ac456d32-35c3-11df-a02c-00219b106224

eng

Paul Bullock

University of Manitoba

(204) 474-8666

Department of Soil Science University of Manitoba

Winnipeg

Manitoba

R3T 2N2

Canada

Paul_Bullock@umanitoba.ca

author

2006-12-31

ISO 19115

ISO19115:2003/Cor 1 2006

Prairie Census Agricultural Region Yield Deviation

2006-12-31

creation

Statistics Canada collects comprehensive data on the area, yield and production of all crops across Canada. Crop statistics by Census Agricultural Region (CAR) covering the period 1976 through 2006 were utilized for this analysis. The data from 4 crops (all spring wheat, barley, canola, field peas) for 40 CARs (8 in Alberta, 20 in Saskatchewan and 12 in Manitoba) were included. For each crop in each CAR, a linear time trend for yield was calculated and used as a baseline. The length of record available for trend line calculation varied by crop and by CAR and ranged from 12 to 31 years. Most trend lines had a positive slope but some crops in some CARs displayed a negative slope. However, the trend line was used in each case without any adjustment. For each year of record for each crop in each CAR, the yearly percentage deviation from the trend was calculated as (reported yield - trend yield)/trend yield. For each crop and CAR, the percentage yield deviations were transformed to percentile values such that a value of 100 was the highest positive deviation in the data series and 0 was below the most negative deviation in the data series. The percentile yield deviation from trend for each crop, in each CAR and each year of record is provided in the database.

The delay between the time that Statistics Canada collects crop data and when it is reported means that this information cannot be used to provide a timely response to drought or other natural disasters that reduce agricultural production. However, by relating more timely indicators such as weather data and satellite images to these crop statistics, quantitative measures can be derived to delineate the spatial extent and intensity of agricultural drought in a more timely manner. The objective of this research was to produce quantitative and representative agricultural measures of the impact of the drought period spanning the period from 1999 to 2004 using crop yield statistics for defined regions of western Canada. The mixture of crops grown on the prairies varies spatially and temporally, therefore, defining a single yield value that could be considered representative of the annual variation in yield in each region is complex. These yield values will be useful for assessing the ability of weather station data and satellite information to quantify the impacts of weather on annual variability in crop yield.

completed

Paul Bullock

University of Manitoba

(204) 474-8666

Department of Soil Science University of Manitoba

Winnipeg

Manitoba

R3T 2N2

Canada

Paul_Bullock@umanitoba.ca

author

notPlanned

Agriculture > Agricultural Plant Science > Crop/Plant Yields

Biosphere > Vegetation

Land Surface > Land Use/Land Cover > Land Productivity

GCMD

2007-04-01

Publication

vector

eng

farming

-120

96

49

60

01/01/1976

31/12/2006

annual

ASCII - Formatted for text attributes

2003

DRI Data Legacy

http://www.drinetwork.ca/data-legacy/portal.php

Distributor

0.04

Dataset

Statistics Canada's survey frame is derived from the Census of Agriculture and stratified according to farm size. A random sample of farms is selected from each stratum along with a weight. These weights are used to expand the data obtained from the sample to an estimate of the entire stratum. Data are collected directly from survey respondents and compared both on a year-to-year basis and between provinces. Imputation is done for yield at the provincial level using an average yield by crop calculated from all complete records. The average is imputed into records with missing values for yield. The information collected in the probability sample is weighted to produce unbiased survey estimates called level indicators. These level indicators are representative of the entire farm operator population. Also from two level indicators, it is possible to derive unbiased ratio indicators such as the pair change ratio of the seeded acreage from year to year. Ratio indicators are used to estimate seeded/harvested area and average yield estimation. Crop production estimates are based on level indicators obtained from a probability survey of farming operations. The potential error introduced by sampling can be estimated from the sample itself by using the coefficient of variation (CV). Over repeated surveys, 95 times out of 100, the relative difference between a sample estimate and what should have been obtained from an enumeration of all farming operations would be less than twice the coefficient of variation (the confidence interval). While published estimates may not exactly equal the level indicators (due to the validation and consultation process), these estimates do remain within the confidence interval of the survey level indicators. CVs at the national level range from 5% to 20% for the major crops. Statistics Canada is prohibited by law from releasing any data that would divulge information obtained under the Statistics Act that relates to any identifiable person, business or organization. Various confidentiality rules are applied to all data that are released or published to prevent disclosure of any information deemed confidential. If necessary, data are suppressed to prevent direct or residual disclosure of identifiable data.