Madsen, Henrik Breuning & Holst, Kristian Aagaard: A nationwide
mapping of dry soils for plant production - A case study
from Denmark. Geografisk Tidsskrift 88: l-8. København, 1988.

For economical or environmental reasons the arable land in Denmark will probably decrease in size in the nearby future. Therefore some areas will change into marginal land, especially wetlands, undulating areas and areas with dry soils. A nation-wide mapping of those areas has been carried out, and this paper describes the methodology used for mapping dry soils for grass and barley production.

Keywords: Marginal land, Root zone capacity.

Henrik Breuning Madsen, lektor, dr.scient, Geographical Institute,
University of Copenhagen, Øster Voldgade 10, DK-1350
Copenhagen K.

Kristian Aagaard Hoist, Vid.ass., lic.agro, Ministry of Agriculture,
Bureau of Land Data, Enghavevej 2, DK-7100 Vejle.

Being too poor for farming, some areas in Denmark have changed from agricultural land to forest or scrub during the last century (Jensen, 1976). These soils are called marginal land. Because of surplus production of agricultural products within the EEC and for environmental reasons the arable land in Denmark is expected to decrease significantly during the next decades. Hereby large areas will be marginal and change from intensively cultivated land to forests, grassland or more specific biotopes.

In order to elucidate what areas will become marginal land in the future, Jensen & Kock (1987) questioned farmers of their reasons for changing agricultural land to marginal land. The main reason was that the soil was too wet, too dry or too steep. In 1986 a nation-wide mapping of these soil types was carried out (Madsen & Holst, 1987), and this paper describes the methodology used for mapping dry soils, defined by the root zone capacity for barley and grass production.

METHODOLOGY

The method used for calculating and mapping the root zone capacity is as follows. First, approximately 36,000 soil profiles were constructed with information on texture and organic matter content at three depths: 0-30 cm, 30-60 cm and 60-120 cm. The available water content in the soil layers was calculated from regression equations combining the water content at field capacity and wilting point with texture and organic matter content. An effective root depth is defined for different crops in relation to soil type, and on the basis hereof the root zone capacity is calculated.

The calculations are exclusively based on computerized nation-wide soil maps combined with analytical databases containing point measurements. In the following, a short presentation of the databases used will be given, followed by a more detailed presentation of the different steps in the calculation of root zone capacity.

The soil database

The soil database contains information from the Danish soil classification and from later investigations improving this classification. Furthermore, information from earlier mappings, e.g. wetlands, has been computerized. All information has been digitized with reference to the UTM coordinate system in order to provide facilities for production of computer-drawn soil maps at different scales and with different combinations of parameters.

The soil data can be divided into two main types, parameter data and line area data (Platou, 1984) of which the following have been selected for mapping the root zone capacity.

Parameter data

There are two major types of parametric data used for the mapping: information on texture and organic matter from the Danish soil classification and analytical data from pedological investigations.

The primary data in the Danish soil classification are texture analyses. From approximately 36,000 sites soil samples were taken by local agronomists. At each site a sample was taken from a depth of 0-20 cm and at selected sites from a depth of 35-55 cm. Texture, organic matter content and calcium carbonate were determined in all samples. The following grain sizes (_[l m) were determined <2, 2-20, 20-63, 63-200 and 200-2000. The hydrometer method was used for determining the grain sizes < 2 and 2-20, while sieving was used for the others. The organic matter content was determined on an IR-Leco apparatus and calcium carbonate content was determined by dilute acid titration (Mathiesen & Nørr, 1976). Since 1979 this database has been increased with several thousand

analyses, e.g. from field experiments carried out on
private farmland by local agronomists.

Pedological investigations have been carried out along the main gas pipeline system from the North Sea gas field across Denmark. More than 800 profiles have been excavated and described (Madsen & Jensen, 1985). On an average five to six samples were taken from each profile. In the laboratory, texture, organic matter content, calcium carbonate content and pH were determined on nearly all samples, and from selected profiles dithionitecitrate and pyrophosphate-soluble iron and aluminium, CEC, exchangeable bases, total N and P, mineralogy, soil water retention and root densities were measured.

Line area data

Four nation-wide line area databases with soil and climatic
information have been used for mapping the root zone
capacity. These are:

- soil boundaries from the Danish soil classification,
- boundaries between loamy and sandy subsoil,
- landscape boundaries drawn on topographic maps,
- boundaries between wetlands and non-wetlands

In 1975-84 soil maps covering Denmark were produced at 1:50,000. They were mainly based on texture analyses of the topsoil (0-20 cm) and, according to these, the agricultural land was divided into eight classes, Table 1. The soil maps also give information on the location of forest areas and urban zones (Mathiesen, 1980). Fig. 1 shows the location of sandy (MCC 1,2), loamy sandy (MCC 3), loamy (MCC 4,5,6) and peaty (MCC 7) topsoils in Denmark. In the western part of the country sandy soils dominate, while in the eastern part loamy soils dominate.

A map showing the location of loamy and sandy subsoil within non-marine landforms has been constructed based on a geological map at 1:500,000 (Bornebusch & Milthers, 1935) and on other geological maps at 1:100,000. Fig. 2 shows the location of sandy and loamy subsoils within the non-marine landforms in Denmark. The eastern part of the country has mainly loamy subsoil, while the western part has mainly sandy subsoil.

A landscape map has been produced at 1:100,000. It is based on topographic maps, earlier landscape maps (Schou, 1949; Smed, 1979) and geological surveys published at 1:100,000. The country is divided into nine different landforms: morainic areas from the Saale glaciation, morainic areas from the Weichsel glaciation, outwash plains, dune sand areas, old marine deposits (Yoldia), marsh areas, younger marine forelands including littorina, reclaimed areas, and rock. Some mixed areas were outlined where it was not possible to separate the landforms. Fig. 3 shows the location of the different landforms in Denmark.

A wetland map was outlined from the landscape map and from old topographic maps (1:20,000) showing the extent of wetlands 60-80 years ago. From the landscape map, marsh areas, younger marine forelands including littorina, and areas below sea-level are considered as wetlands, while in the other six landforms all areas with wetland signature on the old topographic maps are considered as wetlands. These old maps were preferred for younger ones because of the recent decrease in wetlands due to drainage. Fig. 4 shows a nation-wide wetland map. The wetlands cover approximately 20 % of the total area.

CALCULATION AND MAPPING ROOT ZONE CAPACITY

The root zone capacity (RZC) is defined as the amount of water in the soil profile which can be utilized by the plants before wilting due to lack of water supply. RZC is calculated by combining the available water content in the soil layers with information on root profiles from different crops. The RZC has not been determined in wetland areas because in these regions capillary rise of groundwater into the root zone is significant.

Construction of soil profiles

Approximately 36,000 soil profiles have been constructed with information on soil texture and organic matter. The profiles are located at the sampling sites from the Danish Soil Classification and they are based on analytical data from this classification and information from the subsoil map. The soil profiles are divided into three sections 0-30 cm, 30-60 cm and 60-120 cm, and within these sections the texture and organic matter content are considered as uniform. The assessment of texture and organic matter content in the sections is as follows:

- The texture and organic matter content in the soil sam

pie 0-20 cm from the Danish soil classification determine
the texture and organic matter content in the
uppermost 30 cm of the profile.
- The texture and organic matter content in the soil sam

pie 35-55 cm from the Danish soil classification determine
the texture and organic matter content in the
section 30-60 cm depth.

If sampling has not been carried out at 35-55 cm depth, the texture from 0-30 cm is used with an organic matter content reduced to the average values for samples taken at 35-55 cm depth.

These values are between 1.9 % and 1.4 % depending
on map colour code (Madsen & Platou, 1983).

- In section 60-120 cm no texture analyses are available from the Danish soil classification. The assessment of texture and organic matter content is therefore based on the texture in section 30-60 cm and information on type of subsoil from the subsoil map.

The organic matter content is in all samples arbitrarily set to 0.5 %. If the texture in section 30-60 cm has less than 10 % clay and the subsoil is sandy, the texture in 60-120 cm will be equal to 30-60 cm with a reduction in organic matter content to 0.5 %. If the subsoil is loamy the following mean MCCS texture is used: 18.6 % clay, 15.1 % silt, 42.8 % fine sand, 23.0 % coarse sand, and 0.5 % organic matter.

- If the texture in section 30-60 cm has more than 10 %

clay and the subsoil is loamy, the texture in 60-120 cm
will be equal to that in 30-60 cm with a reduction in
organic matter content to 0.5 %.

If the subsoil is sandy the following mean MCC3
texture is used: 7.1 % clay, 8.5 % silt, 45.9 % fine sand,
38.0 % coarse sand, and 0.5 % organic matter.

Available water content in the different soil profiles

The available water content for plant production (AWC) is defined as the water content between field capacity (FC) and the permanent wilting point (PWP). For Danish soils this means a water content between pF2.O and pF4.2 (Madsen, 1983). For a calculation of AWC by means of the data from the soil profiles it is necessary to establish a relationship between soil texture and soil water retention. Soil profile investigations have been carried out in Denmark showing this relationship (Hansen, 1976; Madsen & Platou, 1983).

In the actual study, AWC in the soil layers is calculated from the regression equations based on samples from profile investigations in southern Jutland (Madsen, 1986). The equation for pF2.O is based on 285 samples and the equation for pF4.2 on 111 samples. Calciumcarbonate-containing samples, samples with bulk density below 1.0 g/cm³, and samples with organic matter content above 10 % were excluded. The following classes have been used in the regression analyses: clay 0-2 |i, silt 2-20 , fme sand 20-200 \i, coarse sand 200-2000 M, and the mean values for the pF2.O-samples were clay 9.1 %, silt 5.7 %, fine sand 33.7 %, and organic matter 1.4 %.

Root depth

The effective root depth (ERD) is defined as the depth of soil in which AWC is equal to the amount of soil water utilized by the plants until wilting occurs due to lack of water. At this point the soil water potential will be about -15 bar in the upper profile part, because of a high root density there, while in the deeper sections of the profile, where the root density is low, only part of AWC has been utilized by the vegetation, and the soil water potential will be higher than -15 bar. Fig. 5 shows a decreasing utilization of soil water with decreasing root density, and ERD is equal to the soil depth where area a is equal to area b.

Based on investigations of Wiklert (1962), Vetter & Scharafat (1964), Bohm (1978), and Madsen (1979, 1983, and 1985) a mean ERD can be established for grass, spring-sown and winter-sown cereals, Table 2.

RESULTS

Based on the analytical data from the 36,000 soil profiles, the regression equations (1) and (2) and Table 2, the following mean root zone capacities for grass and barley in relation to soil types and landforms can be calculated, Table 3. The root zone capacity increases with increasing MCC-number, except for MCC 3. For a specific soil type, RZC is highest on the Yoldia landform and lowest on outwash plains and dune landforms. The subsoil strongly influenced the RZC for barley and for a specific soil type (MCC-number) the RZC is higher on loamy subsoil than on sandy. This is not the case for grass because the effective root depth for this crop does not exceed 60 cm. A classification of the RZC for barley and grass production in three categories will be as follows: Low RZC is less than 80 mm; medium RZC is between 80 mm and 110 mm for grass and between 80 mm and 140 mm for barley, and high RZC is above 110 mm for grass and 140 mm for barley. Dry soils for plant production are soils with low root zone capacity. These soils have a high irrigation need and in practice many of them are irrigated. Soils with a medium RZC are normally not irrigated.

Fig. 6 shows a root zone capacity map of Vejle county for barley production. Each point is a sampling site from the Danish soil classification with information on texture and organic matter content. In the eastern part of the county loamy Weichsel moraines dominate, while in the western part sandy outwash plains and dune landscapes dominate. Most of the eastern part of the county has a high root zone capacity, frequently above 170 mm but not above 200 mm. Dry soils are rare in that region. In the west, sandy soils cover great areas. The RZC is low, and especially in north-west dry soils dominate. The major part of the dry soils has a root zone capacity between 50 mm and 80 mm, but in between root zone capacities below 50 mm may be found. The greater white areas on Fig. 6 are urban zones or forests.

Fig. 7 shows the location of dry soils for plant production on a nation scale. Most of them are situated in Jutland, where they cover approximately one third of the non-wetland area. Less than 1 % of the islands are dry soils. Table 4 shows the relationship between dry soils and landforms. Dry soils are especially situated in dune landforms and in outwash plains, where they cover more than three quarter of the area. In Saale glaciation landforms dry soils cover roughly half the area, and in the Weichsel glaciation landscape less than 10 %. Dry soils are not found on the Yoldia plateau.

Summary

A nation-wide mapping of root zone capacity for grass and barley production has been carried out. The mapping was based on computerized soil maps and analytical data. Approximately 36,000 soil profiles have been constructed with information on texture and organic matter content. The plant available water content in the different soil layers was calculated by regression equations combining the texture and organic matter content with the water content at pF2.O and pF4.2. An effective root depth for different crops in relation to soil type has been defined, and the root zone capacity was calculated by combining the effective root depth with the available water content in the soil layers. The mapping shows that dry soils defined as soils with a root zone capacity below 80 mm for grass and barley production cover approximately one-third of the non-wetland area in Jutland and less than 1 % on the islands. The dry soils are mainly situated in dune landforms, outwash plains, and Saale glaciation landscapes in mid- and west Jutland.

References

Bornebusch, C. H. & Milthers, K. (1935): Jordbundskort over
Danmark. Danm. Geol. Unders.. 111, 24

Bohm, W. (1978): Untersuchungen zur Wurzelentwicklung bei

Winterweizen. Z. Acker- und Pflanzenbau, 147:264-269.
Hansen, L. (1976): Jordtyper ved statens forsøgsstationer. (En-

glish Summary). Tidsskr. Planteavl, 80:742-758

Jensen, F. S. & Koch, N.E.(\ 987): Landbrugeren og de marginale
jorder. Marginaljorder og Miljøinteresser, samlerapport 111
Miljøministeriet, Skov- og Naturstyrelsen, København.

Jensen, Kr. M. (l 976): Opgivne og tilplantede landbrugsarealer i
Jylland. (English and German summary). Atlas over Danmark,
serie 11, 4. C. A. Reitzel, København.

Madsen, H. B. (1979): Jordbundskartering og bonitering, Ph.D
Thesis (English summary). Folia Geographica Danica X,5. 202
P

Madsen, H. B. (1983): Himmerlands jordbundsforhold, Doctorial
Thesis (English summary). Folia Geographica Danica XVI
329 p.

Madsen, H. B. (1985): Distribution of spring barley roots in
Danish soils of different texture and under different climatic
conditions. Plant and Soil 88:31-

Madsen, H. B. (l986): The construction of root zone capacity
maps basedon computerized soil maps and pedological data.
Pedologie XXXVI: 133-154.

Madsen, H. B. & Jensen N. H. (1985): The establishment of
pedological soil databases in Denmark. Geografisk Tidsskrift
85:1-

Madsen, H. B. & Platou, S. W. (1983): Land Use Planning in
Denmark. Nordic Hydrology: 267-276.

Mathiesen, F. D. (1980): Soil Classification in Denmark. Its
results and applicability. EEC-report on land resource evaluation.
EUR 6875.

Mathiesen, F. D. & Nørr, A. H. (1976): Den danske Jordklassificering, Teknisk Redegørelse. Landbrugsministeriet Arealdatakontoret, Vejle. Platou, S. W. (1984): The ADK database systems (PDS and LADS) used in Denmark. In Soil Information Systems Technology (eds. P. A. Burrough & S. W. Bie). Proceedings of the sixth meeting of the ISSS Working group of Soil Information System, Bolkesjø, Norway. Pudoc Wageningen.

Schou, A. (1949): Atlas of Denmark. Kgl. Danske Geogr. Selskab,
Copenhagen.

Smed, P. (1979): Landskabskort over Danmark. Geografforlaget,
Brenderup.

Vetter, H. & Scharafat, S. (1964): Die Wurzelverbreitung landwirtschaftlicher Kulturplanzen im Unterboden. Z. Acker- und Pflanzenbau, 120:75-298. Wiklert, P. (1962): Om sambandet mellam markstruktur, rotutveckling och upptorkningsförlopp (English summary). Grundförbättring, 14:221-259.