Dr. Will Osterholz - What's More Important For Water Quality: Recent P Applications Or Legacy Soil Test P?

What’s more important for
water quality, recently
applied P or legacy soil P?
William Osterholz1, Vinayak Shedekar2, Kevin King1
1USDA-ARS Soil Drainage Research Unit, Columbus OH
2The Ohio State University - Department of Food, Agricultural, and
Biological Engineering
SOIL DRAINAGE RESEARCH UNIT

Sources of agricultural P runoff
1. Newly applied P: incidental P, episodic, direct
fertilizer, event-specific losses
• P from recent P applications
• Short term spikes – weeks to months
2. Legacy soil P: chronic losses
• P from historic P applications now stored in soil pools
• Persistent losses over long period of time

• Several pounds of new P per acre can be lost within
weeks of fertilizer application via surface runoff
• Legacy P is a significant source of DRP from NW
Ohio cropland (King et al. 2017)
• Relative contributions of new and legacy sources to
DRP and TP loads are unknown
• How important is fertilizer management?
New P vs. Legacy P

USDA-ARS Edge-of-Field network
• 40 paired fields
• Established 2012-2017
• Examine practices including
fertilizer management, cover
crops, crop rotation, gypsum
application, drainage water
management
Columbus
Cleveland
Toledo
Percent of county with
subsurface drainage
0-5
5-15
15-30
30-50
50-80
Major city
USDA-ARS edge-of-field site
Cincinnati
Dayton
Williams et al. 2016

USDA-ARS Edge-of-Field network
Surface Runoff (35) Tile Drainage (43)
• Automated water samplers and flow meters

EOF findings: P losses from WLEB
cropland
• P concentrations greater in surface runoff than tile
drainage
• But ~80% of runoff volume occurs through tile
drains
• Surface runoff: ~35% of P loading
• Tile drainage: ~65% of P loading
• Dissolved reactive P (DRP) and total P (TP) are
important in both pathways

How can we distinguish new P from
legacy P?
• Following a P application (3 month risk period for new
P loss)
1. Predicted the daily “expected” legacy P concentration using
data measured during the legacy periods (weighted
regression)
2. Calculated the expected legacy P load
3. Subtracted the expected P load from the measured load;
difference was assumed to be “New P”
• New P load = measured P load – expected legacy P load
• Legacy P load = expected legacy P load + full P load
from legacy periods

New P load = 0.117 - 0.011 = 0.106 lb DRP
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
5/25 6/8 6/22 7/6 7/20 8/3 8/17
CumulativeDRPLoad(lb/acre)
Example New P Load Calculation
Observed DRP load = 0.117 lb DRP
Expected legacy DRP load = 0.011 lb DRP
Papplication6/1

0
0.5
1
1.5
2
2.5
10
12
13
18
19
20
21
22
23
24
25
26
27
28
29
30
31
42
43
44
45
46
47
52
53
54
55
56
57
58
59
14A
14B
15A
15B
AnnualDRPLoad(lb/acre)
Field Number
Old DRP New DRP
Legacy
DRP
Legacy DRP Loads
(Surface + Tile)
Average = 0.37 lb P/ac

0
0.5
1
1.5
2
2.5
10
12
13
18
19
20
21
22
23
24
25
26
27
28
29
30
31
42
43
44
45
46
47
52
53
54
55
56
57
58
59
14A
14B
15A
15B
AnnualDRPLoad(lb/acre)
Field Number
Old DRP New DRP
Combined Legacy & New DRP Loads
(Surface + Tile)
Legacy
DRP

Contribution of New P to Overall DRP Load
(Surface + Tile)
0
10
20
30
40
50
60
70
80
90
100
10
12
13
18
19
20
21
22
23
24
25
26
27
28
29
30
31
42
43
44
45
46
47
52
53
54
55
56
57
58
59
14A
14B
15A
15B
NewDRP(%oftotalDRPload)
Field Number
Average = 17%

Combined Legacy & New TP Loads
(Surface + Tile)
0
0.5
1
1.5
2
2.5
3
3.5
4
10
12
13
18
19
20
21
22
23
24
25
26
27
28
29
30
31
42
43
44
45
46
47
52
53
54
55
56
57
58
59
14A
14B
15A
15B
AnnualTPLoad(lb/ac)
Field Number
Legacy TP New TP

-0.5
0
0.5
1
1.5
2
2.5
3
3.5
13
12
18
22
19
56
42
22
29
28
28
42
21
46
57
23
57
13
21
29
56
23
23
54
23
21
20
52
22
18
20
56
29
19
53
13
21
46
28
43
29
12
28
12
21
55
21
31
20
18
31
54
19
30
20
20
21
46
55
30
57
56
21
20
10
12
44
15
10
14
10
15
14
27
45
26
26
14
44
15
10
23
27
45
10
15
44
45
10
25
24
14
13
13
25
12
54
25
24
25
24
24
NewTP(lb/acre)
ManureChemical fertilizer
-0.5
0
0.5
1
1.5
2
2.5
3
3.5
13
12
18
22
19
56
42
22
29
28
28
42
21
46
57
23
57
13
21
29
56
23
23
54
23
21
20
52
22
18
20
56
29
19
53
13
21
46
28
43
29
12
28
12
21
55
21
31
20
18
31
54
19
30
20
20
21
46
55
30
57
56
21
20
10
12
44
15
10
14
10
15
14
27
45
26
26
14
44
15
10
23
27
45
10
15
44
45
10
25
24
14
13
13
25
12
54
25
24
25
24
24
NewDRP(lb/acre)
New P from Individual P Applications
(Surface & Tile)
March
(5)
April
(10)
May
(3)
June
(4)
Aug
(5)
Sept
(1)
Oct
(9)
Nov
(6)
Dec
(3)
All Broadcast Applications
Surface Tile

P fertilizer form and placement: New DRP
(Surface + Tile)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Chem, solid,
broadcast
(27)
Chem, solid,
incorporated
(19)
Chem, liquid,
injected (18)
Manure,
liquid,
broadcast (7)
Manure,
liquid,
incorporated
(3)
Manure,
liquid,
injected (14)
Manure,
solid,
broadcast
(12)
Manure,
solid,
incorporated
(2)
NewDRP(lb/acre)

P fertilizer form and placement: New TP
(Surface + Tile)
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
Chem, solid,
broadcast
(27)
Chem, solid,
incorporated
(19)
Chem, liquid,
injected (18)
Manure,
liquid,
broadcast (7)
Manure,
liquid,
incorporated
(3)
Manure,
liquid,
injected (14)
Manure,
solid,
broadcast
(12)
Manure,
solid,
incorporated
(2)
NewTP(lb/acre)

P application rate: New DRP Loads
-0.2
0.2
0.6
1
1.4
1.8
2.2
2.6
0 30 60 90 120
NewDRP(lb/acre)
P Rate (lb P/acre)
Surface
-0.2
0.2
0.6
1
1.4
1.8
2.2
2.6
0 30 60 90 120
NewDRP(lb/acre)
P Rate (lb P/acre)
Tile

P application timing
(Surface + Tile)
-0.5
0
0.5
1
1.5
2
2.5
March
(5)
April
(10)
May
(3)
June
(4)
Aug
(5)
Sept
(1)
Oct
(9)
Nov
(6)
Dec
(3)
NewDRP(lb/acre)
Surface Tile
-0.5
0
0.5
1
1.5
2
2.5
March
(5)
April
(10)
May
(3)
June
(4)
Aug
(5)
Sept
(1)
Oct
(9)
Nov
(6)
Dec
(3)
NewTP(lb/acre)

Legacy P loss: Influence of STP
0
0.5
1
1.5
2
2.5
0 100 200 300 400 500
Tile+surfaceannuallegacyDRP(lb/ac)
Field average STP (M3 ppm, 0-8 in)

Legacy P loss: Influence of STP
0
0.2
0.4
0.6
0.8
1
0 50 100 150 200
Tile+surfaceannuallegacyDRP(lb/ac)
Field average STP (M3 ppm, 0-8 in)

Summary of initial findings
• Losses of new P averaged 17% of overall DRP loss (but up to
70% from individual fields)
• 11% of TP
• Riskiest applications were broadcast liquid manure (DRP lost
through tile drains)
• Risk of new P loss from broadcast applications:
Late fall/winter > spring > summer/early fall
• No relationship between new P load and P rate
• Legacy P loss is related to STP, other factors TBD

Implications
• P fertilizer management (i.e. 4Rs) will have a limited
short term impact on overall DRP and TP loads
• Need additional focus on management of legacy
soil P losses
• Draw down soil P, but this is a long term solution
• Manage hydrology?
• Manage P stratification?
• Capture at field edge or downstream?

Acknowledgements
• Cooperating farmers
• Jed Stinner
• Phil Levison
• Katie Rumora
• Sarah Henderson
• Eric Fischer
• Marie Pollock
• Brittany Hanrahan
• Funding sources
• USDA-ARS
• USDA-NRCS
• US-EPA
• Ohio Farm Bureau
• International Plant Nutrition
Institute
• The Nature Conservancy
• Becks Hybrids
• Delaware County Soil &
Water Conservation District
• Ohio State University,
• Ohio Corn and Wheat
Growers
• Ohio Soybean Association

Combined results - P applications
0
10
20
30
40
50
60
70
80
0 10 20 30 40 50 60
%NewP
Annual average P applied (lb/acre)
0
10
20
30
40
50
60
70
80
0 0.5 1 1.5 2 2.5
%NewP
P applications per year

P fertilizer form and placement
NewDRP(lb/acre)
-0.5
0
0.5
1
1.5
2
2.5
Surface New DRP
Chem, solid, broadcast (24)
Chem, solid, incorporated (18)
Chem, liquid, injected (18)
Manure, liquid, broadcast (7)
Manure, liquid, incorporated
(2)
Manure, liquid, injected (14)
Manure, solid, broadcast (17)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
NewDRP(lb/acre)
Surface New DRP

Individual P applications: Tile drains
-0.5
0.5
1.5
2.5
3.5
13B
12B
17B
16B
18B
22B
19B
56B
42B
47B
57B
22B
17B
29B
16B
28B
28B
47B
42B
21B
46B
57B
16B
23B
16B
57B
13B
21B
29B
56B
23B
51B
17B
23B
54B
23B
21B
16B
20B
52B
22B
18B
17B
20B
56B
29B
19B
53B
13B
21B
46B
28B
43B
29B
12B
28B
12B
21B
55B
21B
17B
31B
16B
20B
50B
18B
17B
31B
54B
19B
30B
20B
20B
21B
46B
55B
30B
57B
56B
21B
20B
10B
12B
44B
15B
15C
14C
10B
10B
15B
15C
14B
14C
27B
14B
50B
51B
45B
15B
14C
51B
26B
26B
50B
51B
44B
51B
10B
23B
15C
14B
15C
27B
45B
10B
44B
45B
15B
10B
25B
14B
24B
14C
13B
13B
25B
12B
54B
25B
24B
25B
24B
24B
NewDRP(lb/acre)
New DRP
-1
0
1
2
3
4
13B
12B
17B
16B
18B
22B
19B
56B
42B
47B
57B
22B
17B
29B
16B
28B
28B
47B
42B
21B
46B
57B
16B
23B
16B
57B
13B
21B
29B
56B
23B
51B
17B
23B
54B
23B
21B
16B
20B
52B
22B
18B
17B
20B
56B
29B
19B
53B
13B
21B
46B
28B
43B
29B
12B
28B
12B
21B
55B
21B
17B
31B
16B
20B
50B
18B
17B
31B
54B
19B
30B
20B
20B
21B
46B
55B
30B
57B
56B
21B
20B
10B
12B
44B
15B
15C
14C
10B
10B
15B
15C
14B
14C
27B
14B
50B
51B
45B
15B
14C
51B
26B
26B
50B
51B
44B
51B
10B
23B
15C
14B
15C
27B
45B
10B
44B
45B
15B
10B
25B
14B
24B
14C
13B
13B
25B
12B
54B
25B
24B
25B
24B
24B
NewTP(lb/acre)
P applications
New Total P

Individual P applications: Surface
-0.4
0.1
0.6
1.1
30A
12A
56A
19A
28A
22A
13A
46A
43A
23A
28A
12A
56A
21A
29A
20A
57A
56A
23A
29A
52A
28A
55A
21A
55A
29A
22A
54A
19A
54A
42A
20A
42A
56A
18A
22A
46A
18A
21A
57A
23A
21A
20A
21A
18A
20A
57A
28A
31A
29A
21A
53A
20A
23A
30A
31A
19A
20A
21A
46A
44A
24A
25A
45A
44A
45A
12A
15A
21A
14A
15A
24A
26A
45A
15A
15A
27A
44A
13A
14A
14A
25A
26A
13A
44A
44A
14A
12A
23A
25A
10A
27A
24A
45A
44A
45A
10A
45A
54A
10A
NewDRP(lb/acre)
New DRP
-0.4
0.1
0.6
1.1
30A
12A
56A
19A
28A
22A
13A
46A
43A
23A
28A
12A
56A
21A
29A
20A
57A
56A
23A
29A
52A
28A
55A
21A
55A
29A
22A
54A
19A
54A
42A
20A
42A
56A
18A
22A
46A
18A
21A
57A
23A
21A
20A
21A
18A
20A
57A
28A
31A
29A
21A
53A
20A
23A
30A
31A
19A
20A
21A
46A
44A
24A
25A
45A
44A
45A
12A
15A
21A
14A
15A
24A
26A
45A
15A
15A
27A
44A
13A
14A
14A
25A
26A
13A
44A
44A
14A
12A
23A
25A
10A
27A
24A
45A
44A
45A
10A
45A
54A
10A
NewTP(lbP/acre)
New Total P

P fertilizer form and placement
NewDRP(lb/acre)
-0.5
0
0.5
1
1.5
2
2.5
Tile New DRP
Chem, solid, broadcast (29)
Chem, solid, incorporated (26)
Chem, liquid, injected (29)
Manure, liquid, broadcast (9)
Manure, liquid, incorporated
(3)
Manure, liquid, injected (26)
Manure, solid, broadcast (12)
Manure, solid, incorporated (2)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
NewDRP(lb/acre)
Tile New DRP

Michalak et al. 2013
• Binational P reduction targets:
• 40% reduction in total P (TP) and dissolved reactive P (DRP)
loads entering Western Lake Erie Basin by 2025

EoF dataset scope
• 41 fields monitored
• 43 tile drains
• 34 surface runoff flumes
• Examined data through September 2019
• Average monitoring period = 5 years, 9 months
• 178 P applications
• 90 chemical fertilizer
• 88 manure

Factors – Time period for newly
applied P “risk period”
• 6 month period changes estimate of “new” P
contribution – DRP increases from 17% to 19% DRP
but TP decreases from 10 to 9%
• 1 month period misses a large number of P
applications due to little or no runoff, especially
surface
• Use precipitation to define the period?

Factors affecting likelihood of
recent P fertilizer losses?
• Factors
• Precipitation
• Tillage – Williams et al. 2017
• Slope
• Manure vs. chemical fertilizer

Next steps
• Use precipitation to define New P loss risk period
• Assess the uncertainty of New P estimates
• Delve into Old P losses – effects of tillage? cover
crops?

Dr. Will Osterholz - What's More Important For Water Quality: Recent P Applications Or Legacy Soil Test P?

Recommended

Recommended

More Related Content

What's hot

What's hot (18)

Similar to Dr. Will Osterholz - What's More Important For Water Quality: Recent P Applications Or Legacy Soil Test P?

Similar to Dr. Will Osterholz - What's More Important For Water Quality: Recent P Applications Or Legacy Soil Test P? (20)

More from John Blue

More from John Blue (20)

Recently uploaded

Recently uploaded (20)

Dr. Will Osterholz - What's More Important For Water Quality: Recent P Applications Or Legacy Soil Test P?