Conceptual Background

Transcription

1 21//15 Plantation Intelligence Analysis of Commercial Data for Yield and Fertilizer Management in Oil Palm OBERTHÜR, Thomas Director South East Asia Program, International Plant Nutrition Institute 1 Conceptual Background 1

2 21//15 IPNI 4R Nutrient Stewardship Decision Making Uncertainty Complex interactions in an agronomic system render outcomes of any management decision Uncertain 2

3 21//15 Decision Making Uncertainty METRIC Uncertainty about rate & placement of fertilizer to support a yield target Example: ROI in Fertilizer TRANSLATIONAL Uncertainty from external factors that reduce fertilizer performance, e.g. harvest, mill and transport efficiency TEMPORAL Uncertainty about timing of fertilizer applications, e.g. drought interference STRUCTURAL Uncertainty from internal factors that influence fertilizer efficiency, e.g. EFB applications Cook et al. Better Crops 97, 17 - (13) Business Intelligence Hans Peter Luhn, IBM Journal, 1958 A Business Intelligence System (business) intelligence is the ability to apprehend the interrelationships of presented facts in such a way as to guide action towards a desired goal. 3

4 21//15 Business Intelligence Plantation Intelligence System is Monitored in Extraordinary Detail 4

5 21//15 Plantation Intelligence Exogenous Organize existing performance data (VISUALIZE) Quantify performance change for management (DECIDE) Generate performance indicators and metrics (ANALYSE) Devise specific performance intervention options (EVALUATE) Exogenous Plantation Intelligence An adaptive learning process based on the analysis of plantation data 5

6 21//15 Plantation Intelligence Business Opportunity PALMSIM Model Measure Performance Plantation Intelligence Estate Networking The Benchmark Club Continuous Improvement Best Management Practices 2 Application Examples 6

7 FFB (t/ha) 21//15 Current Protocols 7 Yield Age Profiling Yield Trends Yield Soil Interactions Yield Soil Climate Interactions Fertilizer Response Analyses Naïve Gross Margins Yield Labor Interactions Yield Age Profiling TREEAGE 7

8 FFB (t/ha) FFB (t/ha) 21//15 Yield Age Profiling TREEAGE Yield Benchmarking Average yield profile mirrors that of potential 3. Blocks approach yield potential 4. Blocks above potential require further assessment 0 2. Yield gap between average and potential PALMSIM Model Agricultural Systems, 131:1- (14) The Planter, 91: (15) TREEAGE 8

9 Average naïve gross margin in USD per hectare Average naïve gross margin in USD per hectare 21//15 Naïve Gross Margins 5500 Cost = 500 $US per ha per year i.e. a favorable benchmark Individual Block Naïve Gross Margins Individual Block 9

10 Average naïve gross margin in USD per hectare 21//15 Naïve Gross Margins Individual Block 3 Response to Fertilizer

11 21//15 Return on Investment NPKMg BREAK EVEN 8~ kg kg point fresh fruit bunches per kg nutrients Return on Investment NPKMg Sum of NPKMg 2.5 years to 0.5 year before harvest 11

12 Fresh fruit bunches in tons per hectare Fresh fruit bunches in tons per hectare Fresh fruit bunches in tons per hectare 21//15 Tree Age Effects Steep Ascent : Palm age from y= (0.011*x) 25 Plateau : Palm age from y= (0.005*x) NPKMg3yW_IN NPKMg3yW_IN Annual Response NPKMg3yW_IN 12

13 Fresh fruit bunches in tons per hectare Fresh fruit bunches in tons per hectare 21//15 Annual Response y= (0.024*x) y= (-0.015*x) y= (-0.022*x) y= (-0.003*x) SMG A SMG B SMG C SMG D NPKMg3yW_IN Annual Response NPKMg3yW_IN 13

14 Fresh fruit bunches in tons per hectare 21//15 Annual Response y= (-0.054*x) y= (0.025*x) y= (0.025*x) y= (0.036*x) SMG A SMG B SMG C SMG D NPKMg3yW_IN Visualizing Local Response Contrast Method Yield Response (t FFB ha y) = F 1 (PV 1, PV 2, PV N, PV NPKMg ) Yield Response (t FFB ha y) = F 2 (PV 1, PV 2, PV N ) Response Contrast (t FFB ha y) = F 2 F 1 14

15 21//15 Visualizing Local Response Contrast in t FFB per ha per year relative to yield without fertilizer Visualizing Local Response Contrast in t FFB per ha per year relative to yield without fertilizer

16 Fresh Fruit Bunch Yield in t per ha 21//15 4 Disclaimer Confounding Effects Harvest Man days 16

17 Fresh Fruit Bunch Yield in t per ha Fresh Fruit Bunch Yield in t per ha 21//15 Confounding Effects Expected yield level Harvest Man days Confounding Effects Fruit is grown but not harvested Only with enough labor, all fruit is harvested Expected yield level Harvest Man days 17

18 21//15 Confounding Effects EXPECTATION Yield Made VS. Yield Taken REALITY IPNI Southeast Asia Program 4WA -22 October 15 18