Project: MLMMI 01-01-04
Objective
In Phase 1 ECOMatters developed a Nutrient Loading Model (NLM) to establish loading limits for P. Year 2 is to review very recent developments - the State of Maryland Phosphorous Site Index Model, compare to the models, ensure that the NLM has all required processes included in it and then validate the model.
Performer
Steve Sheppard
ECOMatters Inc.
Details
Status: Completed
Started: 2001-06-18
Completed: 2002-10-01
Funding Partners: who have contributed to MLMMI in support of this project:
ARDI - $28,600
Manitoba Pork Council and industry groups - $28,600
Amount Funded: $57,200.00
Performer Funded: $21,300.00
Total Cost: $78,500.00
Activity
First interim report received Oct. 29/01.
Second Progress Report received March 21/02.
Final report due Sept. 1, 2002.
Request granted for extension to Oct.1/02.
Final Report received Oct. 2002.
Summary
With the increase in the livestock industry in Manitoba, there is more manure that must be managed in an environmentally acceptable manner. Nutrient content of the manure and its addition to land is one of the aspects that needs to be managed. For nitrogen (N), the issue is more or less resolved by ensuring the N addition matches the crop requirements, so that on average there is little or no residual environmentally important (soluble) N. For phosphorus (P), the issue is not as simple because crops cannot in a single season remove all the environmentally important P. The reason is that the P sorbs strongly to soil particles, and the environmental importance comes to a large extent from erosion. The Nutrient Loading Model (NLM) was designed to be a tool that a regulator could use to set P loading limits (P fertilization or manure-loading limits) based on the effect of P on stream water quality. Phase 1 of the project (1999-2000) developed the model. Phase 2, described here, considered improvements to the model, benchmarked the model against other models available for this purpose, and validated the model.
The benchmarking process involved comparison of the capabilities of 7 models to that of the NLM. These models varied from complex research-level models with far too many parameters for practical application, to simple look up tables in use now for regulatory purpose. In general, these other models were not readily useable with available data and did not have attributes that were important to add to the NLM.
The validation of the NLM was the major undertaking. It was not possible to conduct experiments or measurements within this project, so the emphasis was to identify existing data and programs in Manitoba that met the requirements. There are quite a few programs in Manitoba related to stream quality. The drawback is that they are conducted by a number of agencies with differing mandates and differing abilities to consistently measure the important parameters. Four case studies were selected. These included reaches of the Whitemouth River, Joubert Creek, the Seine River and the west watershed of the South Tobacco Creek. The river reaches were about 10 km long, so there were many crops and landscape features included. The reaches were chosen to be predominantly through crop land, but there were inevitably other sources of P to the river, such as leaching from overhanging vegetation. In all cases, drawing the required information from the available data required assumptions and some interpolation. Also, no calibration for each river reach was carried out, this created an unbiased test. For better results, the NLM could be calibrated before application and there is data to do this for the Whitemouth River, Joubert Creek as well as the South Tobacco Creek watersheds. Given these uncertainties, the results of the NLM for the rainfall events considered agreed reasonably well with the validation data.
There are many ways the NLM could be implemented in a management, guidelines or regulatory role. It was envisioned to be a tool for a regulator, where an upper limit application rate of P to soils is determined based on the concentration of P in the receiving water. This application probably needs to be quite generic with respect to watershed characteristics. For example, it would not be fair to impose guidelines on a farmer in the lower regions of the watershed that are different from those for a farmer in the upper regions. Several alternative application modes are suggested, one of which is the concept of a water parcel. This water parcel is the water contributed by a unit of land to the total drainage effluent of a watershed, and it contains the P eroded (or leached) from that unit of land. This concept is for an annual average water and P loss, on the assumption that the appropriate integration time is a year. The P in this parcel of water may be subject to some fluvial processes such as sedimentation, but ultimately it is this P concentration that may, for example, impact the North Basin of Lake Winnipeg. It is proposed that the NLM is ready now for such an application.