitability evaluation

It is integrated in Visual Basic, it is stand-alone software that follows the land evaluation

process based on multi-criteria analysis in GIS environment of Nguyen et al., 2015 LSE is

based on open source GIS - DotSpatial It performs all operations in Geotif format Eventually, two converters were built with conversion vectors (.SHP input format) and raster data (Grid, IMG and BGD file formats) in Geotif LSE includes the main parts: (1) Interface with input fields and operation keys; (2) Spatial database including maps; (3) Model of land evaluation to carry out calculation and production of results maps; (4) Supporting tools: Data conversion, Slope data building, Grouping, Standardizing, Verifying results, Viewing maps)

The steps for conducting an land suitability evaluation in LSE is shown in Error! Reference source not found and is described as follows: Develop criteria data according to

ecological adaptation table and select standardized functions Use algorithm functions in software to standardize data The standardized data was developed by: the S membership,

a trapezoidal, the Kandel function Determine and calculate weights The weights are determined by the rank reciprocal equation Evaluate component adaptation according to each criteria group Overall suitability evaluation according to all criteria groups

LSE is a compact, scientific-based and flexible software with friendly interface It is very useful in evaluating regional land suitability, it saves time and ensures accuracy and objectivity LSE is a new software, the influence calculation tool has not been fully integrated into the software While running, users need to use an additional supporting software to

calculate The selection of the standardized function is difficult for users Automatic selection feature has not developed yet

Figure 2 Process of the study

2.2.3 Method of evaluating soil erosion and mapping orange and tea areas

Annual soil erosion for orange and tea cultivation is estimated according to the

formula Universal Soil Loss Equation (USLE) developed by Wischmeier et al (1978) and

shown Figure 9

A= R x K x L x S x C x P (1)

Where, R factor is calculated by the fomula: R= 0.082 x P - 21 of Lai (1999), P is annual rainfall Soil erodibility factor K is estimated based on the interaction with soil organic

matter (OM) of Stone et al (2012) (

Table 10) LS factor is is calculated by the fomula: LS = (L/22,1)0,7(6,432sin(S0,79)cos(S) Where, L is length of slope (m); S is slope angle (radian) Map of LS coefficient is referenced

by Dung et al (2013)

Study to apply cover management factor C=0.2 for tea trees according to Kuok et al (2013), C=0.13 for orange trees according to Shi et al (2004) The support practice factor P applied to tea trees is 0.6 Kuok et al (2013) and 0.5 for orange trees according to Prasannakumar et al (2012)

(a) (b)

Figure 9 Annual soil erosion corresponds to growing orange (a) and tea (b)

Table 10: Estimates of K factor based on soil texture and organic matter (OM) content

Average OM<2% OM>2%

2.3 Data

The criteria selected to land evaluation of the Western Nghe An for oranges and tea, were divided into three parameters There are parameters of climate (annual precipitation, mean annual temperature, length of dry season, mean temperature of 2 months after harvest, mean temperature of flowering period of orange trees); parameters of topography (slope, flooding, drainage); parameters of soil (soil depth, soil texture, pH, OM, CEC, available phosphorus, available potassium) Soil physical data including soil depth and soil texture were taken from soil map of Nghe An province at scale 1:100,000 (NIAPP, 2004) (Figure 10) Soil nutrient criteria: pHH2O, OM, P2O5, K2O, CEC are calculated from the results of chemical analysis of 79 soil profiles in the study area and surrounding areas

(a) (b) (c)

Figure 10 Spatial distribution of: (a) Slope, (b) Soil depth, (c) Soil texture, (d) Drainage, (e)

Flooding

Value of pHH2O, OM and available potassium, available phosphorus is the result of

calculating the average of soil layer 0-25 cm (Sys et al., 1991) These nutritional criteria of the

study area are shown in Figure 5

(d) (e)

Figure 11 Nutritional criteria of the study area: (a) Soil pH, (b) Soil Organic Matter, (c) Soil CEC,

(d) Available potassium, (e) Available phosphorus

Climate data include annual precipitation, mean annual temperature, length of dry season, mean temperature of 2 months after harvest, mean temperature of flowering stage (Figure 12) These maps were established on a series of monitoring data including: precipitation data of 22 regional hydrological stations; especially temperature and precipitation data at 04 meteorological stations in Con Cuong, Tuong Duong, Quy Hop and Quy Chau for a period of 35 years from 1980 to 2014 (DCG, 2016) The mean temperature of flowering of orange trees in the Western Nghe An is calculated as the mean temperature in January and February The mean temperature of 2 months after harvest of orange trees is calculated as the mean temperature of 3 months (December, January, February)

Topography data: Slope angle were extracted from the 30m resolution DEM (Figure 10) Flood data is referenced from flooding map of the North Central region at scale 1:250,000 from Thuy, (2015) and presented in Figure 10e On the basis of soil map, drainage

is determined based on the occurrence of the glei layer in soil profile, shown in Figure 10d

(Sys et al., 1991)

In addition, to make an overall suitability evaluation in the LSE, there is also data on the distance to the market, the distance to the road Market data and traffic axes are taken from the topographic map of 1: 50,000 scale and supplemented from the land use map at 1: 10,000 scale in 2016 Then, the distance to the market and distance to the road are calculated

by Euclidean Distance in Arcgis 10.1

3.1 Ecological requirements for orange and tea in the Western Nghe An

Table 10 summarizes 15 land suitability criteria for orange cultivation in the Western Nghe An These are selected based on the natural conditions of the region and classified

according to Sys et al., (1993) Particularly, 2 important criteria of orange cultivation, which

directly affects orange yield, fruiting rate and fruit quality, namely: mean temperature of 2 months after harvest and mean temperature flowering stage Criteria of OM and cation

exchange capacity CEC are taken from Olusegum et al., (2015) Available phosphorus and

available potassium were specially selected to assess orange trees in the Western Nghe An according to FAVRI, (2011)

Very suitable soil (S1) for orange trees is well-drained soil, not flooded, pH of 7.6; OM > 1.8%; CEC > 16.0 cmol (+)/kg of clay, available phosphorus > 5.0 mg/100g soil, available potassium > 7.0 mg/100g soil Slope of 0°-8°, soil depth >100cm The very suitable soil texture for orange: sandy clay loam (SCL), sandy loam (SL), loamy sand (LS), and clay loam (CL) The mean annual temperature suitable for orange cultivation is over 19-24oC, in

5.5-the condition that R ≥ 1,200mm, 5.5-the mean temperature of two months after harvest is from 8-18oC, the mean temperature of flowering stage > 10oC and dry season less than 4 months

The soil is unsuitable for orange (N) if one or more criteria are at the limit It means annual precipitation > 800 m; length of dry season over 6 months; mean annual temperature

< 16oC Slope > 30o, poor drainage or flooding, soil depth < 50 cm are unsuitable for orange The soil has pH <5.0; OM < 0.8%; CEC < 10.0 cmol (+)/kg of clay should not be planted with orange

Table 11 Agro-ecological, environmental and socio-economic criteria for orange cultivation in the

Western Nghe An

Very suitable (S1)

Moderately suitable (S2)

Marginally suitable (S3)

Unsuitable (N)

Climate

Annual precipitation (mm) > 1200 1000-1200 800-1000 <800 Number of dry months

Mean temperature in the 2

months after harvest (oC)

Physical soil fertility

b C: clay; CL: clay loam; SCL: sandy clay loam; SL: sandy loam; LS: loamy sand

c (a) Soil depth <50 cm; (b) Soil depth 50-70 cm; (c) Soil depth 70-100 cm; (d) Soil depth

Table 12 Agro-ecological, environmental and socio-economic criteria for tea cultivation in the

Western Nghe An

Very suitable (S1)

Moderately suitable (S2)

Marginally suitable (S3)

Unsuitable (N)

b C: clay; CL: clay loam; SCL: sandy clay loam; SL: sandy loam; LS: loamy sand

c (a) Soil depth <50 cm; (b) Soil depth 50-70 cm; (c) Soil depth 70-100 cm; (d) Soil depth

3.2 Evaluation of ecological suitability by ALES

The results of land suitability evaluation by ALES for orange showed that there was

no very suitable land (S1), Moderately suitable (S2) accounted for about 16.6% (228,439 ha), concentrated in Quy Chau, Quy Hop and Tan Ky Marginally suitable (S3) and Unsuitable (N) occupy large areas, accounting for 41.3% and 38.4% respectively (Table 13) The limiting criteria in these areas are mainly steep slopes (> 15o), soil depth (50-70 cm for S3, <50 cm for

N Soil is flooded or poorly drained also Not suitable for orange, other criteria such as pH

<5.0, OM < 0.8%, CEC <10.0 cmol (+)/kg The spatial distribution of the suitability levels is shown in Figure 13

Table 13 Results of land suitability evaluation for orange cultivation by ALES

1 S2 S3 N