The concern about soybean production in South America continues although the conditions have improved in the last days.
After a late start to the season, conditions are generally favourable for crop production in the main areas of the country. One state however has suffered more from drought and has concerned us about the impact on soybean yield potential.
We are talking about Rio Grande do Sul (highlighted in yellow on the map).
In the state, located in the extreme south of the country, the climatic conditions at the beginning of the cycle were unfavourable due to low rainfall and soil moisture in December. This has made us maintain a less optimistic view compared with CONAB (local source).
So far, when we analyse the data, such as NDVI and soil moisture, we see a strong similarity with 2020, when the yield was much below the trend year (-35%), due to the drought in December and in February and March. In 2019, the rainfall was enough to assure a good production (14% above the trend).
This similarity between 2021 and 2020 is causing us to maintain the less optimistic view so far.
Last year, in this period (from the last days of January to the end of the cycle), rainfall was low, which resulted in a decrease in soil moisture that had a negatively impact on the yields.
Fortunately, the forecast for the next days this year is different. According to the European model (ECMWF), Rio Grande do Sul should receive a high level of precipitation in the next days, maintaining the good development of the crops.
If this scenario continues, we could have a production higher than we are currently expecting.
However, and very important, forecast is about probability and not certainty. That is why we remain our view less optimistic than CONAB.
So far, CONAB has estimated the soybean production in Rio Grande do Sul in 19,86 million tonnes. Last week our estimative was in 13,90 million tonnes for the state.
For Brazil, Conab has estimated 133,69 million tonnes and USDA estimated 133,00 million tonnes. Geosys has estimated 130,65 million tonnes (average range).
It’s worthy to say that with the improvement of the conditions in the last weeks we increased our figures and we are considering a scenario already slightly better than last year. If this movement continues, we may increase our figures even more depending on the climatic conditions.
It is important to say that in the last few days another concern has arisen. The high precipitation in Parana has attracted attention and, if this scenario, of high level of rainfall, the state production can be revised and, consequently, national production. However, it is too early to make this statement, but we’re monitoring the situation.
US soybean production
So far, the local source has a higher soybean production estimate than our estimate in Brazil. This also happened for some time last season in the United States, where the USDA (local source) had a higher number than ours for a period. So, let’s see what happened in US and why we had a different view over the season.
When we analyse the data, we see a similarity with 2013 (which was a bad year in terms of yield).
In both case we have low level of soil moisture between July to beginning of October, in the case of 2013. In 2020, the soil moisture remained below average most of the season.
The drought that hit part of the soybean area made us more cautious about soybean production.
In mid-August the USDA estimated soybean production at 120.42 million tonnes and us at 116.33 million tonnes. The final estimate, made in December, the USDA estimated 113.50 million tonnes (decreased of 5.8% from the August report) and our estimate in December was 114.11 million tonnes (decreased of 1.9% in the same comparison).
In the first report of this year, the USDA made an adjustment to 2020 U.S. soybean production and estimated production at 112.55 million tonnes.
Despite the difference between our numbers, our last estimate was, considering the lower range, 111.84 million tonnes, so the final production in the US (according to the USDA) is within our expectations (considering our lower and middle range).