Tải bản đầy đủ - 0 (trang)
Chapter 4. Hydrological Factors for Phosphorus Transfer From Agricultural Soils

Chapter 4. Hydrological Factors for Phosphorus Transfer From Agricultural Soils

Tải bản đầy đủ - 0trang

154



P. M. HAYGARTH ET AL.

ment, where problems occur. Our aim is to provide a simplified basis for classifying the otherwise complex hydrochemical regimes which result in P transfer.

© 2000 Academic Press.



I. INTRODUCTION

The consequences of land use on environmental quality have an increasingly

high profile. From an agricultural standpoint, one area of critical current interest

is the transfer of pollutants from soil to water because soils play a pivotal role in

the protection of groundwaters and surface waters (National Rivers Authority,

1992). There is a need to identify and describe sources, media, and pathways of

pollutant transfer in order to aid the developing interactions of soil science with

hydrology (Boorman et al., 1995) and/or land use specialists with aquatic biologists. Such multidisciplinary links have the potential to create ambiguous, imprecise, and contradictory terminology or to omit key information. Definitions are

particularly important when describing environmental phosphorus (P) transfer

from agricultural soils to inland water bodies. Most studies on P transfer are from

an agronomic standpoint in which the role of hydrology is not fully considered

(Haygarth and Jarvis, 1999). We therefore deliberately take this opportunity to describe the hydrological factors in P transfer because, to date, there has been a tendency for such factors to be neglected.

In many of the world’s agricultural soils, P accumulates in intensively managed agricultural soils because of farm imports of fertilizer and livestock feeds

(Brouwer et al., 1995; Haygarth et al., 1998b). Consequently, agricultural soils

are now considered to be the main diffuse source of P reaching freshwaters (Foy

and Withers, 1995), in which concentrations as low as 35 –100 ␮g total P literϪ1

may contribute to eutrophication (Organization for Economic Cooperation and

Development, 1982). Previous research on phosphorus transfer (PT) has emphasized measurable “soil factors” such as defining threshold P concentrations

using soil extractants (Heckrath et al., 1995; Sharpley et al., 1996), with relatively few studies focusing on “hydrological factors” (Heathwaite, 1995). According to Heathwaite, soil characteristics and factors define the initial chemical form of P export, but the hydrological conditions determine whether or not

mobilization occurs and along which pathway. The scarcity of studies on hydrological interactions with PT is perhaps understandable because, unlike soil

properties, hydrological factors are less easy to measure, classify, and interpret

than many other soil properties. Hydrological conditions are also temporally and

spatially dynamic and have the added difficulties of variations with changes in

scale.



HYDROLOGICAL FACTORS FOR PHOSPHORUS TRANSFER



155



Conventionally, general-purpose soil classifications have been defined by pedological properties, although broad hydrological status might be implied from class

names, as in Avery’s (1980) “stagnogley.” Although pedological classifications

place high significance on expression or gleying (indicating duration of waterlogging), it required the hydrology of soil types (HOST) classification of Boorman et

al. (1995) to provide a detailed description of the hydrologically driven system in

England and Wales. Inferences on overall hydrology, pathways, seasonal behavior patterns, and responses to rainfall can be derived from HOST. Until the Host

classification was developed, only the very simple winter rain acceptance potential classification (Farquharson et al., 1978) and U.S. Department of Agriculture

(USDA, 1972) hydrologic soil grouping had, in a consistent way, addressed the

hydrology within the soil.

Studies of the fundamental role played by water flow characteristics on PT are

invaluable because hydrology provides both the energy and the carrier for the

transfer regime (Haygarth and Jarvis, 1999). Moreover, soils are hydrologically

diverse (Boorman et al., 1995), with very different pedological and hydrological

properties as well as varied responses to rainfall and land-use practices. Nevertheless, contrasting soils are often found in close proximity on hillslopes and even

within individual fields. Such differences are illustrated by the soils in Table I

which, along with the hydrologically similar soils they represent, are important in

various regions of the United Kingdom. The contrast is shown by the impermeable, clayey Hallsworth series soils (Avery, 1980), which are waterlogged from

October to May, whereas the permeable, loamy Denbigh series soils are waterlogged only in very wet weather. Thus, each soil series has different implications

for PT.

The aim of this review is to help to redress the current shortfalls in information

by providing a considered review of the hydrological issues and defining the role

of hydrology in context with other aspects of PT. The treatment of hydrological

factors in this review is provided in two sections dealing with temporal and spatial variables.



II. TEMPORAL VARIABLES

A. EFFECTIVE RAINFALL

Precipitation, with its temporal variation, provides the energy source and the

physical carrier mechanism within the PT process (Haygarth and Jarvis, 1999).

Critical to PT is the relationship between rainfall input and runoff generation. In

catchment/watersheds in which the climate is characterized by low-intensity rain-



Table I

Hydrological Characteristics Describing the Contrast between the Denbigh and Hallsworth Soil Series

Soil classification

Soil Survey and Land Research Centre (Avery, 1980)

USDA (1972)

Food and Agriculture Organization

Wetness (Hodgson, 1997)

Class

Duration of waterlogging (days)

Ͻ40-cm depth

Ͻ70-cm depth

Saturated hydraulic conductivity (m/day)

Porosity

Retained water % (moisture content after drainage, i.e., at FC)

Air capacity % (drainable pores)

Workability/trafficability (Findlay et al., 1984)

Good machinery workdays relative to field capacity season

HOST (Boorman et al., 1995)

General description



Hydrological pathways



HOST class

SPR% (standard percentage runoff; from catchment scale

interpretation)

Base flow index (BFI% 1 ϭ total BF 0 ϭ no BF; from

catchment interpretation)



Denbigh



Hallsworth



Typical brown earths

Dystrochrepts

Dystric cambisols



Pelostagnogley soils

Typic haplaquepts

Dystric gleysols



I



IV



0 days

Ͻ30 days

Ͼ1 (topsoil), Ͼ1 (subsoil)



Ͻ180 days

Ͼ180 days

0.8 (topsoil), 0.002 (subsoil)



42 (topsoil), 32 (subsoil)

12 (topsoil), 21 (subsoil)



47 (topsoil), 49 (subsoil)

12 (topsoil), 0.1 (subsoil)



ϩ30



Ϫ55



1:No impermeable or gleyed layer

within 1 m

2: No significant aquifer or GW

3: Over impermeable or hard substrate

Vertical unsaturated flow; bypass flow

in the substrate; some surface runoff



1: Gleyed layer within 40 cm



17

32



2: Over slowly permeable substrate

3: No significant aquifer

Surface runoff likely; prolonged seasonal

saturated flow; short seasonal bypass

flow to the substrate

24

51



0.6



0.31



HYDROLOGICAL FACTORS FOR PHOSPHORUS TRANSFER



157



fall (Ͻ15 mm hϪ1), the current soil moisture state is more important in controlling the occurrence and magnitude of runoff than rainfall magnitude (Istok and

Boersma, 1986). This has implications for PT because most export occurs when

the land is at or close to field capacity (FC). For example, in the Slapton catchment, Devon, England, 80% of PT occurred in winter (Heathwaite et al., 1989).

Snowmelt has been associated with high rates of PT from soil (Hawkins and

Scholefield, 1996; Timmons et al., 1977).

The concept of “field capacity” is useful in this context. Field capacity occurs

when a soil is thoroughly wetted and drainage starts. For example (see Table I), a

freely draining soil such as the Denbigh series would “drain” and reach FC in approximately 2 days from thorough wetting, whereas this takes weeks (Hodgson,

1997) in a poorly drained soil such as the Hallsworth series. The respective hydrological properties of these two soil types are summarized in Table I. Meteorological field capacity (MFC) occurs when received rainfall is greater than evapotranspiration, provided no soil moisture deficit exists. MFC is readily presented in

cartographic form (Jones and Thomasson, 1985). Excess precipitation, which occurs when soils are at FC, can be called hydrologically significant precipitation.

This effective rainfall is an important driver for the PT process.



B. LEVELS OF HYDROLOGICAL ACTIVITY

Temporal variations in rainfall intensity, duration, and intervals between

storms (return period) affect the magnitude of discharge along various hydrological pathways (Burgoa et al., 1993; DeWalle and Pionke, 1994; Evans, 1978;

Sharpley, 1980a,b; Thornes, 1979). Although flow rate is a continuum from low

to high discharge, different hydrological pathways may be triggered at different

rates of discharge or rainfall input, with different consequences for PT. However, division into base-flow and storm-flow conditions is a helpful starting point

when considering discharges. Storm flows occur infrequently and result in overland and macropore flows, whereas base flows occur more frequently but may

only transfer water along predominately subsurface pathways. Pionke et al.

(1996) found that storm flow was important for P discharge from a 7.4-km2 agricultural watershed. Dils and Heathwaite (1996) found that individual storm

events had different capacities to transport P; moreover, the forms of P varied between events. They suggested that the most important controls on P fractionation

were antecedent soil moisture (determining the likelihood of surface runoff ) and

the interval between storm events (determining the incidence of “old” and “new”

soil water and hence the amount of time for interaction between mobilized P and

soil water).

Base flow comprises non-storm-flow periods in which groundwater discharge,

including springs and near-stream seepage, may form the main component of flow



158



P. M. HAYGARTH ET AL.



(Pionke and DeWalle, 1994). For base flow, subsurface hydrological pathways

such as throughflow are most important for PT. Despite the obvious importance of

storm events, the significance of subsurface pathways in transporting P during

base-flow or low-flow conditions should not be discounted and needs further research in terms of PT.

A limitation of the storm-flow/base-flow concept is that it pertains solely to discharge per se, whereas a more generalized temporal hydrological classification,

which accounts for precipitation (and thus rainfall erosivity, for example), may be

more useful. Haygarth and Jarvis (1997), Haygarth et al. (1998a), and Fraser et al.

(1998) identified two populations of data when studying PT, implicating an interaction of rainfall with flow. We therefore suggest that an effective means of classification must recognize at least two levels of hydrological activity. Level 1 activity occurs during light or little rainfall for a high proportion of time (and will

incorporate base flows). In contrast, level 2 activities will be of low frequency but

high intensity, operating with greater energy than level 1 activity, and have a high

propensity for PT during a short period and resulting in storm flows. When level

2 occurs it will produce high-intensity rainfall having a greater erosivity and will

therefore result in larger particulate PTs than those of level 1 rain. In reality, a twotier classification may be too simplistic and a sliding scale of increasing activity

may be more appropriate: Nevertheless, this system provides useful initial conceptual classification.



C. TIMESCALES

Shrinkage of soils such as the Hallsworth series (Table I), with coarsely structured, often clayey horizons (HOST classes 18 –25 of Boorman et al., 1995), during summer soil moisture deficits opens vertical fissures. These permit vertical bypass flow particularly during autumnal rewetting (level 2; Fig. 1a). When rewetted,

the horizons revert to an impermeable state, dominated by saturated lateral flow

(level 1). The temporal extent of either state depends on the strength of the soil

moisture deficit and the duration of the field capacity period.

In Table II we present a nonexhaustive list of hydrological pathways which

we have attempted to classify in relation to timescales (more detailed discussion

of the relevance of this table is given in the following section, which defines scale).

At the slope/field scale, the time taken for overland flow to travel 100 m may be

of the order of minutes during intense rainfall (Horton, 1945), whereas water in

subsurface pathways to underlying aquifers may take months to years (thus giving rise to the terms new and old water; Bohlke and Dener, 1995). Again, many of

these variations can be defined by soil class, for example, as expressed in the

HOST classification of Boorman et al. (1995) and illustrated by data in Table I for

the Denbigh and Hallsworth series.



Figure 1 Soil profile scale pathways. The situation common (a) in late summer/autumn, when soils are dry and prone to bypass flow, and (b) in winter/spring,

when soils are saturated and prone to matrix flow.



Table II

Terminology Commonly Associated with Hydrochemical Transfer Pathways, Nominally Classified by Discipline, Time, and Spatial Scale

Term (sorted

alphabetically)



Scale



Generic



160



Arterial drainage



Catchment



Agronomic



Base flow



Slope/field



Hydrological



Bypass flow



Soil



Hydrological



Darcian flow



Soil



Hydrological



Ditching



Slope/field



Agronomic



Interflow

Land drainage



Slope/field

Subcatchment



Hydrological

Agricultural



Leaching



Soil



Chemical



Leakage



Slope/field



Macropore flow



Soil



Hydrological and

chemical

Hydrological



Matrix flow



Soil



Hydrological



Overland flow



Slope/field



Geomorphological/

pedological



Nominal

timescale



Definition

Major artificial drainage channel used to remove water after its

discharge from field drains

Nominally the transfer of water underground in “background”

flow conditions, not pathway specific; also used to describe

a low magnitude of flow

Implies a type of soil water movement—in the case of vertical

movement along larger subsoil pathways, e.g., wormholes

and fissures, often occurring in unsaturated conditions

Not a pathway but describes discharge through the soil as related

to the hydraulic gradient and hydraulic conductivity

Artificial (human-made) first-order drainage streams, often used

on agricultural land to compliment arterial drainage

Lateral flows below the soil surface

Water and solute (ϩ entrained solids) export to catchment

resulting from land drainage practices: anthropogenic

Eluviation of chemicals vertically through the soil profile and

vadose zone; despite misconceptions, this is a mechanism

and not a pathway

General nonspecific term describing water and chemical

movement

As bypass flow; macropores are large enough (Ͼ60 ␮m) to

allow gravitational drainage

Implies a type of soil water movement—in this case uniform

vertical movement downwards, common in very porous

media such as sandy textures; only occurs under saturated

conditions

Movement of water exclusively over the soil surface, down

slope, during heavy rain



Hours/days



Hours/weeks/months

Minutes/hours





Minutes/hours

Minutes/hours

Minutes/hours

Variable



Not applicable

Minutes/hours

Days



Minutes/hours



161



Percolating water

Pipe flow

Piston flow

Preferential flow

Return flow

River

Roadway



Soil

Slope/field

Soil

Soil

Slope/field

Subcatchment

Subcatchment



Hydrological

Geomorphological

Hydrological

Hydrological

Hydrological

Hydrological

Engineering



Runoff



Slope/field



Hydrological



Saturated (soil) flow

Seepage



Soil

Slope/field



Hydrological

Hydrological



Soil solution



Soil



Chemical



Stream



Subcatchment



Subsurface flow

Surface runoff

Throughflow



Slope/field

Slope/field

Soil and

Slope/field

Slope/field



Hydrological and

geomorphological

Geomorphological

Hydrological

Hydrological



Unsaturated flow

Vertical saturated

flow

Vertical unsaturated

flow



Hydrological



General nonspecific term describing water movement

Lateral subsurface preferential flow

As matrix flow

As bypass flow

Where a subsurface flow pathway emerges at the soil surface

Large-order drainage network

Human road or path which can assist water transfer from slope/

field to catchment; little studied

General hydrological term describing the lateral movement of

water off land above and below ground, causing a short-term

increase in flow at the catchment outlet; can refer to pathway

when qualified (e.g., surface runoff), but also has been used

to describe processes and water samples

As piston flow, but lateral and not vertical

General nonspecific term describing water movement; implies

emergence at or near the ground surface

Nonspecific term describing water sampled from the soil

environment by whatever means; not a pathway

Small-order drainage network



Not applicable

Minutes/hours

Variable

Minutes/hours

Minutes/days

Hours/days

Hours/days



Lateral flows below the soil surface

As overland flow

As percolating water



Minutes/hours

Minutes/hours

Not applicable

Minutes/hours

Days

Minutes/hours



Soil



Hydrological



As preferential flow, but occurring laterally over capped,

compacted, or slowly permeable horizons

As piston flow



Soil



Hydrological



As bypass flow



Minutes/hours



Days

Not applicable

Not applicable

Hours/days



162



P. M. HAYGARTH ET AL.



III. SPATIAL VARIABLES

A. SCALE

Effects of space and scale are important when attempting to understand PT

along soil hydrological pathways (Beven et al., 1993; Konikow, 1991). Kirkby

(1988) demonstrated the importance of scale (drainage basin area) on hillslope

flow processes (Fig. 2). The concepts incorporated in Fig. 2 may be extended to

an examination of PT where lag times to peak concentration (Fig. 2a) are impor-



Figure 2 (a) Lag times and (b) peak runoff rates integrating spatial and temporal factors. NB Hortonian ϭ infiltration excess overland flow. (Redrawn from “Hillslope Hydrology,” M. J. Kirkby. Copyright John Wiley & Sons Limited. Reproduced with permission.)



HYDROLOGICAL FACTORS FOR PHOSPHORUS TRANSFER



163



tant in determining exchange and uptake reactions with soil solution and consequently the form of P transported. In contrast, peak runoff rates (Fig. 2b) also aid

our understanding of the potential P load reaching the drainage network. Thus, for

cultivated land, the higher peak runoff rates associated with Hortonian (infiltration-excess) overland flow (Fig. 2b) will have greater capacity to detach and transport soil particles and sorb P. Although this pathway is restricted in catchment/watersheds in temperate climes such as that in the United Kingdom, where the soil

infiltration capacity is rarely exceeded (Kirkby, 1988), poor land management

(e.g., overgrazing) will exacerbate its incidence (Heathwaite, 1997).

Because soil hydrological and chemical processes operate at different scales, it

is essential that their spatial dimensions be identified. The challenge is to identify

mechanisms that connect soil profile and slope/field scale process studies with

larger scale catchment/watershed effects. Recent attempts to incorporate chemical–physical links initiating PT using a combination of runoff and erosion modeling (Gburek et al., 1996) have advanced knowledge of pathways of P delivery and

their interaction with soil profile characteristics. Similarly, knowledge of hydrological pathways of water movement from land to stream has developed with increasingly detailed means of field monitoring. Thus, simple models of pipe-flow

channeling of new water via the soil matrix to the stream have been replaced by

new concepts incorporating new/old water and bypass flow (Bohlke and Dener,

1995).

For the purpose of clarifying approaches to understanding the spatial controls

on PT, we have nominally selected two scales smaller than that of catchment/watershed. The size boundaries selected are arbitrary but provide a convenient means

for subdivision of landscape units. The smallest is the soil profile scale (centimeters to meters). The second is the slope/field scale; this has also been called the

hillslope or the subcatchment/watershed and will range from meters to hectares

and is generally restricted to catchment/watershed headwaters and/or first-order

streams (Haygarth and Jarvis, 1997; Heathwaite and Johnes, 1996). This scale is

likely to be relevant to individual soil mapping units (scales of 1:25,000 and

greater). For PT from agricultural land, this scale is perhaps the most critical because it is possible to integrate detailed soil process studies with characterization

of the patterns of PT at the slope/field scale (e.g., P fractionation in different hydrological pathways; Dils and Heathwaite, 1996; Haygarth et al., 1998a). The

largest scale is the catchment/watershed; although we will not focus on catchment/

watershed pathways, it is necessary to consider some definitions. Catchments (English term) or watersheds (American term) are variable in size and there are many

examples of catchment/watershed studies and classifications (Burt et al., 1996;

Gburek et al., 1996; Heathwaite and Johnes, 1996; Johnes et al., 1996; Kronvang,

1990; Molden and Cerney, 1994). Catchment/watersheds are appropriate for interpretation and mapping of “reconaissance” maps (1: M in the United Kingdom,

as in appendix D of Boorman et al., 1995). Our scheme for subdividing the spa-



164



P. M. HAYGARTH ET AL.



tial dimensions of PT is similar to that proposed by Kirkby (1988) in order to examine the time lag (in hours from precipitation input to stream output within a

catchment/watershed; see Fig. 2). Kirkby argues that “catchments may be thought

of as a sequence of moisture stores, some in series, some in parallel.” Temporal

rates of water movement (introduced in Section II) clearly interact with the spatial scales and can give rise to the classification of (i) surface detention (0.1–1 h),

(ii) infiltration (1–20 h), (iii) unsaturated vertical percolation (1–50 h), (iv) saturated downslope flow (1–12 h), (v) channel flow [depends on catchment/watershed area: 0.5 h (1 km2), 7 h (100 km2), 100 h (104 km2)]. Here, scales i–iv equate

with our soil profile scale, scale iv and part of scale v equate with our slope/field

scale, and v refers to the wider catchment/watershed. The store with the longest

residence time exerts the greatest control on water movement. This has implications for P form and the magnitude of P export because it indicates the amount of

time available for equilibration between rainfall and soil water and hence the potential for P mobilization and transfer.



B. PATHWAYS

Generalized overviews and definitions of hydrological pathways have been provided by many authors (Anderson and Burt, 1990; Boorman et al., 1995; Cox et

al., 1997; Kirkby, 1978; Mangold and Tsang, 1991; McGechan and Wu, 1996;

Miyazaki et al., 1993; Youngs and Leeds-Harrison, 1990). When precipitation

reaches the soil, it is partitioned between overland flow and subsurface flow (Kirkby, 1988), each having varying potentials to entrain and retain P. Terms used to describe these two pathways (and their various forms) are sometimes confused in the

scientific literature and problems arise because

1. Hydrological pathway terms can become confused with process terms. The

terms “runoff” and “leaching” are good examples. Runoff has been used in the

context of a (rather vaguely defined) pathway (Boorman et al., 1995; Haygarth and

Jarvis, 1996, 1997), a water medium (Harms et al., 1974; Loehr, 1974), or, in occasional circumstances, a process (Zobisch et al., 1994). Leaching has also been

used in an ambiguous way. It does not describe a pathway, although it has sometimes been used in this context to characterize an amalgam of all pathways of water drainage through soil (Bromfield and Jones, 1972; Heckrath et al., 1995; Jordan and Smith, 1985). Leaching is a process term and describes the elluviation of

solutes, such as P, down through soil and is common in porous soils (Wagenet,

1990; Weaver et al., 1988a,b).

2. Pathways tend to be classified according to research background and discipline and may be defined in terms of geomorphological, hydrological, pedological, chemical, or land-use criteria.



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

Chapter 4. Hydrological Factors for Phosphorus Transfer From Agricultural Soils

Tải bản đầy đủ ngay(0 tr)

×