|
|
|
4th International Conference on Integrating GIS and Environmental Modeling
(GIS/EM4):
Problems, Prospects and Research Needs. Banff, Alberta, Canada, September
2 - 8, 2000.
Integrating climate, soil and crop information:
a land suitability study using GIS
GIS/EM4 No. 129
Abdolali Ghaffari
Hadrian F. Cook
Howard C.Lee
Abstract
Assessing the suitability of an area for crop production requires a considerable effort in terms of information collection that presents both opportunities and limitations to decision-makers. A GIS has been used to match the suitability for main crop potatoes based on the biological requirements of the crop and the quality and characteristics of land. The Stour Catchment, Kent, UK is the study area. The methodology adopted combines climate, and those land quality attributes that most influence crop suitability (long-term average annual rainfall, accumulated temperature, field capacity duration, topography data (slope and altitude) and soil-water deficit). Good management is assumed; including the use of appropriate crop varieties, fertilisers and sowing date and irrigation. Social and economic factors are excluded. Overall suitability is recognised by the Simple Limitation Approach (SLA) in preference to a weighted GIS model that scores attributes. The results show that 58% of the area is 'highly' or 'moderately' suitable for main crop potatoes.
Keywords
Land suitability, potato, climate, soil, Simple Limitation Approach (SLA), GIS.
Introduction
It has long been recognised that land suitability is assessed
as part of a 'rational' cropping system (FAO 1976) and optimising the use
of a piece of land for a specified use (Sys et. al. 1991) should be based
upon its attributes (Rossiter 1996). Furthermore, land may be considered
either in its present condition, or after specified improvements. Although
criteria may vary, they are essentially based on climate, soil, topography
and water availability which are the most important categories of environmental
information required for judging land suitability.
The integration of climate-soil-site modelling using Geographic Information
Systems (GIS) is finding increasing application in crop-specific modelling
of agricultural production. It is necessary to establish the best regime
for the range of conditions which may be sub-optimal (but still acceptable),
and also those which are deemed unsatisfactory. For instance, Van Lanen
et. al. (1992) assessed the suitability of Dutch land for potato by a mixed
qualitative/quantitative evaluation approach, and found about 65% of the
area to be potentially suitable. This 'mixed evaluation' approach predominantly
uses geographically referenced input data; therefore a linkage with GIS
is a prerequisite for its efficient use. Liengsakul et. al. (1993) compared
GIS and digital remote sensing methods for locating new sites for permanent
cropland in the highlands of northern Thailand with the older methods currently
in use. The results show that, apart from considerable time saving and improvements
in data quality, the integrated information can locate potential new cropland
sites. Boje et. al. (1998) calculated the actual and potential lowland suitabilities
for nine different food crops in NW-Tanzania, finding the local food crop
priorities disagree strongly with climatic suitability.
This study presents a GIS-based land suitability model based upon a 'Simple
Limitation Approach (SLA)'. This evaluates the capabilities of the study
area for main crop potatoes, as a main summer crop cultivated under temperate
conditions. Suitability for this crop was determined, based on matching
the biological requirements of crop to the quality and characteristics of
land within the Stour Catchment, Kent, UK. The methodology adopted combines
most aspects of climate, site and soil attributes that influence crop suitability
appropriate to Western Europe.
Material and methods
Study area
This is the Stour Catchment and the Isle of Thanet in the north-eastern corner of Kent, UK. It is located between 51º 04' to 51º 24' N and 00º 40' to 01º 27' E. The area exceeds 1000 km2 and altitude varies, between 0 and 201 m above Ordnance Datum.
Soil
Complete coverage of soil survey data is only available for England and
Wales at a scale of 1:250,000 (Potter et. al. 1993) and soil information
at this scale is presented in terms of soil association (geographically
associated soils) rather than a true soil series (describing soils alike
in soil characteristics behaviour). Drift geology maps were therefore used
as a surrogate for soil survey data (Cook 1991, Osborn and Cook 1997), because
these maps provided better spatial resolution (1:50,000) than soil survey
maps (1:250,000) and drift geology is convertible into soil survey criteria.
Whole or some parts of solid and drift geology maps of sheets: 273, 274,
288, 289, 290, 305 and 396 at a scale of 1:50,000, published by the Geological
Survey of Great Britain (England and Wales), were digitised and stored in
GIS. These maps, with more than 30,000 polygons, were converted to a soil
base map.
Eighteen soil series are distinguished in this study. Soil attribute data
(soil texture, pH, etc.) have been compiled from previous accounts (Jarvis
et. al. 1984, Green and Fordham 1973, Fordham and Green 1980, Totolo 1995).
Characteristics of the dominant or 'lead' soil series were either provided
by the soil profile data (e.g. soil texture and depth as single factor maps),
or were derived by an interpretation procedure (e.g. drainage and leaching
classes). For instance, drainage class is determined based on geology and
parent material, with Class I being the best drained land and class IV the
worst.
Topography
Topographical maps are used to select site slopes and altitude information relevant to land suitability. This study used a landform panorama Digital Terrain Model (DTM) of raster format, 10 m resolution, supplied by the Ordnance Survey of Britain.
Climate
The most important climate characteristics are temperature, rainfall, radiation balance, wind, relative humidity and evaporation. Average accumulated temperature above 0 °C between January to June (1961-1980), field capacity duration (FCD) and also, soil water deficit (1941-1970) for both crops were recorded for each 5 km intersection of the National Grid and accessed via the MORECS database (The Met. Office 1989). This information was used as point data for GIS in order to generate agro-climatic maps.
Interactive limitations
The physical limitations which result from interactions between
climate, site and soil, are the extremes of wetness and droughtiness. For
land suitability purposes, these were assessed separately by relating soil
profile characteristics to appropriate climatic parameters.
Wetness is an extremely important characteristic because it combines soil,
topographic and climate interactions. It takes into account the drainage
class of the profile, the texture of the topsoil and the climatic parameter
of the duration of field capacity (FCD). A combination of these variables
affects land suitabilities, presented in Table 1. For instance, the wetter
the soil (the higher the drainage class), the heavier the topsoil texture
and the wetter the climate (high field capacity duration days) the more
likely the soil is to be assessed as unsuitable because of limitations and
poorer crop growth. 'Droughtiness', on the other hand, is a consequence
of complicated interactions of the available water capacity of the soil,
crop water requirement and climate (Cook and Dent 1990). Comparing the ability
of a soil profile to supply water to crops with dryness of the climate during
the growing season, gives an assessment of droughtiness limitations. The
method used here to assess droughtiness was based on Thomasson (1979).
Land characteristics and qualities as a basis for evaluation
A mixture of land characteristics (simple attributes of the land that can be directly measured or estimated e.g., topsoil texture) and land qualities were used in this project to indicate the degree of suitability of land for main crop potatoes. Land qualities represent complex attributes of land that acts in a distinct manner in its influence on the suitability of the land for a specified kind of use e.g., soil water availability. The overall suitability is expressed in three classes: highly suitable (HS), moderately suitable (MS) and marginally suitable (MG). Moderately suitable and marginally suitable land was expected to have a crop yield of 60-80% and 40-60% of the yield under optimal conditions with practicable and economic inputs, respectively. Unsuitable (U) land was assumed to have severe limitations which could rarely or never be overcome by economic use of inputs or management practices (FAO 1976, Dent and Young 1981).
Crop characteristics
Crop-specific properties such as the physiological and phenological crop parameters were gathered from literature review, particularly from Jarvis et. al. (1985), MAFF (1988), and Sys et. al. (1993). Climatic, edaphic and site requirements for selecting land suited for the cultivation are shown in Table 1.
Geographical Information Systems (GIS)
The GIS approach used in this study identifies input data for the land suitability models and develops a modelling procedure for processing and output presentation. Digitised maps, the geographical distributions of soils, topography and agroclimatic regions were captured together with attribute data (e.g. soil texture, soil depth for each mapped soil unit). Overlaying was carried out using SPANS and ARC/VIEW softwares. The results are presented as tables and maps. Overall suitability is recognised by the Simple Limitation Approach (SLA) illustrated in Figure 1. This method utilities the concept of "most limiting factor" which corresponds to Leibig's "Law of the Minimum". An example path is displayed in dotted lines. Here, a combination was deemed "highly suitable" in the first four factors. However, soil depth was found to be "moderately suitable" and the slope "unsuitable". The sieving process therefore reported the combination as "unsuitable" in terms of overall suitability (dotted line). It was used in preference to a weighted spatial GIS model derived from scoring attributes (cf. Cook 1991), because it is seldom clear how to derive the weightings of successive overlays.
|
|||||||||||||||||||||||||||||||||||
|
|
AP: crop-adjusted available water capacity of the soil profile, WD: soil water deficit, HS: highly suitable, MS: moderately suitable, MG: marginally suitable, U: unsuitable
|
|
|
|
Findings
Altitude and slope
In the first instance, suitability was assessed in terms of topography. The highest recorded elevation in the study area was 201 m O.D. Elevation alone did not affect land suitability since the whole study area was highly suitable for this crop under consideration. Neither did slope affect land suitability very much. About 4% of the area was marginally suitable with slopes between 7 to 11%. Just 1.23% of the study area had steepest slopes (more than 11%) which were unsuitable for crop production in general.
Wetness
A map of 'wetness' as a criterion for assessing suitability for this crop was produced by overlays of field capacity, drainage class and soil texture. Approximately 66% of the study area was found to be 'highly suitable' with very little 'unsuitable' (Table 2). The 'marginally suitable' land is concentrated in the north-west, south-west and north-east coastal area.
Droughtiness
A 'droughtiness' map (not shown) was produced by overlaying available water and water deficit maps. Table 2 shows 61.23% and 26.92% of the area is placed in 'highly suitable' and 'moderately suitable' categories, respectively. Only 1.99% of the study area was unsuitable and the remainder (9.86%) was in the marginal category.
Accumulated temperature
Accumulated temperature is an important variable affecting land suitability for potato. Although there are no 'unsuitable' areas attributed to this factor, only 30.3% of the study area was 'highly suitable' (Table 2) with 63.6% of the area 'moderately suitable.
|
||||||||||||||||||||||||||||||||||||||||
|
|
Soil depth
The study area contains no unsuitable portions for this crop, but only 74.02% of the area was classified as 'highly suitable'. Around 10% of the area was 'moderately suitable' with the remainder marginally suitable.
Soil pH
Regarding soil pH, 46.26% of the area was 'highly suitable' and minuscule
amount was unsuitable.
The suitability map for main crop potatoes under water-limited (rainfed)
conditions was created by overlaying accumulated temperature (January-June),
droughtiness, wetness, slope, soil pH and soil depth maps (Figure 2). As
Table 2 shows, 10% and 47.7% of the study area is 'highly' and 'moderately
suitable', respectively. Some 36.89% of the study area was marginally suitable
and the remainder unsuitable.
Discussion
Aspects of land suitability for main crop potatoes are determined based
on climate, soil and topographic variables. Combining physical land evaluation
models through a GIS improves land evaluation models and enables an analysis
more relevant to policy-making than the original basic data.
Highly suitable areas have a high potential production and sustainability
of yield from year to year. In average years there is an opportunity for
establishment at, or near, the optimum sowing time, while harvesting is
rarely restricted by poor ground conditions. Even in wet years (up to a
frequency of about one in four) working conditions are acceptable and do
not prevent crop establishment; yet there are normally sufficient soil water
reserves to meet the average requirements of the crop. Moderately suitable
areas can allow high or moderate potential crop production, which can be
lower in years when soil-water is insufficient to sustain full growth, or
when crop establishment is unsatisfactory due to untimely sowing or poor
soil structure.
|
|
|
Marginally suitable areas are those with variable potential production
from year to year, with considerable associated risks of low yields, high
economic costs, or difficulties in maintaining continuity of output, which
are due to the climate interacting with soil properties or disease and pest
problems (Jarvis et. al. 1984, Jones and Thomasson 1987). In this study,
unsuitable areas did not meet the criteria of high, moderately and marginally
suitable areas. For instance, a slope of greater than 11% is too steep for
efficient use of machinery.
The criteria, which defined unsuitability in this study area, were based
on climate and slope rather than on soil properties. Dent and Young (1981),
assumed that under rainfed agriculture, expected crop yields (as a percentage
of yields under optimal conditions) were more than 80%, 40-60%, 20-40% and
less than 20% in high, moderate, marginal and unsuited areas, respectively.
Because in the present study, evaluation is based on an average of 30 years
of climate data, we may also assume that expected crop yields are close
to the potential production during more than 80%, 40-60%, 20-40% and less
than 20% of years for high, moderate, marginal and unsuited areas, respectively.
All the factors that affect land suitability for main crop potatoes in
the study area are summarised in Table 2. Average accumulated temperature
above 0 °C between January and June (the first 6 months of the year) uses
the criteria of McRae (1988) considered to be a good measure of the heat
energy available for plant growth in Britain. This is affected by planting
time, elevation of growth and storage, season, genetics, and interaction
with cultivation techniques. Table 2 shows that the accumulated temperature
criterion was found to be a limiting factor for this crop. Jarvis et. al.
(1985) assumed, for England and Wales, that only regions with more than
1550 °C day degrees above 0 °C (January-June) were 'highly suitable' for
potato. Average maximum accumulated temperature in this area was less than
1500 °C therefore, by this variable, the study area could not be placed
in the 'highly suitable' category. However, potatoes are grown in the catchment
and there are some very suitable areas, especially in the north of the area
(McRae 1999 personal communication). It was therefore decided to change
the recommended accumulated temperature for original 'highly suitable area,
(at 1550 °C) to new criteria according to Table 1. By the new criteria,
30.33% of the area is found to be 'highly suitable'.
Doughtiness limits the areas of suitability for this crop. Highly suitable
areas amounted to 61.23% (Table 2). Because potatoes are shallow-rooting
and unable to exploit soil water from below about 0.7 m depth (Durrant et.
al. 1973), droughtiness is frequently a main limitation to crop yield unless
irrigation is applied. Harris (1992) summarised information on rooting depth,
the maximum length and mass of root systems of potato crops and concluded
that the depth of rooting of all potato crops never exceeded 1 m even on
deep and uniform soils.
The duration and degree of waterlogging are described by the system of
wetness classes, where wetness class I is highly suitable and well drained
to wetness class IV is unsuitable and almost permanently waterlogged within
0.4 m depth (Hodgson 1976). One of the advantages of wetness classes, however,
is that the drainage status of a soil can be expressed -not only in its
natural state but also how it might be following drainage improvement. This
factor restricts land suitability. Slowly permeable horizons are those with
textures which are clayey (clay, sandy clay and silty clay), fine loamy
(sandy clay loam) or fine silty (silty clay loam). Wetness can affect workability,
trafficability and aeration conditions. Soils could not be worked during
the field capacity period because, during this period, there was not any
soil water deficit and the soil was too wet for any machinery operation.
Thus, land with a high wetness class can limit cultivation in the early
spring, particularly for early potatoes.
Most soil series are suitable for this crop but some shallow soils are
only marginal, and rooting can be evaluated according to the soil depth
criterion. Higher soil pH is restricted to highly suitability areas. Overall,
acidity and aluminium saturation are not important in the area because only
about a quarter of one percent of the study area predicted that pH was less
than 5. Marginally suitable areas with respect to pH related to higher pH
(8.1-8.2).
Slope, an important element of landform, plays an important role where
mechanisation is concerned. Sys et. al. (1991) believe that, on slopes steeper
than 20%, mechanisation becomes impossible and for slopes less than 20 percent
there are still important variations in productivity according to variation
in slope. Navas and Machin (1997) state that, in order to avoid soil erosion
and other problems derived from the use of machinery, only land with slopes
below 8° should be used. Fortunately, most of the Stour Catchment was found
suitable with respect to topography; only 1.23% had the steepest slope category
and was therefore unsuitable for full-mechanised cultivation.
Conclusion
In general, the climate in the study area is favourable for arable crops
such as winter cereals, spring barley, oilseed rape and potatoes. Land classified
as highly and moderately suitable can sustain production of good quality
ware potatoes in arable rotation. As potatoes are shallow-rooting and unable
to exploit soil water from below about 0.7 m depth, droughtiness is often
the main limitation to crop yield. On well-drained soils the crop responds
economically to irrigation, and allowance is made in the suitability assessment
for upgrading easily worked but droughty land where irrigation is available.
Some parts of the north-west and south-west in the study area have marginal
suitability for this crop in the case of water deficit, heavy texture soils
and low soil pH.
This study furthermore shows that with appropriate soil, topographic and
climate information, GIS used in a discerning fashion is a powerful tool
for agronomic decision making.
Acknowledgments
This research was supported by a fellowship for the lead author from the Islamic Republic of Iran, Ministry of Agriculture. We would like to thank Drs Stuart McRae and Paul Burnham for their technical advice; Kris Roger, and Jim Connor for their GIS assistance. Geological maps were digitised from Geological Survey maps of Britain (England and Wales) with permission.
References
Boje G, Rucker G, Senzige S, Skowronek A. 1998. Land suitability for crop diversification and yield potential of a drained swamp area in NW-Tanzania. Tropenlandwirt 99 (1): 65-75.
Cook HF, Dent DL. 1990. Modelling soil-water supply to crops. Catena 17(1): 25-39.
Cook HF. 1991. Nitrate protection zones, targeting and land use over an aquifer. Land Use Policy, January 1991: 16-28.
Dent D, Young A. 1981. Soil survey and land evaluation. Allen and Unwin, London.
Durrant MJ, Love BJG, Messem AB, Wdraycott AP. 1973. Growth of crop roots in relation to soil moisture extraction. Annual Applied Biology 74: 387-394.
FAO. 1976. A framework for land evaluation. Soils Bulletin, No. 32.
Fordham SJ, Green, RD. 1980. Soils of Kent. Soil survey bulletin, No.9, Rothmasted experimental station.
Green RD, Fordham SJ. 1973. Soils in Kent 1. Soil survey record No. 14. Sheet TR04 Ashford, Harpenden, UK.
Harris PM. 1992. The potato crop the scientific basis for improvement. Second edition, Chapman & Hall, London, UK.
Hodgson JM. 1976. Soil survey field handbook. Soil survey technical monograph, No. 5. Soil survey of England and Wales, Harpenden Herts.
Jarvis MG, Allen RH, Fordham SJ, Hazelden J, Moffat AJ, Sturdy RG. 1984. Soils and their use in south east England. Soil survey of England and Wales.
Jarvis MG, Batey T, Ragg JM. 1985. An agroclimatic databank for England and Wales. Soil survey of England and Wales.
Jones RJA, Thomasson AJ. 1987. Land suitability classification for temperate arable crops. In: Beek KJ, Burrough PA, McCormack DE, editors. Quantifield land evaluation procedures. ITC publication (1): 29-35.
Liengsakul M, Mekpaiboonwatana S, Pramojanee P, Bronsveld K, Huizing H. 1993. Use of GIS and remote sensing for soil mapping and for locating new sites for permanent cropland - a case study in the highlands of northern Thailand. Geoderma 60(1-4): 293-307.
MAFF. 1988. Agricultural land classification of England and Wales. Ministry of Agriculture, Fisheries and Food, UK.
McRae SG. 1988. Practical pedology studying soils in the field. Ellis Horwood Limited, Chichester, England.
Navas A, Machin J. 1997. Assessing erosion risks in the gypsiferous steppe of Litigio NE Spain. An approach using GIS. Journal of Arid Environments 37(3): 433-441.
Osborn S, Cook HF. 1997. Nitrate vulnerable zones and nitrate sensitive areas: A policy and technical analysis of groundwater source protection in England and Wales. Journal of Environmental Planning and Management 40 (2): 217-233.
Potter C, Cook HF, Norman C. 1993. The targeting of rural environmental policies: An assessment of agri-environmental schemes in the UK. Journal of Environmental and Management 36(2): 199-216.
Rossiter DG. 1996. A theoretical framework for land Evaluation. Geoderma 72: 165-190.
Sys Ir C, van Ranst E, Debaveye J. 1991. Land evaluation. Part I. Principles in land evaluation and crop production calculations. International training centre for post-graduate soil scientists, University Ghent.
Sys Ir C, van Ranst E, Debaveye J, Beernaert, F. 1993. Land evaluation. Part III. Crop requirements. International training centre for post-graduate soil scientists, University Ghent.
The Met. Office. 1989. Climatological data for agricultural land classification. UK.
Thomasson J. 1979. Assessment of soil droughtiness. Soil survey, Tech, Monograph, p.43-50.
Totolo O. 1995. The use of a geographic information system GIS (SPANS) to facilitate detailed evaluation of soil and land. PhD thesis, Wye College, University of London.
Van Lanen HAJ, Hacktenbroeke MJD, Bouma J, Degroot WJM. 1992. A mixed qualitative quantitative physical land evaluation methodology. Geoderma 55(1-2): 37-54.
Authors
Abdolali Ghaffari, Assistant Researcher in Crop Production
West Azerbaijan Agricultural Research Centre, P.O. Box 365, Orumieh, Iran
Email: abdolali_ghaffari@yahoo.co.uk, Tel: +98-441-775480, Fax: +98-441-771253
Hadrian F. Cook, Senior Lecturer in Hydrology
Imperial College at Wye, Wye, Ashford, Kent TN25 5AH, UK
Email: H.Cook@ic.ac.uk, Tel: +44-1233-812401, Fax: +44-1233-812855
Howard C. Lee, Senior Lecturer in Crop Production
Imperial College at Wye, Wye, Ashford, Kent TN25 5AH, UK
Email: H.Lee@ic.ac.uk, Tel: +44-1233-812401, Fax: +44-1233-812855