Learning Conservation Agriculture the Innovation Systems way

Grain-SA Smallholder Farmer Innovation Programme

Erna Kruger, Ngcobo P, Dlamini M and Smith H

CA-Farmer Innovation ProgrammeKey objectives and activities

Farmer-centred Innovation

System

Awareness raising and Access to Information

Incentives and Market Based Mechanisms

On-farm, farmer-led Research

Education and Training

Farmers days, symposiums, cross visits, conferences,

popular articles

Subsidies, Village Saving and Loan

Associations, farmer centres, group based access to equipment

and infrastructure

Farmer experimentation; intercropping, crop

rotation, cover crops, livestock integration.

Learning groups; practical

demonstrations, workshops, field

assessments

Stakeholder interaction, partnerships, horizontal

and vertical scaling

CA Farmer led Trial summaries

Midlands Bergville EC, SKZN

Season 2017 2013 2014 2015 2016 2017 2013 2014 2015 2016 2017

No of villages 6 3 9 11 17 18 4 10 8 8 13

No of trial participants 42 28 83 73 212 259 23 16 43 54 93

Area planted (trials) - ha 1,36 2,8 7,2 5,9 13,5 17,4 0,36 0,3 0,37 1,18 3,58

Average yield maize (t/ha) 2,04 3,74 3,63 4,12 5,03 5,7 0,95 0,7 1,37 2,52 2,17

Min and max yield maize (t/ha) 0,4-7,1 2-4,3 1-6,7 0,6-7,4 0,3-11,7 0,5-12,2 0,3-1,7 0,3-1,8 0,5-4,4 1,1-5,2 0,2-6,7

Average yield beans (t/ha) 0,62 1,24 0,26 0,79 1,05 1,22 1,26 0,34 0,69 1,28 0,35

Trial summaries over 5 seasons; Bergville,SKZN and EC

• For CA plots the pH is higher on average and acid saturation lower than on control plots

• The required P has reduced on CA plots

• And % Org C and % N increased significantly compared to control plots

• Savings of around R400/ha made on inorganic N in three seasons

• C:N ratios in the soil decrease over time for the CA plots

• Soil health scores are higher for CA plots than control plots

The CA system and effect on soil fertility and soil health

Intercropping with legumes (beans and cowpeas) and use

of cover crops increase soil fertility

and soil health FASTER than

monocropping

Increased % Organic C and %

N under CA

• Visual and quantitative indicators

• Visual Soil Assessments: soil cover, soil structure, run-off, crusting, earthworms, root size, soil porosity, soil texture

• Measurements: infiltration, run-off plots, weather stations

• Soil health analysis

Soil health; methods

• Colour and texture –more an indicator of soil type – so doesn’t change too much with tillage options

• Soil depth- no distinction between CA and Conv – how to measure?

• Soil cover- ? Which version?

• Infiltration – how to measure

May-18

Visual soil Indicators Stulwane Eqeleni Ezibomvini

NAME OF PARTICIPANT

K D

lad

la(T

)K

Dla

dla

(C)

D H

lon

gwan

e(T)

D H

lon

gwan

e(C

)T

Dla

min

i (T)

T D

lam

ini (

C)

M D

lad

la(T

)M

Dla

dla

(C)

C B

uth

elez

i(T)

C B

uth

elez

i(C

)P

Sth

ebe(

T)

P S

theb

e(C

)

ThZi

kod

e (T

)

ThZi

kod

e (C

)

T Zi

kod

e (T

)T

Ziko

de

(C)

T M

abas

o (

T)

T M

abas

o (

C)

N Z

iko

de

(T)

N Z

iko

de

(C)

S H

lats

hw

ayo

(T)

S H

lats

hw

ayo

(C

)

C H

lon

gwan

e (T

)

C H

lon

gwan

e (C

)

P H

lon

gwan

e (T

)

P H

lon

gwan

e (C

)

SOIL TEXTURE (X3) 6 6 6 6 6 6 6 3 6 6 3 6 3 6 3 6 3 3 3 3 3 3 6 6 6 3

SOIL STRUCTURE( AGGR) (x3) 6 6 6 6 6 3 6 3 6 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

SOIL POROSITY (x3) 6 3 3 3 6 3 6 0 3 3 3 3 6 6 3 3 3 6 3 3 3 0 3 3 6 3

SOIL COLOUR (x2) 2 2 2 2 2 2 4 4 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 4 2

NO. OF SOIL MOTTLES AND COLOUR (x1) 1 1 1 1 1 1 2 2 2 1 1 1 1 1 1 1 1 0 0 0 1 2 1 1 1 0

EARTHWORM COUNTS (x2) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

SOIL COVER (RESIDUE) (x2) 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0

SOIL DEPTH( CM) (x2) 4 4 4 4 4 4 2 4 4 4 4 4 4 4 4 4 4 2 2 2 2 4 2 2 2 2

RUN-OFF (x2) 4 4 0 2 2 4 0 0 2 2 2 2 2 2 2 4 2 2 2 0 2 2 0 2 2 2

INFILTRATION (x2) 4 4 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

TOTALS 33 3024 26 31 25 28 18 27 23 20 23 23 26 20 25 20 20 17 15 18 18 19 21 27 17

VSAs

• Should be >30%

• 15-30% still considered CA

• <15% a problem BUT little cover in our system – we want to assess increase in cover –

• Cannot assess this before spraying – complicates things too much

• During season: IMPORTANT THT WEEDING DOES NOT REMOVE COVER

• Option 3: A range of different percentages with more categories to be more specific:

• 0 =0-15% cover• 1=16-30% cover• 2=31-45% cover• 3=46-60% cover• 4= 61-90% cover• 5= > 90% cover

Soil cover

• When in the season should we look at soil cover?

• How to relate that to canopy cover?

Soil cover

Soils

O,1,2 CONTROL (CA yes or no) TRIAL

Soil colour (light, ave, dark) (uniformity- specks)– (x3)

Soil structure (aggregates) – (x4)

Porosity (Clods, pores, organic matter)- (x5)

Soil surface (run-off, texture, crusting) – (x3)

Soil cover: 0-15%; 15-30%, >30% (x3)

Bulk density

Bulk Density - VSA

In the pit wall; using a pocket knife

- Knife easily pushed in, soil

disintegrates; 1.4-1.6g/cm3 (2)

- Knife pushed in for about half the

length of the blade (1,6-1,8g/cm3 (1)

- Only knife tip can be pushed in

>2g/cm3 (0)

- So, is it worth doing the VSA

version??? How difficult is the lab

version

- When should the sample be taken?

Vill

age

Pe

rio

du

nd

ue

CA

(yrs

)

Nam

ean

d

Surn

ame

Co

ntr

olC

T

Co

ntr

olC

A

M M+B

M+C

P

SCC

Ave

rage

Ezibomvini 4 Phumelele Hlongwane 1,30 1,36 1,38 1,33 1,38 1,28 1,34

Eqeleni 5 Ntombakhe Zikode 1,35 1,49 1,37 1,32 1,38

Thamela 1 Mkhuliseni Zwane 1,14 1,08 1,09 1,07 1,10

Average bulk density 1,27

Rainfall data

Averages for Ezibomvini, Eqeleni,

Stulwane, Thamela and Ndunwana

Dec Jan Feb March April May

Monthly rainfall (mm) 185 72,25 169,2 114,7 17 5

Monthly rainfall – weather station 92,8 93,2 89,6 148,8 24,8 5,2

Monthly rainfall Ezibomvini 29,5 94 11,2 114,7 17

Mean (mm) per rainfall event 7,9 5,8 8,2 7,6 2,1 0,4

Max (mm) per rainfall event 60 30 30 20 1 3,5

• Generally the rain gauge data has under- estimated the rainfall for each month.

• There are reasonably significant differences between the villages- but we don’t know whether it is real or due to haphazard recording

Run-off data

• Run-off data for Phumelele only and only for two months…

• % conversion should be per rainfall event but these were not correlated.

• Nthombakhe’s run-off plots only recorded for 1 week at end Feb…

Rainfall records Run-off plots litresDate Maize+Beans Maize only Maize+CP Summer CC Control

Feb-18 169 35,61 18,53 37,05 35 57,59Mar-18 114,7 7,5 1,52 8,9 7,7 23,32

Rainfall records Percentage rain converted to runoffFeb-18 169 21% 11% 22% 21% 34%Mar-18 114,7 7% 1% 8% 7% 20%

Infiltration

• Infiltration in CA trials higher for 5 of 11 participants

• Unclear whether controls are also CA (and CA how – mono-cropped?)

• Difficult to say anything

• Continue? And if so how? – double ring, single ring???

Village Name and Surname Yrs under

CA

infiltration rate

(mm/hr) control

infiltration rate

(mm/hr) trial

Stulwane Khulekani Dladla 5 587,4 531,4

Dlezakhe Hlongwane 5 226,2 423,8

Thulani Dlamini 5 422,7 450,0

Makhethi Dladla 5 226,6 587,4

Pasazile Sithebe 5 544,4 478,3

Cuphile Buthelezi 5 429,2 637,7

Ezibomvini Phumelele Hlongwane 4 455,5 282,5

Cabangile Hlongwane 3 183,0 133,9

Eqeleni Tholwephi Mabaso 5 218,8 250,8

Tombi Zikode 5 618,1 177,1

Smephi Hlatshwayo 5 434,8 218,8

• WEOC – sugars from root exudates, plus organic matter degradation

• CO2 – microbial activity/respiration

• WEON –Atmospheric N2 sequestration from free living N fixers, plus SOM degradation

• C/N – Balance between WEOC and WEON

• MAC% - efficiency of cycling of WEOC (WEOC/CO2-C)

Soil health(SH) scores CO2/10+WEOC/50+WEON/10 =SH

score

Joining soil science and ecology into a new

science of soil health

• Developed by Rick Haney – to accommodate for and include the organic fractions of nutrients in soil sample analysis

• Recognising that soil health is a dynamic process of cycling of nutrients, microbial activity and degradation of organic matter

• And the plant roots are active participants in the cycling providing carbon sugars as root exudates to supply microbes with food

• C:N ratio is determined by soil chemical properties and micro organisms present in the soil.

• The lower this ratio is, the more organisms are active and the more available the food is to the plants. Good C:N ratios for plant growth are <15:1.

• You can have a low or optimum C:N (WEOC/WEON) within a range of values of available organic carbon in the soil. (WEOC)

• If this value is low, it will reflect in the C02evolution, which will also be low. So less organic carbon means less respiration from microorganisms, but again this relationship is unlikely to be linear.

• The Microbially Active Carbon (MAC = WEOC / ppm CO2) content is an expression of this relationship. If the percentage MAC is low, it means that nutrient cycling will also be low. One needs a %MAC of at least 20% for efficient nutrient cycling.

• The SH score ranges between 0-50. the scale is generally 0-3; 3-7; 7-15; 15-25; 25-50

Test results

ppmCO2-C

N-Mineralisation Potential Biomass

>100 High-N potential soil. Likely

sufficient N for most crops

Soil very well supplied with organic

matter. Biomass>2500ppm

61-100 Moderately-high. This soil has

limited need for supplemental N

Ideal state of biological activity and

adequate organic matter

31-60 Moderate. Supplemental N

required

Requires new applications of stable

organic matter. Biomass<1,200ppm

6-30 Moderate-low. Will not provide

sufficient N for most crops

Low in organic structure and microbial

activity. Biomass<500ppm

0-5 Little biological activity; requires

significant fertilization

Very inactive soil. Biomass<100ppm.

Consider long-term care

What the values mean CONVENTIONAL SYSTEM: Mostly

decomposer fungi – small hyphal

networks, NB for soil fertility, minor

role in carbon storage

CA SYTEM: Mostly Mycorrhizal fungi

– large hyphal networks, major role

in carbon storage

Mycorrhizal fungi get their

energy in a liquid form, as soluble

carbon directly from actively

growing plants. They access and

transport water - plus nutrients

such as phosphorus, nitrogen and

zinc - in exchange for carbon from

plants.

Soluble carbon is also channelled

into soil aggregates via the

hyphae of mycorrhizal fungi and

can undergo humification, a

process in which simple sugars

are made up into highly complex

carbon polymers.

Below: Mycorrhizal fungi grow very closely

associated with plant roots and create

networks of filaments (hyphae) within the soil)

Reserve Organic, 0

Release from Orga, 19.40

NO3, 1.10

NH4, 10.40Total Inorganic, 11.5

Distribution of the Nitrogen components ppm

Reserve Organic, 7.4

Release from Orga, 1.00

NO3, 83.20

NH4, 2.80

Total Inorganic, 86

Distribution of the Nitrogen components ppm

BIOLOGICAL ANALYSES

Sample #

SOLVITA CO2 Burst

WATER EXTRACTC/N

Soil Health

Calculation (Index)

CommentCO2 - C,

ppm COrganic C

ppm COrganic N

ppm NENZV 206.1 379 19.4 19.5 24.2

Excellent

BIOLOGICAL ANALYSES

Sample #

SOLVITA CO2 Burst

WATER EXTRACTC/N

Soil Health

Calculation (Index)

CommentCO2 - C,

ppm COrganic C

ppm COrganic N

ppm N7.1 129 8.4 15.4 2.6

Soil Organic Matter % 6.2

Microbial Active C (MAC)%

54.4Soil Organic Matter

% 1.3

Microbial Active C (MAC)%

5.5

Comparing the nitrogen profile of natural “veld” with an intensively chemically farmed plot.

Comparing the nitrogen profile of natural “veld” with CA diverse cropped plot; Bergville, 2016/17.

BIOLOGICAL ANALYSES

Sample #

SOLVITA CO2 Burst

WATER EXTRACTC/N

Soil Health

Calculation (Index)

CommentCO2 - C,

ppm COrganic C

ppm COrganic N

ppm NEPHV 81,6 326 18,4 17,7 16,5

Excellent

Reserve Organic, 0

Release from Orga, 18.40

NO3, 0.30

NH4, 2.70Total Inorganic, 3

Distribution of the Nitrogen components ppm: Veld (P Hlongwane)

Soil Organic Matter % 4

Microbial Active C (MAC) % 25

Reserve Organic, 4.2

Release from Orga, 21.20

NO3, 12.90

NH4, 4.30

Total Inorganic, 17.2

Distribution of the Nitrogen components ppm; Maize + Cowpea intercrop (P Hlongwane)

BIOLOGICAL ANALYSES

Sample #

SOLVITA CO2 Burst

WATER EXTRACTC/N

Soil Health

Calculation (Index)

CommentCO2 - C,

ppm COrganic C

ppm COrganic N

ppm NEPHMCP 61,8 296 25,4 11,7 14,6

Excellent

Soil Organic Matter % 3,3

Microbial Active C (MAC) % 20,9

Comparing the nitrogen profile of Mono-cropped Maize with CA diverse cropped plot; Bergville, 2016/17.

Reserve Organic, 4.2

Release from Orga, 21.20

NO3, 12.90

NH4, 4.30

Total Inorganic, 17.2

Distribution of the Nitrogen components ppm; Maize + Cowpea intercrop (P Hlongwane)

BIOLOGICAL ANALYSES

Sample #

SOLVITA CO2 Burst

WATER EXTRACTC/N

Soil Health

Calculation (Index)

CommentCO2 - C,

ppm COrganic C

ppm COrganic N

ppm NEPHMCP 61,8 296 25,4 11,7 14,6

Excellent

Soil Organic Matter % 3,3

Microbial Active C (MAC) % 20,9

Reserve Organic, 1.5

Release from Orga, 21.70 NO3, 16.10

NH4, 2.30

Total Inorganic, 18.4

Distribution of the Nitrogen components ppm; CA Maize control (P Hlongwane)

BIOLOGICAL ANALYSES

Sample #

SOLVITA CO2 Burst

WATER EXTRACTC/N

Soil Health

Calculation (Index)

CommentCO2 - C,

ppm COrganic C

ppm COrganic N

ppm N

EPHC 59,6 254 23,2 10,9 13,4Excellent

Soil Organic Matter % 3

Microbial Active C (MAC) % 23,5

• 4-5 years: Reduced need for herbicide - no spraying on trial plots this season

• Increased organic matter, reduced fertilizer requirements -No basal fertilizer applied- only top dressing

• Reduced runoff

• Increased yields and diversity

Bergville: Case study Mphumelele Hlongwane- Ezibomvini

t/ha 2016 2017

Maize (Control)-CA 7,8 9,7

Maize Trial CA - combined 6,93 8,3

Beans 0,25 1,81

Sunflower 0,3 0,8

• EXPERIMENTS: Inter- cropping, crop rotation, legumes, SCC, WCC

• Runoff plots: CA (1,1mm/event) vs Conventional control (3,1mm/event)

• Infiltration: CA (247mm/hr) vs Control (50mm/hr)

• Soil health 2016, 2017: • Build-up of organic soil carbon and nitrogen in

the trial, with more microbially available carbon and thus a much higher soil health score.

• Lower C:N ratios in CA plots (CCs and legumes

• AGGREGATE STABILITY: CA (43%-55%) CA Control (33%) - higher aggregate stability for the plots with crop diversification- highest for inclusion of SCC mixes and Lab-Lab

• % OM: CA average (3 ,47%), Veld Baseline (2,5%) – accumulation of organic matter in CA plots- now higher than veld baseline benchmark

(10)

M + B

(5)

LL

Control plot

(8)

M + B

(6)

M +LL

(3) M + SCC

+WCC

Contro

l plot

(9)

M + CP

(7)

M + CP

(4)

M + B

(2)Sunn

hemp,

millet and

sunflower

(1)

M + B

Legend: M – Maize; B – Beans; CP – Cowpea; LL – Lab Lab; SCC – summer cover crop WCC –

winter cover crop

Bergville: Case study-continued

4 Participants over 3 seasons-crop diversity (intercropping and cover crops) and crop rotation to a lesser extent. AVERAGE FOR ALL TRIAL PLOTS SHOWN

• The Organic Carbon content has INCREASED for all 4 participants

• The Organic Nitrogen content has INCREASED for all 4 participants

• C:N ratios have decreased for Phumelele Hlongwane only – as she has most coherently implemented the diverse cropping and crop rotation process (including legumes).

• Soil health scores have increased significantly between 2016 to 2017

Soil health Test results: 3 seasons 2015-2017

Average values of different cropping options across three seasons for 9 participants • If one compares

single crops (maize only) to the mixed crop options (intercrops and cover crops) then

- Organic Carbon and Organic Nitrogen are HIGHER

- C:N ratios are LOWER - Soil health scores are

HIGHER for all the mixed cropping options - except for maize and bean intercrop option

Soil health Test results: Different Cropping options 2015-2017

• 3-4 years: Animal drawn traction – larger plot – very sandy , extremely low organic matter (0,6%).

• Germination, growth and yields have increased – albeit slowly; Maize 0,78-1,4t/ha, beans stable at 0,16t/ha

• Increased diversity – legumes and cover crops

• Reduced run-off and erosion

• Reduced weeds

• Increased soil health

Eastern Cape: Matatiele Case study Tsoloane Mapheele– Khutsong- Matatiele

• EXPERIMENTS: single crops, Inter- cropping, legumes, SCC, WCC, Lucerne

• Soil health 2016-2017: • Reduction in C:N ratio and increase in organic N in his intercropped

trial plots as compared to his Maize only control plot, indicating an increase in soil health.

• Soil health scores for the trial plots are higher than the control plots, but still below average given the extremely sandy and infertile soils he is working on.

Case study continued

Cont (M) Trial veld Trial

2016 2017

Average of %OM 0.6 0.7 1 0.6

Average of CO2 - C, ppm C 16.8 14.6 19.2 14.6

Average of Organic C ppm C 118 116 145 73

Average of Organic N ppm N 7.5 8.5 12.4 12.1

Average of C:N ratio 15.7 13.6 11.7 6.0

Average of Soil health Calculation 3.0 3.1 4.3 4.0

0

20

40

60

80

100

120

140

160

Ind

icat

ors

Soil Health Indicators Tsoalone Mapheele; 2016-2017

• In Matatiele (EC) soils are generally poor and veld benchmark values tend to be lower than the CA trial values

• The increase in available nitrogen (2016-2017) amounts to an average Rand value of about R150 more than the veld samples. This indicates a potential saving on bought fertilizer of around 14%

• In addition Immediate release N has increased substantially over the veld bench mark values and is higher for 2017; indicating a cropping pattern under the CA trials that builds organic nitrogen in the soil.

• In Bergville the picture is a bit different. Soils are good and veld benchmarks are high – excellent

• Here only the trials in Ezibomvini have higher total Nitrogen amounts than the veld. The Rand value of available organic N here is R64 more than the veld benchmark, indicating a potential saving on bought fertilizer of around 6%.

Nutrient cycling- Nitrogen; Comparison EC and Bgvl

2016 2017 2016 2016 2017 2016 2016 2017 2016

Trial Veld Trial Veld Trial Veld

Khutsong Nkau Sehutlong

Average of N(kg/ha) Total 215 84 151 308 422 280 325 378 280

Average of R value of OrgN

103 243 175 263 353 129 162 243 84

Average of N Immediaterelease

6 22 10 16 31 8 10 22 5

0

50

100

150

200

250

300

350

400

450

Matatiele; Nitrogen 2016- 2017

• Intercropping and use of cover crops is very important for building soil fertility and soil health

• Crop rotation aids in stabilising high soil health scores over time

• The more crops you use and rotate the better

• Having legumes in the mix speeds up the process

Soil Health SummaryCrop diversity is

crucial

Crop rotation in combination with

crop diversity supports this

process

Lab-Lab and SCC provide for very high organic C and N values

Lower C:N ratios are found in crop mixes

that contain legumes – cowpeas, Lab-Lab