|Main authors:||R.K. Laursen, F. Bondgaard, P. Schipper, K. Verloop, L. Tendler, R. Cassidy, L. Farrow, D. Doody, F. A. Nicholson, J. R. Williams, I. Wright, J. Rowbottom, I. A. Leitão, A. Ferreira, B. Hasler, M. Glavan, A. Jamsek, N. Surdyk, J. van Vliet, P. Leendertse, M. Hoogendoorn and L. Jackson-Blake.|
|Source document:||R.K. Laursen et al. (2019) Evaluation of Decision Supports Tools. FAIRWAY Project Deliverable 5.2 216 pp|
A comprehensive review and survey of Decision Support Tools (DSTs) currently in use in the FAIRWAY case studies is described in »Survey and review of existing decision support tools. Of the 36 DSTs identified as most relevant, 12 were selected for further investigation to see if a tool developed in a particular national context could be used or provide inspiration elsewhere (»Evaluation of decision support tools). Here we describe the tools evaluated for potential use in the Overijssel case study.
|1. Selection of DST to evaluation in Overijssel case study|
The FAIRWAY case study site at Overijssel deals with nitrate leaching to groundwater and how dairy farmers within the capture zone of vulnerable drinking water resources can improve their mineral management with less N-losses to the soil (»Overijssel, NL case study).
This case study site covers five regions in the Provence of Overijssel (Eastern sandy soil in the Netherlands): Archemerberg, Hoge Hexel, Wierden en Herikerberg/Goor, with 16 participating dairy farmers. The drinking water wells were indicated susceptible to agricultural pollution due to the hydrogeological situation (deep draining, sandy soils) and due to land use (dairy farming and arable farming). Each of the five abstractions use groundwater. The soils are characterized by a shallow anthropogenic layer with an organic matter content of 3-5% overlaying a deep layer of yellow sand, very low in soil organic matter.
The main focus in the case is on the reduction of nitrate leaching from concentrations in the upper metre of groundwater in the range of 80-100 mg/l to lower than 50 mg/l. Measures are implemented at 16 farms. The effectiveness of measures are evaluated on the basis of the N surplus on the soil balance (kg/ha), the mineral N content in the upper soil layers in autumn and monitoring of nitrate concentrations in the upper meter of groundwater. Measures that are considered most relevant are: improving the grazing strategy, preventing grazing in autumn, undersow of Italian Ryegrass in maize, improving soil quality and optimizing fertilization and spreading of manure.
The relevant stakeholders are: Province of Overijssel, Vitens (water abstraction company), Dairy farmers participating in the project, Contractors that carry out a part of the practical work on farms, EU and national government. These are the basis of the awareness that more strict legislation may follow if the problem with elevated nitrate concentration is not solved. Farm advisors that visit the farm also feed suppliers.
This assessment and tests were conducted by researchers involved in the case study and discussed with farm advisors and with stakeholders that are also the founders of the project Overijssel. We were interested in exploring the applicability of Düngeplanung because in the case of Overijssel optimized distribution of organic and mineral fertilizers over crops and fields are of significant importance for preserving ground water quality in dairy farming regions. Moreover we feel that in the Netherlands knowledge valorisation on optimized fertilization could be improved. We discussed the assessment with professionals in Germany that developed Düngeplanung and apply it in their region. We tested application of Düngeplanung on the experimental farm De Marke and on one commercial farm in the region. This process was reported and discussed with the farmer.
Implementing the software and getting it running was time-consuming process in spite of accurate and effective assistance and guidance by the developers of the software in Germany. This was possibly caused by safety settings that were incorporated in the software. It also took some time before we developed the skills needed to work with Düngeplanung. Language problems could be overcome rather easily by the researchers involved, but this may not hold for practical implementation on farms. Another aspect is that the interface, although well organized, is quite comprehensive.
Data requirements and outputs
To operate Düngeplanung data the following should be supplied:
- Manager of the farm/address and location of the farm
- Fields on the farm including o Agricultural areal per field (ha)
- Crops and precrop per field (species)
- Soil data including soil type, organic matter content, pH and P, K, Mg status
- Nutrients or nutrient carrier applied, including composition
Some data are not used in the Netherlands, e.g. indicators for P status deviate from the parameters used in the Netherlands. The indicators that are commonly used in the Netherlands could possibly be converted to the indicators used as input in Düngeplanung. However, checks are required on the consequences of such conversions. The categories used to indicate soil type are also used in the Netherlands, e.g. lehmiger sand corresponds to lemig zand. However, before implementation further analyses is required to check whether or not the classification is based on the same criteria in terms of composition of texture classes.
Data input is user-friendly. There are options to import data from data systems that are available for German users. These, of course, do not match with the data systems that are used in the Netherlands. Default settings can be used, but these can be specified if desired.
The Düngeplanung produces a practical fertilization plan, expressed in kg total material applied per ha, ready to use for who carries out the Fertilization. Moreover, it produces information on N and P balances of the crop production per field. If a user allocated fertilizers to fields and crops, a report can be printed that shows how much of a fertilizer type will be distributed according to the plan. This way the user can cross-check if the planned spreading of manure matches with the amount that is produced or bought. The same holds for mineral fertilizer.
Functionality within the Dutch context
A major characteristic of Düngeplanung is that it integrates the fertilizer requirements of fields defined on a farm into a fertilizer plan at farm level. It addresses:
- specific fields with their crop history and their soil characteristics
- a great variety of crop species
- The characteristics and chemical composition of organic materials functioning as carrier for specific nutrients
- Residual effects of organic fertilization applied in previous years
The added value is not so much in the total absence of these aspects in the Dutch recommendations but more in the integration of these recommendations with adequate accuracy and precision to specific farms and their fields. This can explained by a short description of the agricultural context.
Until the 1950s most farms in the Netherlands where integrated meaning that farmers used their land as basis for crops, dairy and/or meat production. Farmers needed to be skilled in both animal management and crop management. However, since the 1950s agricultural production was increasingly based on specialized arable production on the one hand and specialized dairy and meat production on the other hand. This trend resulted in disintegrated knowledge development and different cultures in the arable sector and the dairy production sector. In the dairy sector the major focus was on herd management at cost of a gradual decline of focus on soil and crop management. Management skills on crop production ceased over time and dairy farmers adopted relatively simple and robust approaches of crop production, based on only grassland and silage maize and with low appreciation of accurate planning of fertilization. As a consequence the accuracy and precision of Dutch fertilizer recommendations, with a scientific and empirical basis that is more or less similar to the German fertilizer recommendations, is not fully reflected in the daily practice of dairy farming.
Fertilizer recommendations are established for the relevant nutrients: N, P, K, S, Mg etcetera, whereas a large share of the nutrients are brought to the field in the form of organic carrier material such as farm slurry. To comply with fertilizer recommendations the variability in composition of those nutrients in farm slurry must be addressed. Moreover, the land use history, i.e. the crops cultivated in preceding years should be addressed as well as the soil characteristics and soil fertility. This farm specific information should be integrated to convert fertilizer recommendations into tailor made farm specific fertilizer plans. Many dairy farmers consider this conversion too complex and work according to tacit knowledge or simply use farm quota for N and P rates (i.e. the amount of N and P that can be applied within the legal application standards) as a basis for fertilization.
Also farm quotas for the use of N and P rates to farmland that came into force in 2006 affected practical farm management. These quotas limit both organic and mineral fertilizer N and P input rates to soil which urges intensive dairy farmers to export part of the manure produced by cattle from their farm as the animal production exceed the fertilizer N or P quotas. In fact these quota are crop specific application standards, but farmers are free to decide on the distribution over crops and fields on their farms. Since the implementation of the quotas farmers tend to tune their fertilizer application rates not to recommendations but to the allowed quotas for maize and grass without paying much attention to differences between fields in production capacity and expected yields, soil characteristics and land use in earlier years. This practice lacks precision which causes high risks of nitrate leaching. To improve this there is an urgent need for more accuracy in fertilization in dairy farming and the PerceelsVerdeler was developed to support this accuracy (Oenema et al., 2017).
The PerceelsVerdeler has some similarities to the Düngeplanung. However, the PerceelsVerdeler only addresses grass and maize. Therefore the PerceelsVerdeler falls short where other crops are integrated into crop rotations. In spite of the specialized character of arable production and dairy production crop rotation in which arable crops, e.g. potatoes, grains or beets are alternated with grassland occurs regularly where dairy farmers rent parts of their farmland to arable farmers. In addition, advantages of Düngeplanung are the wider range of nutrients addressed as compared to the PerceelsVerdeler (looking at only N, P, K), the more systematic and formal way to incorporate soil data and the easy way to build up sound records of crop history for specified fields on the farm.
The Düngeplanung does not address N fertilizer quotas but considers crop fertilizer N requirements as threshold for maximum N rates in the fertilizer plans. Rates of organic and mineral fertilizer N and P are limited in the Dutch regulation. On the basis of these limits (specified per crop) expressed in kg per ha and the areal of each crop on the farm a farm budget for N and P is established. This budget can be freely allocated to the crops and parcels of the farm. Thus farm fertilizer plans should respect the farm N quota, and when N quota are lower than the fertilizer recommendations, they should suggest an optimal distribution of the N and P quota. The PerceelsVerdeler uses the quota as a starting point to recommend optimal distribution. As far as we see, the Düngeplanung does not refer to quota set by regulation. In this respect Düngeplanung seems, similar to the approach in Dutch fertilizer recommendations. It is a rational approach for distribution of N as long as farm N quota are not exceeded. But when recommended N rates exceed the level of farm N quota farmers need to know on which fields and on which crops they apply a discount to the N rates in order to be meet the farm N budgets. For these circumstances it would be helpful if Düngeplanung would address farm N quota based on regulation as the PerceelsVerdeler does. For P Düngeplanung tunes to equilibrium fertilization on fields with high soil P status. Therefore, for P the recommendations are well in agreement with P quota in the Netherlands that also are based on P equilibrium fertilization.
The current systems developed to support fertilizer plans for arable famers do address crop history but do not support evaluation of the cropping system in terms of the N and P surplus (kg/ha) on the soil balance, whereas Düngeplannung does. This functionality is relevant for the case Overijssel because in the nearby future the case will be elaborated to arable farmers in drinking water abstractions. In the evaluation of the environmental pressure caused by arable farming the N surplus is one of the relevant indicators.
The assessment of Düngeplanung for Dutch circumstances are summarized in Table 33.
Table 33. Advantages and disadvantages of the use of Düngeplannung as a basis for fertilizer plans
|User-friendly interface, systematic and consistent design||Comprehensive data input needed|
|All relevant crop species are incorporated||Conversion of some parameters e.g. P status needed|
|Integration of relevant farm data on fields, soil fertility, crops and crop history and organic nutrient carriers with their specific composition||N quota on farm level are not addressed|
|Production of a practical list for fertilziation||Check on recommendations or adaptation to Dutch recommendations needed|
|Minor problems concerning software implementation and language need to be solved upon implementation|
Discussions with the project managers, farm advisors, researchers and farmers involved in the case Overijssel resulted in the conclusions that it is not recommended to implement Düngeplannung directly in the case Overijssel for the following reasons:
- Input data used in Düngeplanung are different from the parameters used in the Netherlands.
- Without additional tests and comparisons there is too much uncertainty about the applicability of the German fertilizer recommendations under Dutch circumstances.
- It is important to address the distribution of the N quota on a farm, also indicated as N budget, over the fields of the farm.
- It is important to benefit from data supplied by ANCA. This enables users to use farm specific default values concerning the availability of organic manure on the farm, grazing intensity, the chemical composition of organic manures and crop yields. This link is incorporated in the software of PerceelsVerdeler.
Instead it was suggested to organize a further exchange on the concept of Düngeplanung and the PerceelsVerdeler or Dutch fertilizer recommendations. This could result in a DST that has a higher practical impact and that could stimulate farmers to optimize the distribution of fertilizers over fields. This could be realised by adoption of strong functions of Düngeplanung in Dutch systems and vice versa and could result in mutual gains for the stakeholders involved (science, drinking water company, National and regional governments).
For full references to papers quoted in this article see » References