Plant Soil

DOI 10.1007/s11104-007-9193-9

ORIGINAL PAPER

Long term eVects of manure, charcoal and mineral fertilization on crop production and fertility on a highly weathered Central Amazonian upland soil

Christoph Steiner · Wenceslau G. Teixeira · Johannes Lehmann · Thomas Nehls · Jeferson Luis Vasconcelos de Macêdo · Winfried E. H. Blum · Wolfgang Zech

Received: 16 September 2006 / Accepted: 2 January 2007© Springer Science+Business Media B.V. 2007

Abstract Application of organic fertilizers andcharcoal increase nutrient stocks in the rootingzone of crops, reduce nutrient leaching and thusimprove crop production on acid and highlyweathered tropical soils. In a Weld trial near Man-aus (Brazil) 15 diVerent amendment combina-tions based on equal amounts of carbon (C)applied through chicken manure (CM), compost,charcoal, and forest litter were tested during fourcropping cycles with rice (Oryza sativa L.) andsorghum (Sorghum bicolor L.) in Wve replicates.CM amendments resulted in the highest(P < 0.05) cumulative crop yield (12.4 Mg ha¡1)over four seasons. Most importantly, surface soilpH, phosphorus (P), calcium (Ca), and magne-

sium (Mg) were signiWcantly enhanced by CM. Asingle compost application produced fourfoldmore grain yield (P < 0.05) than plots mineral fer-tilized in split applications. Charcoal signiWcantlyimproved plant growth and doubled grain pro-duction if fertilized with NPK in comparison tothe NPK-fertilizer without charcoal (P < 0.05).The higher yields caused a signiWcantly greaternutrient export in charcoal-amended Welds, butavailable nutrients did not decrease to the sameextent as on just mineral fertilized plots.Exchangeable soil aluminum (Al) was furtherreduced if mineral fertilizer was applied withcharcoal (from 4.7 to 0 mg kg¡1). The resilience ofsoil organic matter (SOM) in charcoal amendedplots (8 and 4% soil C loss, mineral fertilized ornot fertilized, respectively) indicates the refrac-tory nature of charcoal in comparison to SOMlosses over 20 months in CM (27%), compostamended (27%), and control plots (25% loss).

Keywords Black carbon · Brazil · Organic agriculture · Oxisol · Terra Preta de Indio

Introduction

Slash and burn agriculture is practiced by about300–500 million people globally, aVecting almostone third of the planet’s 1,500 million ha of arableland (Giardina et al. 2000; Goldammer 1993).

C. Steiner (&) · T. Nehls · W. ZechInstitute of Soil Science and Soil Geography, University of Bayreuth, 95440 Bayreuth, Germanye-mail: Christoph.Steiner@uni-bayreuth.de

W. G. Teixeira · J. L. V. de MacêdoEmbrapa Amazonia Ocidental, CP 319, Manaus, AM, 69011-970, Brazil

J. LehmannDepartment of Crop and Soil Sciences, Cornell University, Ithaca, NY, 14853, USA

W. E. H. BlumInstitute of Soil Research, University of Natural Resources and Applied Life Sciences (BOKU), 1180 Vienna, Austria

1 3

Plant Soil

This traditional agricultural practice is consideredto be sustainable if adequate fallow periods (up to20 years) are following a short time of cultivation(Kleinman et al. 1995). A growing populationwith changing socio-economic habits may not beable to practice slash and burn in a sustainableway. In most agricultural systems the tendencyhas been for population pressure to increase,leading to shorter fallow periods (Fearnside1997). SigniWcant amounts of nutrients, mainlynitrogen (N) and sulfur (S) (Giardina et al. 2000;Hölscher et al. 1997a; Hughes et al. 2000;Kuhlbusch et al. 1991), and organic matter (OM)are lost by burning forests during the conversionof native vegetation to pasture or crop land.Accelerated OM loss through burning disruptssoil organic matter (SOM) formation andextended cultivation periods further reduce theSOM contents correlating with soil nutrientdepletion (Goldammer 1993; Hölscher et al.1997b; Silva-Forsberg and Fearnside 1995; Zechet al. 1990). In strongly weathered tropical soils,SOM plays a major role in soil productivity (Ties-sen et al. 1994) because it represents the domi-nant reservoir of plant nutrients such as N, P, andS. Generally, SOM contains 95% or more of thetotal N and S, and between 20 and 75% of the P insurface soils (Duxbury et al. 1989). Thus, long-term intensive land use where SOM stocks aredepleted, is not sustainable without nutrientinputs (Tiessen et al. 1994). On soils with low-nutrient retention capacity the strong tropicalrains easily leach available and mobile nutrients,such as inorganic N fertilizers, rapidly into thesubsoil where they are unavailable for most crops(Giardina et al. 2000; Hölscher et al. 1997a;Renck and Lehmann 2004) rendering conven-tional fertilization highly ineYcient.

Reducing losses of nutrients and C resultingfrom forest clearing require alternatives to slashand burn (Lewis et al. 2002) and alternative fertil-ization methods. Depending on the mineraliza-tion rate, organic fertilizers such as compost,mulch or manure release nutrients in a gradualmanner (Burger and Jackson 2003) and maytherefore be more appropriate for nutrient reten-tion under high-leaching conditions than inor-ganic fertilizers. But, as mentioned above, notonly nutrients but also C from the vegetation and

soils is lost over time during deforestation andsubsequent land use. Such a C loss reduces theability of the soil to retain nutrients especially insoils with low-activity clays such as in highlyweathered soils of the humid tropics (Sanchez1976).

Application of charred biomass as an alterna-tive to manures or composts seems to be a prom-ising option to maintain a maximum of C in soilsas charring signiWcantly increases the stability ofC against microbial decay (Baldock and Smernik2002). However, charcoal represents just 1.7% ofthe pre-burn biomass if a forest is converted bythe traditional slash and burn technique (Fearn-side et al. 2001). The purposeful production ofcharcoal for soil application is able to increase theproportion that can be applied in such shiftingcultivation systems (Lehmann et al. 2006).

The existence of an anthropogenic and C-enriched dark soil in diVerent parts of the worldand especially in Amazonia (Amazonian DarkEarths (ADE) or Terra Preta de Indio) makes itlikely that a similar strategy existed before thearrival of the Europeans. Without steel axes andmodern tools for deforestation it is more likelythat soil fertility was maintained by the AmazonIndians with rich organic inputs instead of clear-ing new forests when soil fertility decreased(Denevan 1996). The ADE’s fertility is mostlikely linked to an anthropogenic accumulation ofP and Ca associated with bone apatite (Lehmannet al. 2004; Lima et al. 2002; Zech et al. 1990) andblack C as charcoal (Glaser et al. 2001a). The highpersistence of charcoal is responsible for the sta-bility of the ADE’s SOM. Today and as assumedalso in the past those soils have been intensivelycultivated by the native population. The existenceof ADE proves that infertile Oxisols can in princi-ple be transformed into fertile soils. However,this transformation was not solely achieved byreplenishing the mineral nutrient supply, butrelies on the addition of stable C in the form ofcharcoal.

The sustained fertility in charcoal-containingADE and the frequent use of charcoal as a soilconditioner (Steiner et al. 2004b) in Brazil andother parts of the world (mainly Japan) Ogawa(1994) provided the incentive to study the eVectsof charcoal application to a highly weathered soil

1 3

Plant Soil

(Lehmann et al. 2003). Lehmann et al. (2002) andSteiner et al. (2004b) described slash and char asan alternative agricultural method producingcharcoal from the aboveground biomass insteadof converting it to CO2 through burning. Slashand char practiced as an alternative to slash andburn throughout the tropics could be an impor-tant step toward sustainability by SOM conserva-tion in tropical agriculture. Therefore, weaddressed the following objectives: (1) to quantifythe eVects of charcoal, organic, and inorganic fer-tilization on soil fertility and crop production; and(2) to evaluate the sustainability of charcoal addi-tions in terms of maintaining high-SOM contentsand nutrient availability with special emphasis onTerra Wrme Ferralsols near Manaus. It is hypothe-sized that charcoal additions improve nutrientavailability and that crop productivity is main-tained on a higher level than without charcoaladditions.

Materials and methods

The experiment was established 30 km north ofManaus, Amazonas, Brasil (3°8�S, 59°52�W, 40–50 m a.s.l.) at the Embrapa-Amazônia Ocidental(Empresa Brasileira de Pesquisa Agropecuaria)experimental research station. The natural vege-tation is evergreen tropical rainforest with a meanannual precipitation of 2,530 mm (1971–1997)having its seasonal maximum between Decemberand May, a mean annual temperature of 25.8°C(1987–1997) and a relative humidity of 85% (Cor-reia and Lieberei 1998). After clearing of about3,600 m2 secondary forest and removing theaboveground biomass the experiment was estab-lished on a highly weathered Xanthic Ferralsol(FAO 1990) derived from Tertiary sediments.The soil is Wne textured with up to 80% clay. It isstrongly aggregated and has medium contents oforganic C (24 g kg¡1), low-pH values of 4.7 (inH20), low CEC of 1.6 cmolc kg¡1 and low-basesaturation (BS) of 11.2%.

About 15 randomized treatments were appliedon 4 m2 plots (2 £ 2 m2) in Wve replicates formingan entire Weld area of 1,600 m2 (45 £ 35 m2) witha minimum distance to the surrounding vegeta-tion of 10 m. The plots were protected against

runoV water by metal sheets. Rice was plantedinto the spaces between plots as an active barrierin order to decrease the risks of cross contamina-tion by runoV.

Charcoal amendments (2CC)

Charcoal amendments were considered rather assoil conditioner than fertilizer due to the char-coal’s low-nutrient contents (Table 1). Previouspot experiments revealed advantageous eVects ofsoil charcoal amendments in addition to anydirect nutrient additions (Lehmann et al. 2003).Terra Preta research has shown that oxidation onthe edges of the aromatic backbone and adsorp-tion of other OM to charcoal is responsible forthe increased CEC, though the relative propor-tion of these two processes is unclear (Liang et al.2006).

Charcoal derived from secondary forest wood,was bought from a local distributor. It was manu-ally crushed to particle sizes smaller than 2 mm.The applied 11 mg ha¡1 charcoal corresponded tothe amount of charcoal-C which could be pro-duced by a single slash-and-char event of a typicalsecondary forest on Xanthic Ferralsols in centralAmazonia (Lehmann et al. 2002). The amount ofC added with charcoal was chosen as a referencevalue for adding the compost, litter, and chickenmanure (CM) amendments.

Mineral fertilization (F)

Mineral fertilizer (NPK and lime) was applied asammonium sulfate ((NH4)2SO4), simple superphosphate, and potassium chlorite (KCl) as rec-ommended by Embrapa (Fageria 1998). In addi-tion to one solely inorganically fertilizedtreatment (F) all organically treated plots despiteCM and litter (L) had one replicate with addi-tional inorganic fertilization. Mineral fertilizerwas applied in March 2001 and after the secondharvest in April 2002 (Table 1). At the second fer-tilization the treatments L, and 2CCp + CO + Fwere additionally fertilized with micronutrients(Table 1). Those treatments received mineral fer-tilization for the Wrst time. While the2CCp + CO + F treatment with micronutrientswas paired with one treatment without micronu-

1 3

Plant Soil

Tab

le1

Tre

atm

ents

and

app

licat

ions

of

orga

nic

mat

ter

(mg

ha¡

1 ) an

d nu

trie

nts

(kg

ha¡

1 )

1N

utri

ent

cont

ents

of

the

ash

wer

e no

t de

term

ined

, N w

as p

roba

bly

lost

to

a gr

eat

exte

nt, P

, K, C

a an

d M

g w

ere

assu

med

to

rem

ain

in a

sh a

fter

the

in W

eld

burn

ing

of fo

rest

litt

er

Tre

atm

ent

Org

anic

mat

ter

(mg

ha¡

1 )N

utri

ent c

onte

nts

of

orga

nic

mat

ter

(kg

ha¡

1 )F

irst

fer

tiliz

atio

n (k

gha

¡1 )

Seco

nd fe

rtili

zati

on (

kgha

¡1 )

CC

ontr

ol–

––

LL

itte

r (1

3)N

(11

4), P

(0.

31),

K (

4.30

),

Ca

(13.

28),

Mg

(4.6

9)–

N (

55),

P (

40),

K (

50),

lim

e (2

,800

»61

3 C

a, 3

60 M

g), Z

n (7

),

B (

1.4)

, Cu

(0.6

), F

e (2

.3),

Mn

(1.6

), M

o (0

.08)

LB

1B

urne

d lit

ter

(13

mg

of d

rylit

ter

burn

ed o

n th

e pl

ot)

N (

???)

, P (

0.31

), K

(4.

30),

C

a (1

3.28

), M

g (4

.69)

–N

(55

), P

(40

), K

(50

),

lime

(2,8

00»

613

Ca,

360

Mg)

FM

iner

al f

erti

lizer

–N

(30

), P

(35

), K

(50

), li

me

(2,1

00»

460

Ca,

270

Mg)

N (

55),

P (

40),

K (

50),

lim

e (4

30»

94 C

a, 5

5 M

g)C

MC

hick

en m

anur

e (4

7)N

(77

4), P

(32

4), K

(83

6),

Ca

(784

), M

g (1

43)

––

2CO

Com

post

(67

)N

(68

1), P

(49

), K

(19

1),

Ca

(219

), M

g (1

01)

––

2CC

Cha

rcoa

l (11

)N

(59

), P

(0.

29),

K (

2.52

),

Ca

(9.0

0), M

g (1

.87)

––

2CO

+F

Com

post

(67

)N

(68

1), P

(49

), K

(19

1),

Ca

(219

), M

g (1

01)

N (

30),

P (

35),

K (

50),

lim

e (2

,100

»46

0 C

a, 2

70 M

g)N

(55

), P

(40

), K

(50

),

lime

(430

»94

Ca,

55

Mg)

2CC

+F

Cha

rcoa

l (11

)N

(59

), P

(0.

29),

K (

2.52

),

Ca

(9.0

0), M

g (1

.87)

N (

30),

P (

35),

K (

50),

lim

e (2

,100

»46

0 C

a, 2

70 M

g)N

(55

), P

(40

), K

(50

),

lime

(430

»94

Ca,

55

Mg)

CC

+C

OC

harc

oal (

5,5)

, co

mpo

st (

33,5

)N

(37

0), P

(24

.5),

K (

96.8

),

Ca

(114

), M

g (5

1.6)

––

CC

+C

O+

FC

harc

oal (

5,5)

, co

mpo

st (

33,5

)N

(37

0), P

(24

.5),

K (

96.8

),

Ca

(114

), M

g (5

1.6)

N (

30),

P (

35),

K (

50),

lim

e (2

,100

»46

0 C

a, 2

70 M

g)N

(55

), P

(40

), K

(50

),

lime

(430

»94

Ca,

55

Mg)

2CC

+C

OC

harc

oal (

11),

co

mpo

st (

33,5

)N

(39

9), P

(24

.7),

K (

98.1

),

Ca

(118

.6),

Mg

(52.

5)–

–

2CC

+C

O+

FC

harc

oal (

11),

co

mpo

st (

33,5

)N

(39

9), P

(24

.7),

K (

98.1

),

Ca

(118

.6),

Mg

(52.

5)N

(30

), P

(35

), K

(50

),

lime

(2,1

00»

460

Ca,

270

Mg)

N (

55),

P (

40),

K (

50),

lim

e (4

30»

94 C

a, 5

5 M

g)2C

Cp

+C

OC

harc

oal p

iece

s (1

1),

com

post

(33

,5)

N (

399)

, P (

24.7

), K

(98

.1),

C

a (1

18.6

), M

g (5

2.5)

–N

(55

), P

(40

), K

(50

),

lime

(2,8

00»

613

Ca,

360

Mg)

, Z

n (7

), B

(1.

4), C

u (0

.6),

Fe

(2.3

),

Mn

(1.6

), M

o (0

.08)

2CC

p+

CO

Cha

rcoa

l pie

ces

(11)

, co

mpo

st (

33,5

)N

(39

9), P

(24

.7),

K (

98.1

),

Ca

(118

.6),

Mg

(52.

5)–

N (

55),

P (

40),

K (

50),

lim

e (2

,800

»61

3 C

a, 3

60 M

g)

1 3

Plant Soil

trients, the litter treatment was not identicallypaired, but the treatments C, L, and burned litter(LB) did not show any diVerences at the Wrst har-vest and had no biomass production at all at thesecond harvest.

Organic amendments (L, LB, CM, 2CO)

Organic materials were applied just once at thebeginning of the experiment (February 3, 2001).The amount of applied OM (litter, L; CM; com-post, 2CO; and leaf litter, L) was calculated fromthe total C content of the materials to increasetotal soil C content in 0–10 cm depth by 25%.Compost was prepared from biomass of a second-ary forest, fruit residues, manure, and kitchenwaste. The CM was bought at a nearby chickenfarm and the litter was collected on the experi-mental site. From February 12–20, 2001, the Weldswere hoe-harrowed to 0.10 m depth and theorganic amendments were mixed in with the soil.LB was established with the intention to simulateslash and burn farming. Unburned litter (L) wasof interest in order to study alternative cultivationwithout burning (slash and mulch). Compost andCM are available organic fertilizers. Compost israther expensive in Manaus but could be pro-duced by land-owners themselves. CM is abun-dant due to a large poultry industry in the vicinityof Manaus and therefore relevant for land-own-ers.

Combination treatments (CC + CO, 2CC + CO, CCp + CO)

Combination treatments were established tostudy the eVects of organic nutrients appliedtogether with charcoal. Carbon was appliedabove the reference value (1.5 times) if the fullcharcoal dose (2CC = reference value) wasapplied as a degree of comparison. All combina-tion treatments were established with and with-out mineral fertilization. Charcoal applied aspieces (»10 mm in diameter) instead of pow-dered charcoal (<2 mm) was used to study theeVect of charcoal size. A combination of CM andcharcoal was missing in this study, but was stud-ied in other experiments (Steiner et al., unpub-lished).

Crops

As a Wrst crop rice (Oryza sativa L.) was plantedfollowed by three repeated sorghum (Sorghumbicolor L. Moench) crops. Rice was plantedMarch 10, 2001 in a density of 200 seeds per m2,followed by sorghum planted on October 15, 2001in a density of 12.5 plants per m2. The 3rd cropwas established in a density of 25 plants per m2 onApril 18, 2002, the latter producing two harvestsby ratooning. The stover and grain yields wereassessed on July 7, 2001, February 6, 2002, July21, 2002, and October 16, 2002. To minimize bor-der eVects, two plant rows were left at each har-vest on each side of the plots. The plant materialwas stored in paper bags and immediately dried at65°C until constant weight. Rice grains and strawwere weighed separately. Biomass samples wereground with a ball mill and stored for nutrientanalysis. The dried and weighed crop residueswere brought back to the plots of origin andremained on the Weld for decomposition. Only theground plant samples (»10 g) for foliar nutrientanalyses were removed from the system.

Soil samples and sample analyses

Soil samples (at depths of 0–0.1, 0.1–0.3, and 0.3–0.6 m) were taken at the beginning of the experi-ment (after OM application but before mineralfertilization) and later on after each harvest (Feb-ruary 23, 2001, July 11, 2001, February 15, 2002,July 23, 2002, and November 22, 02). Two sam-ples per plot were taken, combined to one com-posite sample, and then air dried and sieved topass 2 mm.

For the determination of foliar nutrient con-tents a digestion with a mixture of H2SO4, salicylicacid, H2O2, and selenium was used according toWalinga (1995). For the extraction of exchange-able P, K, Ca, and Mg, the Mehlich¡3 extractionwas used without modiWcation (Mehlich 1984).Calcium and Mg in the Wltered solutions were ana-lyzed using atomic absorption spectrometry (AA-1475, Varian Associates, Inc., Palo Alto, CA,USA). Potassium was analyzed with a Xame pho-tometer (Micronal B 262, Sao Paulo, SP, Brazil).Phosphorus was measured using a photometer(He�ios ß, Thermo Spectronic, Cambridge, UK)

1 3

Plant Soil

with the molybdene blue method (Olsen andSommers 1982). pH was determined in water and1 N KCl (1:5 w/v) using an electronic pH meterwith a glass electrode (WTW pH 330, WTW,Weilheim, Germany) and electric conductivity bya conductivity meter (HI 8733, HANNA Instru-ments, Kehl am Rhein, Germany). Total C and Nwere analyzed by dry combustion with an auto-matic C/N- Analyzer (Elementar, Hanau, Ger-many).

Plant-available NH4 and NO3 were determinedphotometrically in soil extracts (in 1 N KCl) usinga rapid Xow analyzer (Scan Plus analyzer, SkalarAnalytical B.V., Breda, The Netherlands).Exchangeable acidity and exchangeable Al weredetermined by titration (McLean 1965) afterextraction with 1 N KCl. Cation exchange capac-ity was calculated as the sum of ammonium ace-tate-exchangeable cations and acidity (Claessenet al. 1997).

Statistical analyses

Treatment eVects were analyzed by general linearmodel (GLM) univariate analysis of variance(ANOVA). Most parameters were not normallydistributed and did not have equal variances. ForreWtting a Box Cox transformation (Box and Cox1964) was used. Homogeneous subsets were sepa-rated by the Student–Newman–Keuls test and theFisher’s LSD (least signiWcant diVerence) wasinserted into Wgures. Statistical analyses and plotswere performed using SPSS 12.0 and SigmaPlot(SPSS Inc., Chicago, IL, USA).

Results

Stover and grain yield

Total cumulative stover production (all harvests)decreased in the order CM > 2CO + F > 2CO =2CC + CO + F = CC + CO + F > CCp + CO = 2CC + F > 2CC + CO = CC + CO = F > L = LB > 2CC > C. Cumulative grain yield showed similarresults, but the diVerence between charcoal con-taining plots to plots without charcoal was evengreater than for total stover. Mineral fertilizedplots with additional charcoal application hadalmost twice as much yield as only mineral fertil-ized plots (Fig. 1a). CM proved to be the mosteVective treatment within this experiment. Totalgrain yield (all harvests together 12.4 Mg ha¡1,Fig. 1a) and total stover production(14.2 Mg ha¡1) were signiWcantly higher than thatof the other treatments, aside from a combinedcompost, and mineral fertilizer application (10.6,11.6 Mg ha¡1 for yield and stover production,respectively). The productivity of only minerallyfertilized plots (F) declined rapidly after the Wrstharvest (Fig. 1b).

First harvest

Organic and mineral fertilization increased thestover production (leaves and stalks) and grainyield of rice (O. sativa L.). The highest stover pro-duction was achieved with a combination of com-post and mineral fertilizer (6.24 Mg ha¡1) whilethe greatest grain yield was harvested using CM

Fig. 1 a Cumulative yields of selected treatments and b the yields as a percentage of the yield at Wrst harvest. Mineral fertil-ization without any OM input is indicated by an arrow (means and standard errors; N = 5)

1 3

Plant Soil

(7.5 Mg ha¡1, Fig. 1). Charcoal alone and litter(L) applications alone had a minor eVect. LB andcharcoal amendments signiWcantly improvedyields compared to control and unburned litteralthough with very little grain production (0.18,0.10, 0.01, and 0.02 Mg ha¡1, respectively). Stoverand grain yield production were increased by 29and 73%, respectively if mineral fertilizer wasapplied on plots with charcoal in comparison tomineral fertilization alone (Fig. 1).

Second harvest

No plant growth was found in control, charcoal,litter, and LB plots. Plots fertilized with CM hadthe highest (P < 0.05) grain yield and stover pro-duction (2.80, 3.56 Mg ha¡1), followed by Weldsreceiving compost + mineral fertilizer (1.59,1.51 Mg ha¡1) and only compost (1.13,1.11 Mg ha¡1, respectively). While charcoal addi-tions alone did not aVect crop production, a syn-ergistic eVect occurred when both charcoal andinorganic fertilizers were applied. The grain yieldand stover production was 0.05 and 0.27 Mg ha¡1,respectively, within plots receiving inorganic fer-tilizer, 0.0 Mg ha¡1 within plots receiving charcoaland 0.46 and 0.72 Mg ha¡1, respectively, withinplots receiving both amendments (Fig. 2). Stoverproduction on plots receiving charcoal plus min-eral fertilizer (2CC + F) was found to be in thesame homogenous subgroup as compost-treatedplots. Bare mineral fertilization had the smallesteVect on grain yield and stover production. Thesynergistic eVects of charcoal in combination withfertilizers improved yields by a factor of 9 and thestover production by a factor of 2.7 in comparisonto only mineral fertilized plots (Fig. 2). In addi-tion, a visible diVerence was observed in plantgrowth during this cropping period. Plants grow-ing on 2CC + F were 24.7 and 42.0 cm tallwhereas plants growing on F-plots were only 13.0and 21.8 cm tall after 50 and 55 days, respectively(P < 0.001, N = 80).

Third and fourth harvest

Crop production with added CM still exceededthose with all other applications (yield 1.13, 0.93,stover 3.05, 1.64 Mg ha¡1, 3rd and 4th harvest,

respectively). Plots receiving CC + CO alwaysproduced less biomass than the 2CC + CO plots.This diVerence was signiWcant at the 3rd harvestand at the 4th harvest where the CC treatmentceased to produce any biomass. Mineral fertil-izer with additional charcoal or compost applica-tion again improved stover production and yield.Although the synergistic eVect of charcoal plusmineral fertilizers was less than observed at the2nd harvest, it still improved yield and stoverproduction by a factor of 1.5 and 2.0 for yieldand 1.3 and 1.4 for stover at the 3rd and 4th har-vest, respectively. Only three out of Wve plotsproduced grain yield at the 4th harvest if onlymineral fertilized. The application of micronutri-ents after the 2nd harvest (Table 1) did not showany signiWcant eVect (P > 0.05), indicating thatthe observed charcoal eVect was not caused bythe charcoal’s micronutrient content (data notshown).

Nutrient contents of plants and grains

The plants fertilized with CM had the highestnutrient contents followed by plants thatreceived compost and/or mineral fertilizer. CMsigniWcantly improved the K and P nutrition incomparison to all other treatments. Charcoalapplications did not show a signiWcant inXuenceon nutrient levels. Only the foliar K contents ofcrop residues originating from plots receiving

Fig. 2 Grain yield production at the second harvest (Feb-ruary 06, 2002). The error bars show the mean § standarderror. The Fisher’s least signiWcant diVerence (LSD) valueis plotted to scale signiWcant mean separation (P < 0.05)

1 3

Plant Soil

charcoal and mineral fertilizer were higher thanthose on plots with mineral fertilizer alone.After the 3rd harvest similar results wereobtained. Although mineral fertilization wasrepeated, plants growing on CM plots still hadsigniWcantly (P < 0.05) higher P and K contentscompared to all other treatments (Fig. 3). Plotsamended with compost or compost plus charcoalbut without mineral fertilizer revealed the high-

est foliar N levels although producing signiW-cantly less biomass.

Mainly P and N were exported due to a highercontent of these elements in grains than crop resi-dues, which remained on the plots. In contrast,crop residues generally contained approximately90% of the K and Ca. The exported K after theWrst harvest due to the removal of grains wasalmost twice as much on fertilized plots plus

Fig. 3 Foliar nutrient contents in crop residues of the 3rd harvest. The error bars show the mean § standard error. The Fish-er’s least signiWcant diVerence (LSD) value is plotted to scale signiWcant mean separation (P < 0.05)

1 3

Plant Soil

charcoal addition (2CC + F) (2.83 kg ha¡1) com-pared to only minerally fertilized plots (F)(1.42 kg ha¡1). After the second harvest almosttenfold more nutrients (K, Ca, Mg, P, and N)were exported from 2CC + F-plots (2.8 kg ha¡1 K,1.0 kg ha¡1 Ca, 0.7 kg ha¡1 Mg, 1.5 kg ha¡1 P, and7.5 kg ha¡1 N) than on F-plots (0.3 kg ha¡1 K,0.1 kg ha¡1 Ca, 0.1 kg ha¡1 Mg, 0.2 kg ha¡1 P, and0.8 kg ha¡1 N) due to their higher yields (Fig. 4).The P export after the following harvests contin-ued to be signiWcantly higher from the charcoalplots. The overall K export with grains from2CC + F-plots (9.2 kg ha¡1 K) during the fourcropping seasons was signiWcantly higher in com-parison to F-plots (4.2 kg ha¡1 K).

Soil nutrient contents (Fig. 5, Table 2)

Mineral fertilization

Soil samples after mineral fertilization were takenafter the Wrst harvest. A signiWcant enhancementthrough mineral fertilization was achieved by lim-ing. The Ca (210 mg kg¡1) and Mg (77 mg kg¡1)levels signiWcantly increased in the surface soillayer (0.1 m) of all mineral fertilized plots but thishad only little eVect on pH. Extractable Al con-tents (4.7 mg kg¡1) signiWcantly decreased in com-parison to the control plots (63.8 mg kg¡1).

The Al concentrations were further decreased(P < 0.05), if mineral fertilizer was applied on

Fig. 4 Cumulative nutrient uptake (grains and crop residues) of selected treatments on a logarithmic scale (means and stan-dard errors; N = 5)

1 3

Plant Soil

charcoal containing plots. Mineral fertilizationsigniWcantly increased BS but CEC was not sig-niWcantly altered.

After the second harvest the mineral fertilizedplots had less available P, K, Ca, and Mg (2.2,20.4, 74.6, and 43.5 mg kg¡1) than those plots thatreceived additional charcoal (4.5, 24.2, 86.6, and45.9 mg kg¡1) although the diVerences were notsigniWcant. Even though signiWcantly more nutri-ents (P, K, Ca, Mg, and N) were exported fromthe charcoal plots the available soil nutrient con-tents of the soil did not decrease in comparison toonly mineral fertilized plots. By the end of theexperiment (after the 4th harvest) 2CC + F-plotsstill revealed higher nutrient availability andhigher nutrient exports due to grain removal incomparison to F-plots (data not shown).

Organic amendments

Compost signiWcantly increased soil nutrient con-tents, mainly P, K, Ca, and Mg as the compost con-tained lime. Only a very small and not signiWcant

increase in soil nutrient contents resulted fromcharcoal, litter or LB application. LB showed sig-niWcantly increased NH4 contents.

Nutrient concentrations were several foldhigher (top 0.1 m) on CM plots (Nmin112 mg kg¡1, P 610 mg kg¡1, K 1,108 mg kg¡1, Ca2,793 mg kg¡1, Mg 373 mg kg¡1) than theamounts measured on control plots (Nmin27 mg kg¡1, P 3 mg kg¡1, K 32 mg kg¡1, Ca16 mg kg¡1, Mg 11 mg kg¡1) and enhanced theplant-available P, K Ca, and Mg contents down toa soil depth of 60 cm (Fig. 5). Also pH was signiW-cantly increased and reached almost neutral lev-els (6.6).

Soils of the CM plots continued to show thehighest nutrient contents after the Wrst harvest(Table 2). No available Al was found in the top10 cm when CM or charcoal + mineral fertilizerwere added. A signiWcant increase in CEC wasonly found in soils fertilized with CM and theacidity was further reduced.

After the 3rd harvest P, Ca, and Mg contentsremained high on CM plots but K contents

Fig. 5 The inXuence of selected treatments on plant available P, K, Ca, and Mg down to a soil depth of 0.6 m. Soil sampleswere taken after the 3rd harvest (means and standard errors; N = 5)

1 3

Plant Soil

Tab

le2

Surf

ace

(0–0

.1m

) so

il ch

emic

al p

rope

rtie

s af

ter

the W

rst

harv

est

Hom

ogen

ous

subg

roup

s ar

e in

dica

ted

by t

he s

ame

lett

er, S

tude

nt–N

ewm

an–K

euls

pos

t ho

c te

st (

P<

0.05

)

Tre

atm

ent

para

met

erC

LL

BF

CM

2CO

2CC

2CO

+F

2CC

+F

CC

+C

OC

C+

CO

+F

2CC

+C

O2C

C+

CO

+F

2CC

p(p

iece

s )2C

Cp

(pie

ces)

pH H

2O4.

50a

4.58

a,b

4.52

a,b

5.04

b5.

58c

4.87

a,b

4.50

a4.

80a,

b4.

71a,

b4.

61a,

b4.

82a,

b4.

66a,

b4.

78a,

b4.

79a,

b4.

75a,

b

pH K

Cl

3.80

a3.

83a

3.76

a4.

26a

5.53

b3.

91a

3.74

a4.

06a

4.21

a3.

83a

4.13

a3.

85a

4.03

a3.

92a

3.92

a

NO

3 (m

gkg

¡1 )

7.47

a,b

13.1

2b12

.79b

5.20

a,b

5.15

a,b

1.66

a9.

27a,

b2.

38a

2.89

a4.

51a,

b1.

88a

3.70

a4.

41a,

b6.

20a,

b7.

61a,

b

NH

4 (m

gkg

¡1 )

8.25

5.92

7.33

7.89

15.0

114

.12

13.2

512

.17

11.0

714

.34

10.6

59.

767.

177.

328.

18N

min

(mg

kg¡

1 )15

.72

19.0

320

.12

13.0

920

.16

15.7

722

.52

14.5

513

.95

18.8

512

.53

13.4

611

.57

13.5

215

.79

Tot

al N

(g

kg¡

1 )1.

821.

531.

801.

511.

851.

651.

801.

921.

651.

871.

671.

761.

82–

–T

otal

C (

gkg

¡1 )2

0.00

a,b

18.1

3a21

.78a,

b17

.28a

18.8

4a19

.51a

24.2

7a,b

23.6

7a,b

21.5

4a,b

24.5

7a,b

22.4

0a,b

21.8

2a,b

27.0

7b–

–C

/N11

.65a,

b11

.71a,

b,c

12.2

a,b,

c,d

11.2

8a10

.81a

11.8

2a,b,

c13

.32c,

d,e

12.3

a,b,

c,d,

e13

.1b,

c,d,

e13

.1b,

c,d,

e13

.1b,

c,d,

e13

.52d,

e13

.80e

––

P (

mg

kg¡

1 )1.

87a

1.03

a2.

54a

5.15

a39

2.27

b5.

91a

3.09

a8.

62a

3.41

a5.

77a

7.62

a4.

87a

4.44

a3.

16a

4.10

a

K (

mg

kg¡

1 )26

.65a

21.6

8a25

.42a

24.3

7a21

2.50

b30

.61a

31.5

4a26

.95a

27.7

2a37

.44a

31.4

1a38

.97a

36.8

7a34

.31a

39.8

1a

Ca

(mg

kg¡

1 )9.

75a

18.7

5a,b

30.2

5a,b,

c20

9.87

c21

25.5

d16

1.88

c15

.30a

268.

15c

183.

73c

110.

19c

234.

21c

84.9

5b,c

212.

69c

92.5

4a,b,

c11

0.57

c

Mg

(mg

kg¡

1 )7.

75a

8.85

a9.

34a

76.8

7b,c

115.

24d

24.7

8a9.

56a

85.1

1c79

.05b,

c22

.32a

71.1

9b,c

20.4

0a48

.62a,

b,c

38.7

9a,b

21.3

0a

Al (

mg

kg¡

1 )63

.82f,g

55.3

4e,f,g

63.3

3f,g4.

71b,

c0.

00a

17.4

6b,c,

d77

.20g

4.65

b,c

0.00

a37

.01c,

d,e,

f2.

09b

40.1

6d,e,

f9.

96b,

c,d

27.1

4b,c,

d,e

27.6

2b,c,

d,e

Aci

dity

(c

mol

c+kg

¡1 )

1.37

b,c

1.12

a,b,

c0.

68a,

b1.

27a,

b,c

0.88

a,b,

c1.

13a,

b,c

1.04

a,b,

c0.

91a,

b,c

1.08

a,b,

c1.

51c

1.18

a,b,

c0.

55a

1.23

a,b,

c0.

75a,

b,c

1.28

a,b,

c

CE

C

(cm

olc+

kg¡

1 )1.

61a

1.52

a1.

73a

2.16

a12

.55b

1.94

a1.

80a

2.45

a1.

94a

1.95

a2.

28a

1.81

a2.

11a

1.65

a1.

67a

BS

(%)

11.2

4a14

.53a

14.1

9a77

.51c

99.5

0d49

.99b

12.6

8a85

.65c,

d82

.02c

39.9

3b73

.48c

36.5

5b73

.32c

50.0

1b44

.04b

1 3

Plant Soil

decreased. CM fertilized plots had signiWcantlylower acidity (0.03 cmolc+ kg¡1) than all othertreatments.

Soil carbon dynamics

Total soil C contents increased due to OM appli-cation. The initial charcoal application signiW-cantly increased the C/N ratio. This diVerence toplots without charcoal remained detectable by theend of the experiment. Although N was fertilizedtwice, the total N content of the soil remainedrather stable and was even reduced to a greaterextent in plots not containing charcoal. Soilsreceiving charcoal lost only 11% of their initialsoil C and 13% of total N in comparison to 23%of C and 23% of N on plots without charcoalregardless of mineral fertilization. Even the con-trol and mineral fertilized plots without any OMadditions lost 25 and 22% of C, respectively. Incontrast plots receiving just charcoal or charcoalplus mineral fertilizer (without compost) lost only4 and 8% of their soil C content even at a higherstarting level.

Discussion

Charcoal amendments

Charcoal appeared to be more stable than theother tested organic amendments as well as thenative SOM. Charcoal containing soil lost only 8and 4% of their soil C content with or withoutmineral fertilization, respectively, in contrast toorganic fertilization (compost or CM) where soilC content decreased by 27%. Even the controlplots had a mean soil C loss of 25%. The resil-ience of soil C in charcoal amended plots showsthe refractory nature of charcoal (Kuhlbusch andCrutzen 1995). While the stability of charcoalleads to low-C losses, nutrient release by mineral-ization is most likely lower than from otherorganic materials. The recalcitrant nature of char-coal and low-nutrient contents (Table 1 for totalN and available P, K, Ca, and Mg contents) there-fore makes charcoal itself unlikely to be a bal-anced fertilizer. According to Duxbury et al.(1989) and Sombroek et al. (1993) it is important

to separate eVects due to OM per se (mainte-nance and improvement of water inWltration,water holding capacity, structure stability, CEC,healthy soil biological activity) from those due toits decomposition (source of nutrients).

A further assumption from Terra Pretaresearch is that slow oxidation on the edges of thearomatic backbone of charcoal forming carbox-ylic groups is responsible for both the potential offorming organo-mineral complexes and theincreased CEC (Glaser et al. 2001b). The periodof this study might not have been suYcient foroxidation. Cheng et al. (2006) demonstrated in anincubation experiment that already 4 months at30°C could signiWcantly increase the CEC. Possi-bly, the amount of applied charcoal was insuY-cient to increase the CEC as observed in ourstudy (Table 2). According to Duxbury et al.(1989) much of the negative charge is notexpressed in soils, both because it is pH depen-dent and because many of the negatively chargedsites are blocked by interactions with Al. SOMwas only eVective at increasing CEC levels abovepH 5.5, which is consistent with blockage ofexchange sites by either Al or Fe at lower pH val-ues (Lobes and Cos, 1977 cited in Duxbury et al.1989). In our study only plots fertilized with CMhad pH values higher than 5.5 and increasedCEC. Much higher charcoal amendmentsincreased pH in a pot experiment by Lehmannet al. (2003) using the same soil as studied in thisexperiment. Again the amount of charcoalapplied in our experiment might not have beenhigh enough to Wnd statistically signiWcantchanges in soil pH.

The conditions of ADE are ideal for maximumbiological N2 Wxation. About 77% of the ADEsampled showed positive incidence of Azospiril-lum spp. compared to only 10% of the Ferralsols(Silvester-Bradley et al. 1980). Charcoal providesa good habitat for the propagation of usefulmicroorganisms such as free-living nitrogen Wxingbacteria and mycorrhizal fungi (Ogawa 1994).Ogawa (1994) holds the charcoal’s weak alkalin-ity, porosity, and ability to retain water and airresponsible for the stimulation of microbes.

We found the largest biomass diVerences duringthe second harvest, which was also the driest crop-ping cycle. The mean daily precipitation was 6 mm,

1 3

Plant Soil

which is 2 mm less than during the Wrst croppingcycle. Even more important is the uneven distribu-tion of rainfall during the second growth season.More than one quarter of the 700 mm fell duringone single rain event in a few hours and up to8 days free of precipitation occurred. Both Dux-bury et al. (1989) and Sombroek et al. (1993)emphasize the importance of OM for the mainte-nance and improvement of water inWltration andwater holding capacity. Glaser et al. (2002) con-cluded from a literature survey that only sandy soilshad higher available moisture after charcoal addi-tions, clayey soils even showed decreased moisturecontents with increasing charcoal additions.

Mineral fertilization

Mineral fertilization resulted in a fast depletion ofsoil nutrients and a rapid decline in grain yieldsafter the Wrst cropping cycle (Fig. 1b). These Wnd-ings corroborate the conclusion of Zech et al.(1990) that solely a replenishment of plant-avail-able nutrients by mineral nutrient additions is notenough to maintain soil fertility in freely drainingsoils. Because SOM is often the major source ofnegative charge in tropical soils, its maintenanceis important for the adsorption of exchangeablecations (Duxbury et al. 1989). In a study con-ducted by Lehmann et al. (1999) 63% of the Napplied as (NH4)2SO4 was lost from the top 1.2 mby leaching and volatilization, but just 1% of theorganically applied N (from mulch). Soils thatreceive high-OM inputs have greater labile Cpools, higher microbial activity and higher soil Nsupplying power compared to solely mineral fer-tilized soil (Burger and Jackson 2003).

Charcoal plus mineral fertilization

The synergistic eVects if both charcoal plus min-eral fertilizers were applied, doubled the cumula-tive grain yield of four harvests, but this studyinsuYciently explains the improved crop perfor-mance. Only extractable Al concentrations werefound to be signiWcantly lower in charcoalamended and mineral fertilized soil in compari-son to mineral fertilized soil alone. Charcoalapplication increased legume production in astudy by Topoliantz et al. (2005) due to decreased

soil acidity and exchangeable Al but increased Caand Mg availability. Reactive Al and Fe surfacescan form complexes with SOM, reducing the CECbut blocking these sites will reduce the capacity ofsoil to Wx phosphate and sulphate. IncreasedSOM may also stimulate desorption of phosphateand sulphate by acting as a competing anion(Duxbury et al. 1989). Aluminum can reduce cropproduction severely (Sierra et al. 2003).

Lehmann et al. (2003) found decreasing Navailability in the Ferralsol similar to ADE, butincreased uptake of P, K, Ca, zinc (Zn), and cup-per (Cu) by plants after higher charcoal additions.The application of charcoal signiWcantly reducedleaching of applied mineral fertilizer N. Theincreased ratio of uptake to leaching due to char-coal application indicates a high eYciency ofnutrients applied with charcoal (Lehmann et al.2003). In this study, we were not able to statisti-cally prove increased availability of soil nutrientcontents but in spite of signiWcantly higher nutri-ent export by means of yield withdrawal, theavailable nutrient contents remained as high orhigher in soils receiving charcoal than only min-eral fertilized soils (Fig. 4). The withdrawal of Pand K due to grain yield removal oVset the addi-tional plant available P and K supplied by thecharcoal (Table 1) by a factor of two.

Microbial immobilization is described as animportant mechanism to retain N in those soilshighly aVected by leaching (Bengtsson et al. 2003;Burger and Jackson 2003). They conclude thatgreater C availability stimulate microbial activityresulting in greater N demand, promoting immo-bilization, and recycling of NO3. Steiner et al.(2004a) found increased microbial reproductionrates after glucose addition in soils amended withcharcoal, although not showing higher soil respi-ration rates. This indicates a low-biodegradableSOM content but suYcient soil nutrient contentsto support microbial population growth. ThisdiVerence between low-soil respiration and high-microbial population growth potential is one ofthe characteristics of ADE (Steiner et al. 2004a).

Organic amendments (chicken manure)

In the case of CM the soil nutrient levelsremained high during the entire study period

1 3

Plant Soil

(four harvests), although the quantity of nutrientsapplied as CM was not comparable to the amountapplied as mineral fertilizer (Table 1). Duxburyet al. (1989) emphasizes the importance oforganic P for plant growth in soils with reducedtotal P contents and high-phosphate adsorptioncapacity. Potassium and N are commonly quicklylost due to leaching and uptake by crops, N mightin addition be lost by denitriWcation (Brady andWeil 2001). Without additional OM applicationno diVerences could be detected in the soil K andN contents due to mineral fertilization. CM sig-niWcantly increased soil K contents until the endof the experiment although a large portion of itsoriginal soil contents was lost and exported dur-ing the four cropping cycles. Nitrogen limitationseemed to be of minor importance compared to Pand K. The initial CM application increased thesoil N contents only for a short time, whereas pro-ductivity was sustained over the four croppingcycles. Elimination of exchangeable Al (from63.8 mg kg¡1 on the control plots) was alreadyobserved after the Wrst cropping cycle and aciditycontinuously declined during the four croppingcycles (from 1.4 on control plots to 0.9 to 0.4 to0.03 at 1st, 2nd, and 4th harvest, respectively).Materechera and Mkhabela (2002) measured aliming eVectiveness of CM of 26% compared tolime. They conclude that the proton consumingability of humic materials might have reducedacidity in their study. The slow reduction of acid-ity might be explained by the steady formation oforganic material with functional groups such ascarboxyl and phenolic groups during decomposi-tion and by the low solubility of CaCO3.

Conclusion

In our experiment plant biomass productionsharply decreased within 1 year when only min-eral fertilizer was applied, but could be main-tained for a longer period of time when OM wasadded. Soils fertilized with CM lost their initiallyhigh N and K contents during the four croppingcycles but remained fertile after the 4th harvestcompared to other OM or inorganic fertilizeradditions and had signiWcantly increased cropproduction even without further input. pH

seemed to be important to ensure high productiv-ity. The charcoal additions proved to sustain fer-tility if an additional nutrient source is given.Even though signiWcantly more nutrients (P, K,Ca, Mg, and N) were exported from the charcoalplots, the available nutrient contents of the soildid not decrease in comparison to just mineralfertilized plots.

The losses of soil C were highest on CM (27%)and compost (27%) treated plots, followed bysoils amended with litter (26%), and the control(25%) whereas the charcoal amended plots lostonly 8 and 4% of their soil C content if mineralfertilized or not fertilized, respectively. The resil-ience of soil C in charcoal amended plots showedthe refractory nature of charcoal. A combinationof charcoal and CM might mimic the favorableproperties of Terra Preta best.

We propose that charcoal applications canimprove soil chemical, biological, and physicalproperties in various ways causing the observedsigniWcant increase in crop production. It is diYcultto isolate single mechanisms, which were responsi-ble for this increase. Further research is needed todiscern the mechanisms of fertility enhancementand optimize charcoal use for soil amelioration.

Acknowledgments The research was conducted withinSHIFT ENV 45, a German–Brazilian cooperation andWnanced by the Bundesministerium für Bildung und Fors-chung (BMBF), Germany and Conselho Nacional de De-senvolvimento CientíWco e Tecnológico (CNPq), Brazil(BMBF No. 0339641 5A, CNPq 690003/98–6). A Wnancialcontribution was given by the doctoral scholarship programof the Austrian Academy of Sciences. We are grateful forthe Weldworkers’ help particularly Luciana Ferreira da Sil-va and Franzisco Aragão Simão and the laboratory techni-cian Marcia Pereira de Almeida. Critical comments andhelp were given by Prof. Dr. William I. Woods, Dr. BrunoGlaser, Jago Birk, and Heiko Grosch.

References

Baldock JA, Smernik RJ (2002) Chemical composition andbioavailability of thermally altered Pinus resinosa(Red pine) wood. Org Geochem 33:1093–1109

Bengtsson G, Bengtson P, Månsson KF (2003) Gross nitro-gen mineralization-, immobilization-, and nitriWcationrates as a function of soil C/N ratio and microbial activ-ity. Soil Biol Biochem 35:143–154

Box GEP, Cox DR (1964) An analysis of transformations.J R Stat Soc Ser B.26:211–252

1 3

Plant Soil

Brady NC, Weil RR (2001) The nature and properties ofsoils. Prentice Hall, Englewood CliVs, NJ

Burger M, Jackson LE (2003) Microbial immobilization ofammonium and nitrate in relation to ammoniWcationand nitriWcation rates in organic and conventionalcropping systems. Soil Biol Biochem 35:29–36

Cheng CH, Lehmann J, Thies JE, Burton SD, EngelhardMH (2006) Oxidation of black carbon by biotic andabiotic processes. Org Geochem 37:1477–1488

Claessen MEC, Barreto WdO, Paula JLd, Duarte MN(1997) Manual de metodos de analise de solo. EMB-RAPA, Rio de Janeiro, p 212

Correia FWS, Lieberei R (1998) Agroclimatological infor-mation about the experimental Weld of the SHIFT-ar-ea, ENV 23, 42, 45, 52. In: Lieberei R, Biachi H, VossK (eds) Third SHIFT workshop, Manaus, March 1998,1998. BMBF, Germany, pp 389–396

Denevan WM (1996) A bluV model of riverine settlementin prehistoric Amazonia. Ann Assoc Am Geogr86(4):654–681

Duxbury JM, Smith MS, Doran JW, Jordan C, Szott L,Vance E (1989) Soil organic matter as a source and asink of plant nutrients. In: Coleman DC, Oades JM,Uehara G (eds) Dynamics of soil organic matter intropical ecosystems. University of Hawaii Press,Honolulu, pp 33–67

Fageria NK (1998) Manejo da calagem e adubacao do ar-roz. Embrapa Arroz e Feijao, Santo Antonio de Goias,GO, pp 67–79

FAO (1990) Soil map of the world, revised legend. FAO,Rome, Italy

Fearnside PM (1997) Greenhouse gases from deforestationin Brazilian Amazinia: net committed emissions. ClimChang 35:321–360

Fearnside PM, Lima PM, Graça A, Rodrigues FJA (2001)Burning of Amazonian rainforest: burning eYciencyand charcoal formation in forest cleared for cattle pas-ture near Manaus, Brazil. For Ecol Manage 146:115–128

Giardina CP, Sanford RL, Dockersmith IC, Jaramillo VJ(2000) The eVects of slash burning on ecosystem nutri-ents during the land preparation phase of shifting cul-tivation. Plant Soil 220:247–260

Glaser B, Guggenberger G, Haumaier L, Zech W (2001a)Persistence of soil organic matter in archaeologicalsoils (Terra Preta) of the Brazilian Amazon region. In:Rees RM, Ball BC, Campbell CD, Watson CA (eds)Sustainable management of soil organic matter. CABIPublishing, Wallingford, pp 190–194

Glaser B, Haumaier L, Guggenberger G, Zech W (2001b)The “terra preta” phenomenon: a model for sustain-able agriculture in the humid tropics. Naturwissens-chaften 88:37–41

Glaser B, Lehmann J, Zech W (2002) Ameliorating physi-cal and chemical properties of highly weathered soilsin the tropics with charcoal—a review. Biol Fertil Soils35:219–230

Goldammer JG (1993) Historical biogeography of Wre:tropical and subtropical. In: Crutzen PJ, GoldammerJG (eds) Fire in the environment: the ecological

atmospheric, and climatic importance of vegetationWres. Wiley, New York, pp 297–314

Hölscher D, Ludwig B, Möller RF, Fölster H (1997a) Dy-namic of soil chemical parameters in shifting agricul-ture in the Eastern Amazon. Agric Ecosyst Environ66:153–163

Hölscher D, Möller RF, Denich M, Fölster H (1997b)Nutrient input-output budget of shifting agriculture inEastern Amazonia. Nutr Cycl Agroecosys 47:49–57

Hughes RF, KauVman JB, Cummings DL (2000) Fire in theBrazilian Amazon 3. Dynamics of biomass, C, andnutrient pools in regenerating forests. Oecologia124:574–588

Kleinman PJA, Pimentel D, Bryant RB (1995) The ecolog-ical sustainability of slash-and-burn agriculture. AgricEcosyst Environ 52:235–249

Kuhlbusch TA, Lobert JM, Crutzen PJ, Warneck P (1991)Molecular nitrogen emissions from denitriWcation dur-ing biomass burning. Nature 351:135–137

Kuhlbusch TAJ, Crutzen PJ (1995) Toward a global esti-mate of black carbon in residues of vegetation Wresrepresenting a sink of atmospheric CO2 and a sourceof O2. Global Biogeochem Cycles 9:491–501

Lehmann J, Campos CV, Macedo JLV, German L (2004)Sequential fractionation and sources of P in Amazo-nian dark earths. In: Glaser B, Woods WI (eds) Ama-zonian dark earths: exploration in time and space.Springer Verlag, Heidelberg, pp 113–123

Lehmann J, da Silva JP Jr, Rondon M, Cravo MdS, Green-wood J, Nehls T, Steiner C, Glaser B (2002) Slash andchar—a feasible alternative for soil fertility manage-ment in the central Amazon? In: Proceedings of the17th world congress of soil science, Bangkok, Thai-land, 14–21. 08. 2002, Ed T I U o S Sciences, pp 449-1–449-12

Lehmann J, da Silva JP Jr, Steiner C, Nehls T, Zech W, Gla-ser B (2003) Nutrient availability and leaching in anarchaeological Anthrosol and a Ferralsol of the Cen-tral Amazon basin: fertilizer, manure and charcoalamendments. Plant Soil 249:343–357

Lehmann J, Feilner T, Gebauer G, Zech W (1999) Nitro-gen uptake of sorghum (Sorghum bicolor L.) from treemulch and mineral fertilizer under high leaching con-ditions estimated by nitrogen¡15 enrichment. BiolFertil Soils 30:90–95

Lehmann J, Gaunt J, Rondon M (2006) Bio-char sequestra-tion in terrestrial ecosystems—a review. Mitig AdaptStrateg Global Change 11:403–427

Lewis J, Vosti S, Witcover J, Ericksen PJ, Guevara R, To-mich T (2002) Alternatives to slash-and burn in Brazil.World Agroforestry Centre (ICRAF), Nairobi, pp 1–100

Liang B, Lehmann J, Solomon D, Grossman J, O’Neill B,Skjemstad JO, Thies J, Luizão FJ, Petersen J, NevesEG (2006) Black carbon increases cation exchangecapacity in soils. Soil Sci Soc Am J 70:1719–1730

Lima HN, Schaefer CER, Mello JWV, Gilkes RJ, Ker JC(2002) Pedogenesis and pre-Colombian land use of“Terra Preta Anthrosols” (“Indian black earth”) ofWestern Amazonia. Geoderma 110:1–17

1 3

Plant Soil

Materechera SA, Mkhabela TS (2002) The eVectiveness oflime, chicken manure and leaf litter ash in ameliorat-ing acidity in a soil previously under black wattle (Aca-cia mearnsii) plantation. Bioresour Technol 859–816

McLean EO (1965) Aluminium. In: Black CA (ed) Meth-ods of soil analysis part 2: chemical and microbiologi-cal properties. Am. Soc. Agron, Madison, pp 986–994

Mehlich A (1984) Mehlich-3 Soil test extractant: a modiW-cation of Mehlich-2 extractant. Commun Soil Sci PlantAnal 15:1409–1416

Ogawa M (1994) Symbiosis of people and nature in the tro-pics. Farming Jp 28(5):10–30

Olsen SR, Sommers LE (1982) Phosphorus. In: Page AL,Miller RH, Keeney DR (eds) Methods of soil analyses:part 2 chemical and microbiological properties. Am.Soc. Agron, Wisconsin, pp 403–430

Renck A, Lehmann J (2004) Rapid water Xow and trans-port of inorganic and organic nitrogen in a highlyaggregated tropical soil. Soil Sci 169:330–341

Sanchez P (1976) Properties and management of soils inthe tropics. Wiley, New York

Sierra J, Noël C, Dufour L, Ozier-Lafontaine H, WelckerC, Desfontaines L (2003) Mineral nutrition andgrowth of tropical maize as aVected by soil acidity.Plant Soil 252:215–226

Silva-Forsberg MC, Fearnside PM (1995) Agriculturalmanagement of caboclos of the Xingu river: a startingpoint for sustaining populations in degraded areas inthe Brazilian Amazon. In: Parrotta JA, Kanashiro M(eds) Management and rehabilitation of degradedlands and secondary forests in Amazonia. Interna-tional Institute of Tropical Forestry, Rio Piedras, pp90–95

Silvester-Bradley R, Oliveira LAD, Filho JADP, JohnSTV (1980) Nodulation of legumes, nitrogenaseactivity of roots and occurrance of nitrogen-Wxing

Azospirillum spp. in representative soils of centralAmazonia. Agro-Ecosyst 6:249–266

Sombroek WG, Nachtergaele FO, Hebel A (1993)Amounts, dynamics and sequestering of carbon intropical and subtropical soils. Ambio 22:417–426

Steiner , Teixeira WG, Lehmann J, Zech W (2004a) Micro-bial response to charcoal amendments of highlyweathered soils and amazonian dark earths in centralAmazonia—preliminary results. In: Glaser B, WoodsWI (eds) Amazonian dark earths: explorations in spaceand time. Springer Verlag, Heidelberg, pp 195–212

Steiner C, Teixeira WG, Zech W (2004b) Slash and char: analternative to slash and burn practiced in the Amazonbasin. In: Glaser B, Woods WI (eds) Amazonian darkearths: explorations in space and time. Springer Ver-lag, Heidelberg, pp 183–193

Tiessen H, Cuevas E, Chacon P (1994) The role of soil or-ganic matter in sustaining soil fertility. Nature371:783–785

Topoliantz S, Ponge J-F, Ballof S (2005) Manioc peel andcharcoal: a potential organic amendment for sustain-able soil fertility in the tropics. Biol Fertil Soils 41:15–21

Walinga I (1995) Digestion in tubes with H2SO4–salicylicacid-H2O2 and selenium and determination of Ca, K,Mg, N, Na, P, Zn. In: Walinga I, Lee JJvd, Houba VJG,Vark Wv, Novozamsky I (eds) Plant analysis manual.Kluwer Academic Publishers, Dordrecht, The Nether-lands

Zech W, Haumaier L, HempXing R (1990) Ecological as-pects of soil organic matter in the tropical land use. In:McCarthy P, Clapp C, Malcolm RL, Bloom PR (eds)Humic substances in soil and crop sciences; selectedreadings. American Society of Agronomy and Soil Sci-ence Society of America, Madison, pp 187–202

1 3