South Africa faces low corn supply after disappointing seasonal rains

Below-average seasonal rainfall and an untimely dry spell during mid-February through mid-March, 2012 will reduce South Africa’s corn yields and corn supply for the 2011/12 season. The 2011/12 seasonal rainfall was below-average this year even though the La Nina year forecast was above-average rainfall for the region (refer to Figure 1). Above-average rainfall returned to South Africa’s grain belt in late-March, but these rains arrived too late to make any significant improvements to potential yields. In summary, USDA lowered South Africa corn yields last month from 3.75 tons per hectare (T/Ha) to 3.59 T/Ha, which is well below the five-year average yield of 3.77 T/Ha. South Africa will also be facing low corn supplies this year with total corn production forecast at 11.5 million tons (MT); total domestic consumption estimated at 10.6 MT (refer to FAS Annual GAIN report); and carry-over stocks at 1.3 MT at the end of March (refer to SAGIS (South African Grain Information Service)).

Two consecutive La Nina years, with seasonal rainfall above-average last year and below-average this year

Seasonal rainfall in South Africa’s grain basket was below-average this year and above-average last year, even though both years were La Nina years. This year’s below-average seasonal rainfall in South Africa was unexpected because seasonal rainfall during La Nina years tend to bring above-average seasonal rainfall for the region (refer to Figure 2).

Figure 1. Two consecutive La Nina years,
with seasonal rainfall above-average last year (left) and below-average this year (right).

Figure 2. The 2011/12 seasonal rainfall in South Africa was below-average (left),
but the La Nina forecast was for seasonal rainfall to be above-average (right).

Another method to monitor crop conditions year-to year is to utilize satellite imagery from the MODIS (Moderate Resolution Imaging Spectroradiometer) sensor onboard NASA’s Terra and Aqua satellites. The USDA/NASA Global Agriculture Monitoring (GLAM) project automatically processes the red and infrared MODIS bands to produce NDVI (Normalized Difference Vegetation Index) composites to measure the vegetation’s relative “greenness” from year to year and compare it with the 10-year average vegetation greenness. Current and historical global MODIS-NDVI composites are archived and made readily available by the USDA/NASA GLAM project. The global MODIS-NDVI archive at GLAM serves as powerful tool for monitoring year-to-year crop conditions worldwide. Figure 3 displays South Africa’s NDVI anomaly composites for April 6, 2011 and April 5, 2012, where last year’s NDVI anomaly image shows above-average vegetation greenness but this year’s NDVI anomaly image shows below-average crop conditions.

Figure 3. Last year’s crop conditions are above-average (NDVI anomaly image on left), and
this year’s crop conditions are below-average (NDVI anomaly image on right).

NDVI values for all cropland pixels within each provincial administrative unit are extracted from the 250-meter resolution MODIS-NDVI imagery provided by the USDA/NASA GLAM project every 16-days. The growing season’s NDVI time-series graphs, or profiles, for the current year, previous year, and 10-year average are then compared to each other for determining relative crop conditions and yields. These NDVI time-series graphs are shown in Figure 4 for the four major corn-producing provinces of South Africa: Free State, Mpumalanga, North West and Gauteng. The four NDVI time series graphs for each province indicate that this year’s crop conditions are below last year and below the 10-year average for most months.

Figure 4. NDVI-time series graphs show below-average vegetation conditions
during 2011/12 growing season for three of four South Africa provinces.

USDA’s estimates of corn yields in South Africa during the past 20-years are shown in Figure 5, and it illustrates that strong El Nino events have reduced corn yields in South Africa for the past 20-years. However, corn yields during the past two consecutive La Nina years did not provide yields above the 20-year trend yield, with last year’s corn yields below the trend yield due to flooding, water-logging and excessive soil moisture. In addition, corn yields this year were below the trend and 5-year average due to below-average seasonal rainfall and a long dry spell during mid-February through mid-March.

Figure 5. Strong El Nino events reduced corn yields during the past 20-years,
but corn yields also below trend during the past two La Nina years (2010/11 and 2011/12).

USDA/FAS Crop Travel during March 5-8, 2012

USDA/FAS personnel traveled within South Africa’s corn-belt during the first week of March, 2012 (refer to Figure 6), when most of the crop was in the grain-filling stages but some late-planted crops, especially in the west, were in the pollination stage.

Figure 6. FAS crop travel route in South Africa from March 5-8, 2012.

Crop conditions in early-March ranged from poor to good, with the poor areas being very dry and some fields reaching the permanently wilted stage. All major corn producing provinces had a mixture of good and poor crops, with the majority of the crop being severely stressed in early March by a long dry spell that occurred from mid-February to mid-March. The location of good and poor crop conditions in late February are shown in the percent median WRSI (Water Requirement Satisfaction Index) images in Figure 7.

The percent median WRSI images in Figure 7 serve as another tool to monitor year-to-year relative crop yields and the affect of dry spells on potential crop yields. The WRSI is a water-balance model originally developed by the Food and Agricultural Organization (FAO) in the 1970s that monitors crop performance based on the availability of rainfall during the growing season. The WRSI is a water balance method that accounts for water gained by precipitation and lost by evapotranspiration in daily time increments, with the plant’s total water consumption estimated and accounted for during the entire growing season. The final WRSI model output is a crop "water satisfaction index" where an index of 100 indicates an excellent crop due to adequate rainfall and an index of 50 indicates crop failure from lack of rainfall or water for the crop.

The WRSI images in Figure 7 compare the percent median WRSI conditions during the current year, with above median crop conditions shown in green and below median conditions shown in brown. Correspondingly, the WRSI percent median images in Figure 8 illustrate how the dry spell from mid-February and mid-March 2012 greatly reduced crop yield potential for all provinces within South Africa.

Figure 7. Good crop conditions during last year’s La Nina (left),
and poor crop conditions during this year’s La Nina (right).

Figure 8. WRSI geospatial model indicates crop conditions turned from average
to below-average during the dry spell from mid-February through mid-March.

Figures 9 and 10 show photos taken on a farm located between Hoopstad and Wesselsbron, South Africa, where the WRSI model will not apply because a shallow ground water table is located 1.2-meters deep. The farm land prices in this region have greatly increased over the years, because a clay hardpan layer was discovered and it creates a shallow water table that extends for miles and provides good crop yields during dry years. The clay hardpan prevents soil moisture from being drained away allowing the crop’s root zone to receive soil moisture during dry years. The WRSI model does not apply to this region since the WRSI model assumes all rainfall water will be drained away from the root zone by deep percolation.

Figure 9. Crops grown on sandy soils that are easily eroded by wind.

Figure 10. Shallow water table above clay hardpans,
located approximately 1.2 meters below the ground surface.

Current USDA area and production estimates for grains and other agricultural commodities are available at PSD Online.

Other FAS Related Links

FAS Annual GAIN report for South Africa’s Grain and Feed (2/16/2012)

USDA/NASA GLAM (Global Agriculture Monitoring) Project with global MODIS and NDVI Imagery