Publications

Health of our Soils | Index

Chapter 4 - Benchmark Sites for Monitoring Agricultural Soil Quality

C. Wang, L.J. Gregorich, H.W. Rees, B.D. Walker, D.A. Holmstrom, E.A. Kenney, D.J. King, L.M. Kozak, W. Michalyna, M.C. Nolin, K.T. Webb, and E.F. Woodrow

Highlights

• A national benchmark program has been put in place to monitor changes in the health of agricultural soils, especially changes resulting from land use and management practices.
• Baseline data sets include farm history; soil and landform descriptions; and measurements of various chemical, physical, and biological properties of the soil. Several climatic factors are measured daily at eight locations.
• Some soil properties are measured annually. Others are measured every 5 or 10 years, depending on their sensitivity (how quickly they change). Soil properties are classified as sensitive, moderately sensitive, and nonsensitive.
• Baseline data sets for all benchmark sites will be completed in 1995; the first set of resampling data has been collected for 6 out of 23 sites.
• Early results of the program allow some conclusions about effects of farming practices on soil health.

Introduction

Discussions of sustainable agriculture in Canada have raised questions about the health of our agricultural soils and the effects of farming practices on soil health, including:

• How do changes in soil properties affect crop yields?
• Do certain crops and crop rotations degrade soil more than others?
• Does the method of tillage have an effect on soil health?
• Is there enough organic matter in our soils? What can be done to build levels up?
• How serious is the problem of soil erosion? Which farming practices increase or reduce the risk of erosion?
• How widespread is the breakdown of soil structure? How can soil compaction by farm vehicles be reduced?
• When and how should agrochemicals be used? What happens when they accumulate in the soil? (National soil-quality monitoring program)

However, the scientific information needed to answer these and other questions in relation to a broad range of agricultural conditions across Canada was generally unavailable. Although studies on various features of soil health were being carried out at several sites across Canada, most of these studies were narrow in scope, focusing on selected soil properties, land uses, and management systems.

What was needed was a system to monitor the quality of agricultural soils all across the country. This system would look at the health of soils under different land uses and farming practices. Over time it would show how soil health changes as a result of climate, landform, and agricultural use.

In January 1988, participants at a national workshop on soil-quality monitoring agreed that we had only a hazy picture of the health of Canada's agricultural soils. It was decided that a national soil-quality monitoring system should be put into place and was agreed to establish agricultural benchmark sites across Canada-sites at which certain soil and environmental characteristics would be measured in a standard way at regular intervals, so that changes could be observed over time.

Seven benchmark sites were set up and sampled in eastern Canada in 1989. The National Soil Conservation Program began supporting the project in 1990, and 16 more benchmark sites were selected and sampled by 1993, for a total of 23 sites across Canada (Fig. 4-1). The collection of baseline data sets (the first set of measurements made at each site) will be completed in 1995. Soil quality will be monitored for at least 10 years after baseline data is collected or until trends in changing soil health become clear.

There is an urgent need to develop the institutional capability for continuously monitoring and assessing the status or quality of the world's resources.

Parr et al. 1992

Objectives of the benchmark system

The system for benchmark monitoring will allow various soil properties and processes to be measured at specific times and places over the long term. In this way, it will generate periodic "snapshots" of soil health and allow researchers to determine changes in soil health over time and to relate these changes to land use and management practices, as well as other factors.

The main objectives of the benchmark system are to provide:

• baseline and resampling data sets for assessing changes in soil health and productivity (soil's ability to produce crops, measured as crop yields) of typical farm production systems
• a way to test and validate simulation models that predict soil degradation and productivity
• a way to evaluate whether farming systems in the major agricultural regions of Canada are sustainable
• a national network of sites that can be used by government and non-government groups to conduct cooperative research.

Benchmark sites

Benchmark sites were selected on the basis of seven factors that were identified by scientists and agricultural workers. The goal was to represent the main landform in Canada's most important agricultural regions. Benchmark sites were to be made up only of cultivated land and were to represent:

• a major soil zone, agricultural region, or both
• a typical landform or a broad textural grouping of soils, or both
• an important farm production system within a region.

They would also:

• fit well with provincial agricultural concerns
• show signs of soil degradation or the potential for soil degradation
• cover about 5 to 10 hectares (a small watershed in some cases).

Emphasis was placed on the first four points, so that monitoring results could be used to evaluate as many similar landforms and farming systems as possible over as broad an area as possible within each agricultural region. (Choosing a benchmark site)

Baseline data sets

To observe changes in soil quality over time and area, one must begin by establishing a set of baseline data. This set of observations, made at the start of the monitoring period, serves as the reference point against which to compare all future observations.

Farm history database

A farm history was recorded for each benchmark site, based on an interview with the farmer. Three types of information were collected:

1. site identification (including legal and ecological descriptions);
2. site history (including land acquisition; first cultivation; land management in the early years; major changes in farming practices; crop rotation; tillage system; crop yields and quality; use of commercial fertilizers, organic fertilizers and soil conditioners, and chemical pesticides or herbicides; drainage improvement; and degradation problems); and
3. current cropping and tillage practices (including crop rotation system; tillage, crop management, and harvesting methods; and an inventory of farm machinery).

Soil and relief maps

The following were drawn for each site:

• a map of the types of soil (Fig. 4-2) found over the entire benchmark site, at a scale of 1:2000 or larger, using observations from at least 40 inspection points
s
• a map of the surface relief (Fig. 4-3A) of the benchmark site, constructed from elevations (relative height of the ground surface) measured at about 100 selected points.

Information provided by the relief map can be used to display landforms, such as ridges and gullies, and drainage direction. It provides the background for overlaying other soil and landform features. It also allows strategic sampling points to be located for future resampling.

Soil sampling

Three types of samples are taken as follows:

• samples of all the major layers in the two most representative soils found on the benchmark site
• 60-100 loose soil-surface samples, taken at selected points throughout the benchmark site (to establish baseline data for the site)
• 60-100 loose soil-surface samples, taken at selected points throughout the benchmark site to assess soil redistribution, using analytical methods involving cesium.

A portion of each loose sample is stored in a soil archive (collection) so that more testing is possible in the future.

Soil properties

A number of chemical, physical, biological, and mineralogical properties of soil were identified at the 1988 soil-quality workshop as key elements of a baseline data set. These properties were classified into the following categories (Table 4-1):

• sensitive properties, which could change significantly in less than 10 years
• moderately sensitive properties, which are likely to change over decades
• nonsensitive properties, which are not expected to change significantly in 100 years.

The nonsensitive properties, although not expected to change much over the duration of this study, are important properties for assessing the overall soil health of the benchmark sites.

Data sets for some physical and biological properties of soil are collected in the field (Table 4-1). Measuring the saturated and unsaturated hydraulic conductivity outlines the movement of water in the crop-root zone of soil. It also identifies changes in soil structure over the long term. Penetrometer readings provide information about soil strength (soil strength influences plant-root growth, which in turn affects the availability of soil moisture to plants).

Biopores, which are important to the movement of water in water-saturated soils and to soil aeration, include root channels and earthworm holes. Earthworms have an important effect on soil structure; a large population of earthworms produces a better structure by aerating the soil and increases the availability of organic nutrients. Crop yield (quality and quantity of the crop) is affected by many soil properties and processes and is the most visible sign of the health of agricultural soils. (Sensitive soil properties)

Progress of the study

Baseline data sets

Much baseline information has already been collected at the benchmark sites. Most of the data have been organized and entered into a national benchmark site database (a collection of data). Besides providing valuable information about trends for particular soil properties, the database allows researchers to observe how a particular feature is affected by other characteristics of the soil.

Example 1

Site 18 in Quebec appears to the casual observer to be a flat field. But, by mapping the contours of the field, we found that the field has a slight slope, and a small ridge extends halfway across it at the mid point (Fig. 4-3A). This ridge, although a small relief feature, clearly affects soil characteristics at this site.

On measuring soil moisture (Fig. 4-3B), we found that the moisture was greatest directly behind, or upslope, from the ridge. The water runoff pattern is also affected; water skirts the ridge and runs to the opposite side of the field downslope, leaving the soil directly below the ridge with a much lower moisture content. Levels of soil organic matter (Fig. 4-3C) were higher behind the ridge, probably because better soil moisture conditions in that location promote better crop growth and higher returns of organic matter to the soil.

A basic principle in the manufacture of products is that quality cannot be sampled [sic] into a product.... The analogy to soil quality is that monitoring soil quality cannot change it. Therefore, sustainable land management requires a deliberate effort to design land management systems that do not degrade the soil.

F.J. Pierce and W.E. Larson
Developing Criteria to Evaluate Sustainable Land Management

Example 2

Site 5 in Alberta (Fig. 4-1) has a hilly surface. Adjacent to the national benchmark site is another agricultural research site, located on native (uncultivated) land. These paired sites provide the opportunity to look directly at the effects of agriculture on native soil. Sampling points are located along a hillslope on both the cultivated (since 1912) and native sites. Soil properties vary widely according to slope position and land use. On the hilltops (crests and upper slopes) of both cultivated and native sites, the topsoil is high in pH (Fig. 4-4A) and low in organic matter (Fig. 4-4B). On lower slopes and depressions, the topsoil is low in pH and high in organic matter. Because of soil erosion and thin topsoil, cultivation has brought subsoil material near the surface. Carbonates (which have a high pH, or are alkaline) from the subsoils have been mixed with the plow layer, producing a higher pH in hilltop positions. The lack of grass cover on the cultivated slope allows greater runoff of precipitation. This runoff in turn results in more leaching by water on the lower slope, producing a lower pH in this position on the cultivated site than on the native site.

The organic matter content is much lower on the cultivated site than on the native site (about 25% on the hilltop and 50% on the lower slope). Cultivation dilutes and reduces the amount of organic matter that is found naturally in prairie soils; the reduction is greatest at hilltop positions. On hilltops (crests and upperslopes), the topsoil is low in organic matter and nitrogen. On lower slopes and in depressions, levels of organic matter and nitrogen are much higher. (Soil organic matter is discussed more fully in Chapter 5.)

Monitoring and preliminary results

Soils at each benchmark site are resampled according to the monitoring schedule (see footnotes to Table 4-1) by taking 60-100 loose surface-soil samples at selected points throughout the site. For hilly sites, measurements of soil properties are grouped by their position in the landscape (all the measurements taken at upper-slope positions are combined, as are measurements taken at mid-slope and lower-slope positions). From this grouping we can observe differences in how soil quality changes at different slope positions. Six benchmark sites were resampled by 1993. Some trends in soil quality were already evident at that time. (A benchmark farm family)

Example

The New Brunswick benchmark sites (Sites 20 and 22, Fig. 4-1) have the same soil type and similar slopes. They are under similar tillage systems, and the crop rotation at both sites is 2 years of potatoes followed by 1 year of grain. (This rotation is not typical in New Brunswick; most rotations are less intensive, such as 1 year of potatoes followed by 1 or more years of grain.) Terraces and grassed waterways are used on Site 22 to prevent soil erosion, whereas Site 20 is cultivated up- and down-slope.

A comparison of the baseline and re-sampling data for these sites (Fig. 4-5) shows that over 3 years there has been little change in the level of soil organic matter at Site 22, but there has been a 5% loss on Site 20, 10 to 20 times more than on Site 22. Nearly all erosion on both sites took place during potato production. This result shows the positive effects of erosion controls in reducing the loss of soil organic matter.

On the Prince Edward Island benchmark site (Site 21, Fig. 4-1), potatoes are grown for 1 year, followed by 1 year of grain and 2 years of forage, which is plowed under. The initial level of soil organic matter at this site was lower than at either of the New Brunswick sites, because a previous cropping system had degraded the soil. But, between 1989 and 1993, soil organic matter at this site actually increased slightly as a result of plowing under the forage crop 2 years in a row (this practice puts organic matter and nutrients back into the soil).

A degraded soil may require many years to be restored. On the Prince Edward Island site, the 4-year crop rotation resulted in only a small improvement in soil organic carbon content. This observation shows that it takes less time and effort to maintain healthy soils and prevent soil degradation than to restore a soil that has been degraded.

The future

The national benchmark system holds considerable promise for monitoring the health of selected agricultural soils. It not only takes a systematic look at how individual soil properties change over time but also shows how these properties are related to each other and how current farming practices affect soil health. Early findings of the program show that the sensitive soil properties selected for measurement every 5 years are indeed measurable and together provide a good indication of changes in soil health.

At a time when financial resources are limited and research needs are pressing, it makes sense for partners in agricultural research to work together. The benchmark system enables researchers from many agencies to use the database, set up complementary programs, and apply findings to a wider range of sites.

But the care of the earth is our most ancient and most worthy and, after all, our most pleasing responsibility. To cherish what remains of it, and to foster its renewal, is our only legitimate hope.

Wendell Berry
The Unsettling of America: Culture and Agriculture