Tải bản đầy đủ - 0 (trang)
IV. Cycling and Flows at the Field Level

IV. Cycling and Flows at the Field Level

Tải bản đầy đủ - 0trang



the same field on which the plant was grown and complete a true s o i l j p l a n t j

soil or soil+plant+animal+soil cycle at the field scale. Nutrients may also be

lost from the cycle as they flow away from the field in both managed and unmanaged flows and do not return. On the other hand, nutrients that did not originate in

the field can flow onto it in the form of manure from animals fed partially or completely with feeds from other fields or farms. Also, crop residues applied to a field

can originate in other fields or farms. These flows, although not nutrient cycles at

the field level, may represent cycling at the farm or regional scales.

There are numerous factors that control the efficiency with which nutrients are

utilized by crops and/or lost from soils. These include the characteristics of the nutrient source, the quantity of nutrients added, the method and timing of the application, the uptake needs and abilities of the crop, soil structure, and the climate.



1. Harvest Removal

Crop harvest normally represents the only intentional loss of nutrients from

fields. The amounts of nutrients removed in the harvested portion of crops are influenced by the characteristics of the plant species and the portion of the plant removed. Although there are differences in biomass production and elemental composition among crop species, some of the most striking differences in nutrient

removal are caused by differences in the amounts of plant material removed. With

a crop such as broccoli, many nutrients are taken up by the crop, but only approximately 20% of the nutrients in the aboveground biomass is removed with the

heads, with the remaining 80% in the unharvested stalk and leaves. Therefore, relatively small amounts of nutrients leave the fields with the harvested broccoli

heads (Table 11). Potatoes offer a contrast to broccoli because a large portion of the

total plant biomass and nutrients are in the harvested tubers. Nutrient removals of

N, P, and K in potato harvests are approximately four to nine times greater than for

broccoli (Table 11).

Occasionally, the same crop can be harvested for different purposes, with resulting effects on the quantity of nutrients removed from the field. For example,

the removal of nutrients from fields is much greater with corn silage, where close

to the entire aboveground corn plant is removed, than occurs with corn harvested

only for grain (Table 11). In a 9-year study, Vitosh er al. ( 1 973) found that N, P, and

K removal in silage was 60, 28, and 509% more from the soil, respectively, than

when grain was harvested.

The relative amounts of the various nutrients removed in crop harvest depends

on the species and the plant part(s) removed. For example, there is much less K

relative to N harvested in grain than when whole plants are harvested (Table 11).



Table I1

Approximate Amount of Nutrients Removed

in Harvested Portion of Selected Crops

(representativecommercial yields)



Corn grain

Corn silage


GrdS hdy




Brussel sprouts







Honeydew mellons'









































20 1










"Derived from data from Pennsylvania State University

State Agronomy Guide," assuming yields of 6,

15, 9, 6.7, and 2.5 tons ha-' for corn grain, corn silage, alfalfa, grass hay, and soybeans, respectively.

"From Lorenz and Maynard (I 980).

'From "New England Vegetable Guide," (Ferro, 19961997).

'/From Westwood (1978).

( 1994),"Penn

Some cropping systems, such as on a dairy farm or vegetable farm, that rely on

mixes of crops in the rotation can export significantly different amounts of nutrients and the net effect may be moderate average yearly field nutrient exports. By

contrast, when high yielding forages, such as corn silage and alfalfa, are grown in

the same field over a long time period, high levels of nutrients will be consistently exported from fields because most of the aboveground biomass leaves the field.

If grains are produced in the same field for some years there will be fewer nutrients exported out of the field than if forages are produced. A field producing vegetables, such as potatoes and tomatoes, for long periods will also be continually

exporting large amounts of nutrients. However, when high-nutrient export pota-



toes and/or tomatoes are rotated with crops such as broccoli and lettuce, the average annual nutrients exported from the field will be decreased.

2. Inadvertent Losses

Although removal of nutrients in crops is a necessary part of agriculture, conventional agriculture systems have had particularly large inadvertent nutrient losses from fields. The pathways of these losses, sometimes referred to as unmanaged

flows, include soil erosion (by water andor wind), soluble nutrients dissolved in

runoff waters, leaching of nutrients to below rooting depth, and gaseous losses

(mainly for one nutrient, N, as NH, volatilization and loss of N, and N,O resulting from denitrification).

These losses are an economic loss to the farm and at the same time may lead to

decreased water or air quality. In the United States, soil erosion alone is estimated to cause losses of $0.5 billion worth of available nutrients and $18 billion of total plant nutrients (Crosson, 1985; Troeh et al., 1980). In addition, off-farm economic damage from runoff and erosion is estimated at between $2 and $8 billion

(USDA, 1987). Soil erosion leads to movement of soil particles that enrich natural waters with N, P, and other nutrients as well as pesticides. Nitrate leaching from

fields contributes to poor groundwater quality and is a special health threat to infants. In many cases, systems leak because of the use of excess nutrients. In the

case of N, this is partially a result of fertilizer recommendations not taking sufficient account of the nitrate in the soil, soil organic matter mineralization, additions

of manure, legumes in the rotation, or nitrate in irrigation water. In addition, some

farmers do not follow the recommendations and apply excess or extra N because

they want insurance that they will have sufficient N or they assume they are starting with little or no sources already in the soil that might provide N for their crops.

However, concentrations of NO;-N in the percolating water underneath a corn

field will most likely be above 10 mg kg-I, even when fertilizer applications do

not exceed the economic optimum rate (Magdoff, 1992).

Holding unintended losses to an absolute minimum is an important goal for

sustainable agriculture. Methods for doing so include many different techniques

and practices. Of particular importance are those that reduce runoff and erosion

such as contour tillage and planting, strip cropping, use of terraces, grassed waterways, minimum tillage, use of sod crops as part of rotations, enhancing soil

structure through building up organic matter, etc. In addition to conserving soil

and water by improving moisture retention and reducing surface runoff and erosion, reduced tillage systems also allow farmers to reduce costs through saving

on fuel, labor, and equipment (Magleby et al., 1985; Phillips and Phillips, 1984).

In order to minimize unintended losses it is also important to maintain nutrient

applications close to levels needed to obtain maximum economic yield (see Quantity of Added Nutrients) as well as follow other nutrient management practices



that reduce losses (see Sources of Added Nutrients and Nutrient Application Timing and Methods).

For one element (N), there are other significant sources of loss aside from runoff

and erosion and leaching: volatilization of NH, and denitrification to gaseous N,O

and N,. Much research has been done on N use efficiencies and loss. In general,

approximately 5-10% of fertilizer N may be lost to the atmosphere in gaseous

forms (Kundler, 1970: Westerman et al., 1972).



1. Sources of Added Nutrients

Some sources are more susceptible to unintended loss than other sources. This

may be related to method of application as when ammonia volatilization occurs

with surface-applied urea or manures. The loss of ammonia from anhydrous ammonia can also be a problem when the soil does not seal quickly after injection.

Some biologically based sources of fertility, such as raw manure, may have high

quantities of available nutrients (Klausner, 1995). However, most of the nutrients

in other biological sources, such as plant residues and composted materials, are

usually less available in the short run and must first undergo mineralization. There

is some evidence that farms relying on slowly available organic sources of fertility might have more efficient soil-plant nutrient flows, thereby lessening the potential for loss in leaching waters or as eroded materials. For example, in a comparison of organic and conventional farms producing tomatoes in California,

Drinkwater et al. ( 1995) found that soils from the organic farms had lower nitrate

levels and a greater portion of the inorganic N in the ammonium form.

In an extensive evaluation of the fate of lSN from fertilizer and legumes, Harris et al. (1994) found that more fertilizer than legume N was recovered by crops

(40 vs 17% of input), but less fertilizer than legume N was retained in the soil (17

vs 47% of input), and similar amounts of N from both sources were lost from the

cropping systems over the 2-year period. Less legume than fertilizer N was lost

during the year of application (18 vs 38% of input), but more legume than fertilizer N was lost the year after application (17 vs 4% of input). Residual fertilizer

and legume IsN were distributed similarly among soil fractions. Soil microbial

biomass C and N was substantially larger in the legume-based system. A larger,

but not necessarily more active, soil microbial biomass was probably responsible

for the greater soil N-supplying capacity in the legume-based system compared

with the fertilizer-based system. These data suggest that legume- and fertilizerbased systems do have some similarities and some very significant differences.

The main difference is the amount of microbial biomass C and N. The yields of

the resulting crops are often very similar, and the N recovered from 1 year’s input



is usually higher in fertilizer systems. This suggests that in the legume system, numerous years of N inputs may contribute to the N-supplying ability of these systems, and that their N supply is related to a much larger microbial fraction. It may

also explain why yield losses result when the system is being changed from one

that relies exclusively on commercial synthetic fertilizers to legume- or organicbased practices.

It is also important to recognize that organic sources of fertility have many potential benefits aside from direct effects on plant nutrient supply. For example,

when a variety of sources of organic residues are added to soils they promote

greater biodiversity of soil organisms and may help reduce outbreaks of plant pests

(Dick, 1992). Organic residues also help microorganisms to produce phytohormones (Frankenberger and Arshad, 1995) and help promote better soil physical properties (Smith and Elliott, 1990), etc.

2. Nutrient Application Timing and Methods

The efficiency of nutrient use by plants and, hence, the amount of unintended

losses, is often influenced by the timing of the application and/or the application

method. [Timing of application is frequently implied by method (for example,

side-dress and top-dress are done during the growing season, whereas in-ground

band placement is usually done near the seed at planting, and broadcast is either

done in fall or spring when the ground is bare or as a top-dress for certain crops

during season).]

When using amendments with low solubilities, such as rock phosphate, limestone, and gypsum, it is important to mix them with the soil as thoroughly as possible and have some time for them to react before corps are planted. Complete incorporation of fertilizer P and K is frequently done to raise the general fertility

level of soils that test low in these elements. Incorporation of manures is also often desirable in order to reduce ammonia losses, reduce risk of runoff losses, and

introduce organic matter deeper into the topsoil. However, either in-season or atplanting localized placement of soluble synthetic sources of N and P (which are

susceptible to loss in runoff, erosion, and leaching or are commonly converted into

forms that are not available to plants) helps to enhance uptake and thus may reduce losses (Mahler et al., 1994; Lathwell ef al., 1970; Welch ef al., 1966, 1971).

There are certainly pros and cons to preplant, at-planting, and in-season N fertilizer application. In-season fertilizer N application relies on precipitation to bring

the nutrient into intimate contact with roots and thus is not a reliable system for

relatively dry but unirrigated conditions. However, under humid conditions, the

longer the time between N application and plant uptake, the greater the possibility for nutrient loss.

Randall and Hoeft (1988), in a review of the literature, related fertilizer efficiency to soil test level and placement with corn, soybeans, and small grain. They



concluded that corn and soybeans usually do not respond positively to P and K on

medium or high soil test levels. At low soil test level corn responds more positively

to banded fertilizer than broadcast fertilizer, whereas broadcast is sufficient with

soybeans on low testing soils. Small grains generally respond better to banded

rather than broadcast fertilizer where a response is achieved.

There has been considerable interest in application systems that involve more

intensive soil sampling and changing the rate of fertilizer application within fields

based on soil test levels. Referred to as variable rate or precision farming technology, these systems involve the use of specialized equipment to map fields, monitor location within the field, and to change the fertilizer application rate rapidly.

Usually used for applying preplant broadcast fertilizers, some systems involve onthe-go sensors, eliminating the need for previously testing and mapping the field

and providing the possibility of use for N side-dress of corn. These systems can

theoretically enhance the efficiency of nutrient use by taking soil variability into

account and applying only the amounts of fertilizer needed to different parts of the

field. However, the degree to which they influence total nutrient application rates

and flows has not been established.

3. Quantity of Added Nutrients

The goal of most soil fertility programs is to have the amount of nutrients present that is needed to produce maximum economic crop yield. If a nutrient is not

supplied in sufficient quantities to achieve this goal, then yields suffer. If a nutrient is added above the level needed to achieve maximum economic yields, then

the extra flow onto the field is both an unneeded expense and a potential environmental hazard.

Many technical and human factors combine to determine the actual quantity of

nutrients added to a particular field. Although the use of soil testing and plant

analysis as means of assessing the nutrient status of fields is common, the majority of farmers do not utilize these tools. They use experiential knowledge based on

their observations of soil conditions and crop yields, intuition, observations of their

neighbors’ fields, and discussions with fertilizer salesmen or other agricultural suppliers and professionals. The land base available for manure application frequently affects the nutrient loading rate on farms with animals.

The soil testing process and the complexities and pitfalls of recommendation

systems are poorly understood by farmers as well as by many agricultural professionals. It is commonly believed that soil testing provides a precise and scientific

assessment of the soil’s nutrient status. In actual practice soil testing and recommendation systems are a mix of science, art, probabilities, compromises, and other human factors.

There are two separate but critically important parts of the process: (i) A good

soil test must reflect plant availability (low, medium, and high soil test levels must



reflect high, medium, and low probability, respectively, of increasing plant yields

by adding a particular nutrient); and (ii) the system for recommending nutrient additions based on the soil test should suggest amounts of added nutrients that will

not sacrifice yield nor recommend greater application rates than are economically

and environmentallyjustified.

Rates of nutrient additions recommended by soil testing laboratories vary substantially for N, P, and K (Liebhardt et al., 1982a,b,c). These variations are partially due to all the technical and human factors involved in soil testing and recommendations, but primarily result from differing soil test philosophies and

motivations of people interpreting the soil test. There are three main recommendation philosophies currently being used. One approach is the “sufficiency level,” which was derived from research that reveals no yield response to an applied

nutrient above a certain “critical” soil test level. Soils testing low and very low

have a high probability of a significant yield response to added nutrients. Crops

grown on soils testing medium have a lower probability of responding to nutrient additions and soils testing high and very high are unlikely to respond to nutrient additions. A second philosophy, the “cation saturation ratio” concept, suggests a certain balance or ratio of Ca, Mg, and K and calls for additions to bring

the ratio of these cations into certain ranges. A third approach is the “build-up and

maintenance” concept. This system calls for adding nutrients (mainly P and K) to

attain high soil test levels and then adding annually what is removed in the harvest. Field research has conclusively shown that the sufficiency level approach,

which generally recommends lower rates of nutrients, is superior to the cation saturation and the maintenance approaches with respect to agronomic, economic,

and environmental factors (Olson et al., 1982; McLean et al., 1983; Liebhardt,


Another factor that affects the accuracy of recommendation systems is that good

soil tests only separate sites with high probability of response to added nutrients

from those with low probability of response. It is frequently found that within the

population of sites that will likely benefit from nutrient additions (i.e., low soil test

values), the extent of the response is not sensitive to soil test level (Magdoff et al.,

1993). Thus, although more fertilizer is generally recommended as soil test levels

decrease further below a critical value, there are little crop response data to justify that approach. Other factors in addition to the soil test level of the particular nutrient, such as rooting depth, organic matter mineralization, bulk density, plant

pests, and weather, are apparently influencing the extent of the response at these

sites. In addition, even when using the sufficiency level system for making recommendations, different models for describing the same data can result in varying fertilizer recommendations. For example, Cerrato and Blackmer (1990) found

that economic returns were greater when using a linear-plateau model to describe

plant response to fertilizer N than for other models, such as a quadratic curve.

Because of the complex nature of the biological and chemical reactions N un-



dergoes, the development of soil testing and recommendation procedures for this

element has been a greater challenge than for P or K. Nitrogen recommendations

for corn, the crop on which the largest amount of fertilizer N is used in the United States, were historically based mainly on the farmer’s yield goal. For example,

if the yield goal was 9408 kg ha-’ (150 bu acre- I ) and it was estimated that crop

needs 0.022 kg N kg-’ (1.25 Ibs N bu- I ) , then N required was approximately 210

kg kg-l (187.5 Ibs acre-’). In this system it was assumed that there was no legume

or manure, no soil N carryover or mineralization, and no N in irrigation water or

in any other source. This “put back what you take out” philosophy is simple but

does not take into account many important factors and results in economic and environmental costs. An improved “tax form” recommendation system continues

where the previous system leaves off and attempts to credit various sources of

available N. In the case in which the estimated total requirement is 2 10 kg ha- I ,

estimates of contributions from sources such as SOM, fertilizer N carryover from

the previous year, legume plowdown, manure additions, or N in irrigation water

can be subtracted from the total estimated need. If these sources are not sufficient,

then N fertilizer should be added. It is an improvement but it still makes many

guesses and assumptions about N behavior, crop growth, and environmental conditions.

The first soil test for N that gained widespread acceptance was in the semiarid

western corn belt where there is little leaching and denitrification and SOM levels are low and not very active. In this system, soil nitrate in the soil profile to at

least 3 ft, and sometimes to 6 ft, is determined in the fall or spring and is used to

estimate fertilizer N needed. The pre-side-dress nitrate test (PSNT) (Magdoff et

al., 1984 and Magdoff 1991a), is a new approach to N testing now used by many

states in the humid regions of the Northeast, mid-Atlantic, and Midwest. The

premise of this concept is to evaluate soil nitrate in the top 30 cm of soil when

corn height is 15-30 cm. The test is late enough in the season to account for a significant quantity of mineralization of SOM, legume residues, and manure and early enough to still be able to side-dress N fertilizer when soil nitrate is below the

critical value.

During the past few decades, there have been improvements in many fertilizer

recommendation systems, but few major advances in soil testing and recommendation systems have been adopted for routine use by testing laboratories. These

have been limited to adoption of a new extract, the Mehlich 3 (Mehlich, 1984),

by a number of laboratories, and the PSNT (Magdoff et al., 1984; Magdoff,

199I a).

Soil testing will become more important in the future as farmers, testing laboratories, and various governmental agencies try to “fine-tune” recommendations

to take into account concerns of both yields and the environment. More research

is needed to enhance the accuracy of the quantity of nutrients recommended for

sites that will probably show a positive response.






Soil and crop management practices that do not involve direct additions of nutrients from off the field or harvest losses also have profound impacts on nutrient

flows. These include the use of crop residues, system(s) of tillage practiced, use of

cover crops, and the characteristics of the growing crop and the other crops grown

in rotation. The influences of these factors are primarily on the magnitude of nutrient loss, but there are a number of influences on nutrient additions as well.

1. Crop Residues

Cycling within the field setting is mainly confined to the soil+crop+soil pathways through residues remaining in the field after harvest and by use of cover

crops. When residues are returned to the soil they help retain plant nutrients, maintain soil porosity and tilth, enhance water infiltration, and act as an effective control against water erosion (Lindstrom and Holt, 1983). Crop residues may contain

substantial amounts of biomass as well as nutrients, as in the case of broccoli or

corn grown for grain. On the other hand, certain crops such as corn silage leave

few residues except for the root systems. Cycling of nutrients in residues is relatively rapid under conventional moldboard plow-harrow tillage systems in which

residues are annually incorporated into the soil. Under reduced tillage systems, especially no-till, residues accumulate near the surface where they generally decompose more slowly. When residues are incorporated into soil by tillage or biological activity or accumulate on the surface, organisms begin the process of

decomposition, during which nutrients are mineralized. The amount of nutrients

contained in crop residues in the United States is sizable, with estimates that they

contain 24% of the N, 13% of the P, and 34% of the K applied to cropland (Follett

et al., 1987). Nutrients in Midwest crop residues as a percentage of those in commercial fertilizer are also large, averaging approximately 75% for N, 35% for P,

and 90% for K (Holt, 1979). In addition to the direct effects on cycling nutrients,

residues left on the soil surface reduce wind and water erosion, thus reducing loss

of nutrients from the field.

Although crop residues may be plentiful on a national basis, not all available

residues contribute to a field-scale nutrient cycle. Residues are sometimes removed to become another commodity for sale as when straw from small grain is

sold to another farm for use as animal bedding. Other potential uses of crop

residues that are under exploration are to produce “biofuels” and feedstocks for industrial biochemical products. In some regions of the United States crop residues,

such as rice and wheat straw, are routinely burned, with loss of most of the N and

S as volatilized gases. This practice is being discouraged because of the resulting

air pollution. Competition for crop residues is particularly critical in developing

countries, where they commonly serve as important sources of fuel or building ma-



terials. Although there may be good economic reasons to remove crop residues,

this results in the loss of valuable nutrients as well as a source of material to help

maintain SOM.

2. Tillage Systems

Tillage practices have both short-term and long-term consequences, particularly in relation to soil erosion and soil structure, and thus are key components of soil

management strategies for sustainable land management, crop production, and

agriculture as a whole (Cannell and Hawes, 1994). The early criticisms of the

widespread reliance on the moldboard plow, such as by Faulkner (1943), were considered controversial. However, it is now widely accepted that there are potentially severe problems associated with reliance on it as the main form of tillage. Hillel(1991) summarized the situation as follows:

Although the development of the plow represented a huge advance in terms of

convenience and efficiency of operation, it had an important side effect. As with

many other innovations, the benefits were immediate, but the full range of consequences took several generations to play out, long after the new practice became entrenched. The major environmental consequence was that plowing

made the soil surface-now loosened, pulverized, and bared of weeds-much

more vulnerable to accelerated erosion.

The soil disturbance caused by plowing and subsequent harrowing also accelerates SOM decomposition, and more organic materials must be returnedadded

to the soil to maintain or build up SOM levels. Thus, conventional tillage systems

tend to promote loss of nutrient stocks from soils by wind and water erosion as

well as by accelerated SOM mineralization. Loss of cation-exchange capacity with

decreasing SOM also encourages the leaching of cations such as Ca, K, and Mg.

Plowing is different from a natural system in that it disrupts soil and buries organic matter deep in the soil rather than allowing it to be worked in at the top layers by soil organisms. One of the main ways to decrease runoff and erosion losses of soluble nutrients and soil solids with their associated nutrient stock is to

follow practices that enhance the development of macropores for conducting water deep into the profile. In addition, a stable and porous structure should be maintained at the soil surface so that water can infiltrate rapidly even under the potentially damaging impact of rainfall and agricultural machinery traffic. Reduced

systems of tillage allow macropores to develop (such as by earthworms under notill practices) and maintains more of a vegetative cover than with moldboard plow

systems. Organic residues on the surface provide food for earthworms and also

lessen the force of rainfall on the soil surface. Higher organic matter content of notill surface soils also helps to build and stabilize soil structure.

Tillage methods are changing rapidly. Conservation tillage in the United States



is increasing and is currently used on approximately 30% of cropland, whereas notill methods are employed on approximately 10%of cropland (Cannell and Hawes,

1994). Experiments in Mississippi demonstrate the dramatic positive effects of the

conservation tillage systems, where no-till planting with the residues left on the

surface reduced soil losses due to water erosion by 85% (from 17.5 to 2.5 Mg hayear-’) compared to plowing (McGregor etal., 1975). Another study on soils that

were susceptible to severe soil erosion compared conventional tillage with flutedcoulter and in-row chisel techniques and found that the latter two methods essentially eliminated sediment losses (Langdale and Leonard, 1983). Fluted-coulter

tillage reduced runoff 50% compared to conventional tillage, whereas in-row chisel tillage reduced runoff 90%. Although concentrations of soluble N and P increased in runoff in the reduced tillage treatments, except for PO,-P associated

with the fluted-coultertillage, nutrient loss in runoff was drastically decreased with

both reduced tillage systems. Stinner etal. (1988) found that tilled agroecosystems

are much more susceptible to nutrient loss than are natural systems such as forests.

No-tillage conditions tend to favor nutrient conservation, as measurements in both

sorghum and soybean systems demonstrate.

In addition to positive effects on runoff and erosion, conservation tillage reduces

costs of fuel, labor, and equipment and conserves soil and water by improving

moisture retention and reducing surface runoff and erosion (Magleby et al., 1985;

Phillips and Phillips, 1984). In addition, no-till creates conditions resembling

undisturbed ecosystems (Blumberg and Crossley, 1983; Fleige and Baeumer,

1974). Despite similarities between natural and no-till soils, it is not clear how

much conservation practices modify decomposition process, thereby altering nutrient cycles in agroecosystems (Hendrix et al., 1986). Hendrix et al. (1986) concluded that tillage increases nutrient mobility, and no-till soils are usually physically stratified with more nutrients localized near the surface.

3. Type of Crops

a. Rotations

Rotations have long been the cornerstone of well-managed agricultural systems.

There are a number of sound reasons for rotating crops, including an increase in

yield that is frequently obtained over that obtained under monoculture-sometimes called the “rotation effect.” Rotations help manage insect, weed, and disease

cycles and confer a degree of protection against weather and pest problems. In addition, crop sequences, which include forage legumes or fibrous-rooted grass sod

crops, often enhance soil microbial populations and activity (Bolton e f al., 1985)

as well as soil structure. Although NOT-N leaching to groundwater from soils under corn is especially problematic (Magdoff, 1991b), little leaching of NO;-N occurs under sod crops (Olson et al., 1970) except where very high N rates are used

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

IV. Cycling and Flows at the Field Level

Tải bản đầy đủ ngay(0 tr)