I find this thread really interesting. I am currently developing an algorithm to do just that within the Soil and Water Assessment Model. My current geographic area is one where sowing is controlled by temperature but delayed by excess moisture. One factor that does not seem to be part of the discussion here is that farmers have limited resources and cannot sow all their land on the same day. Reasons include anything from equipment that limits how much land can be sown at once, lack of labor, to other personal reasons such as health or financial reasons. Thus farmers do not always plant (and that is true for any field operation) at optimal times. At a watershed scale, one can take that into account by limiting the amount of land that can be sown on the same day.

Good day Brian.

Not put to laugh, but it could be that many farmers choose the sowing date based on the stage of the moon, as for thousands of years.

Regards

From my experience there is heavy priority on not working the soil when it is too wet but still strong emphasis on sufficient moisture and likely ongoing moisture for germination. Days wait after rainfall really varies mainly based on amount of rainfall, preexisting soil moisture and local knowledge of the behaviour of the soil type and could be judged as I describe below.

This is very variable depending on crop, soil type, region and many other factors including the farmer. Optimal topsoil moisture for germination is something that is approaching but not past field capacity (FC) and this needs to remain for a sufficient period of time (e.g. 5 or 7 days or longer depending on the crop species) for emergence and establishment. In terms of percentage of field capacity this could be say around 60% of FC (perhaps at least 50%) or more. Anyway it sounds like you already have a thorough understanding of the theory approach.

Good farming practice certainly tries to avoid tractor traffic and cultivation/sowing when the soil is too wet to avoid compaction and other structural damage. Farmer assessment may simply be a quick visual assessment and seeing how the soil responds to some work but hopefully includes at least some digging, feeling the soil, rolling the soil into a "sausage" in their hands to ensure that there is reasonable moisture and not excess (especially above FC) or they may use more technologically advanced methods (this is becoming more common).

If contemplation of sowing is coming after a dry period followed by some period of rain, some farmers will assess whether the rain has been sufficient to moisten the soil and reach the depth where the soil was already moist (this connection is important for reducing the risk in sowing.

In some regions, there may be a practice of dry sowing (sowing in dry soil to avoid germination until there will be a hopefully reliable period of wet weather) in which case farmers will wait after some rain for the soil to dry out sufficiently to ensure there is not premature germination which is then hit by dry conditions). Sowing into the dry soil then proceeds in advance of a coming wet period.

Dr. Jenkins,

Thank you very much for your comprehensive discussion, very useful for me.

In our work, we are going to implement some kind of sowing date estimation procedure as part of a crop model. I reviewed various types of algorithms proposed for specific regions or for continental or global applications (e.g. Waha et al., 2012; JRC, 2007). My main question is about precipitation / soil moisture, as it seems there is no certainty about how farmers should react to them. Some Algorithms choose the sowing date on a wet day, while I believe many farmers (if not all of them) do not sow on a wet day, as you thoroughly discussed. Others, propose values as optimum soil moisture (like the values you mentioned) at sowing, but there is no theoretical or experimental background for their choice.

On the other hand, we are not willing to change the problem from estimation of sowing date into optimization of sowing date. I believe the latter case would not be close to farmer practices as they do not have any knowledge of the future climate. So, we must just rely on the past weather conditions and model simulations on the days before sowing.

So, let change the question into this: are you aware of any comprehensive study aiming at reviewing various sowing strategies (based on field experiments or model simulations), specially for wheat and across Europe?

Regards.

Out of curiosity, could you use remote sensing to estimate emergence instead of needing to rely on a sowing date? 

Hi Dr. Ababaei,

I fully agree with the view of Dr. Jenkin regarding soil moisture status required for germination. Of course soil moisture during sowing time is an important factor. I am not an agronomist but so far my knowledge goes based on my experience there are some other determinants that play important role in determining sowing time of a crop. The major determinants are temperature, rainfall/water regime, photo period sensitivity of the crop etc.

Temperature: If a crop is temperature sensitive we must have to consider temperature of the growing season and the sowing time should be adjusted based on the temperature requirement of the crop. For example, we cannot grow winter crops like wheat, mustard, potato, winter pulses etc. in summer season and summer crops in winter season. Wheat is more susceptible to high temperature during its reproductive phase, starting from flowering to maturity, which must be considered in determining its sowing time. Maize is a temperature insensitive crop and can be grown both in winter and summer season.

Rainfall/water regime: Rainfall/water regime is a major determinant of sowing time of any crop. For example, rice grows well in wet lands but the upland crops like wheat, mustard, potato, pulses etc. are sensitive to water logging and cannot survive at all under long time water logging. For pulses, if there is rain during reproductive phase, especially during flowering stage the crop stops flowering and turns again into vegetative phase. So we must consider these factors in determining sowing time of any crop.

Photo period sensitivity: Some crops/varieties are photo period (day length) sensitive and some are not. For the photo period sensitive crops/varieties specific day length is a must for turning from their vegetative phase to reproductive phase. There are many local/traditional rice varieties which need specific day length for flowering. They flower only in a particular month of the year; sowing time is immaterial for those varieties. But most of the modern rice varieties are photo period insensitive and they don’t need specific day length for flowering.

These are the major determinants of sowing time of any crop. Definitely there are some other environmental factors like humidity, sunshine, cloudiness etc., which might also play role in determining sowing time of a crop but they may not be the major factors. Overall, crop requirement of the environmental factors must be considered in determining sowing time of any crop.

Hi Dr. Ababaei,

I fully agree with the view of Dr. Jenkin regarding soil moisture status required for germination. Of course soil moisture during sowing time is an important factor. I am not an agronomist but so far my knowledge goes based on my experience there are some other determinants that play important role in determining sowing time of a crop. The major determinants are temperature, rainfall/water regime, photo period sensitivity of the crop etc.

Temperature: If a crop is temperature sensitive we must have to consider temperature of the growing season and the sowing time should be adjusted based on the temperature requirement of the crop. For example, we cannot grow winter crops like wheat, mustard, potato, winter pulses etc. in summer season and summer crops in winter season. Wheat is more susceptible to high temperature during its reproductive phase, starting from flowering to maturity, which must be considered in determining its sowing time. Maize is a temperature insensitive crop and can be grown both in winter and summer season.

Rainfall/water regime: Rainfall/water regime is a major determinant of sowing time of any crop. For example, rice grows well in wet lands but the upland crops like wheat, mustard, potato, pulses etc. are sensitive to water logging and cannot survive at all under long time water logging. For pulses, if there is rain during reproductive phase, especially during flowering stage the crop stops flowering and turns again into vegetative phase. So we must consider these factors in determining sowing time of any crop.

Photo period sensitivity: Some crops/varieties are photo period (day length) sensitive and some are not. For the photo period sensitive crops/varieties specific day length is a must for turning from their vegetative phase to reproductive phase. There are many local/traditional rice varieties which need specific day length for flowering. They flower only in a particular month of the year; sowing time is immaterial for those varieties. But most of the modern rice varieties are photo period insensitive and they don’t need specific day length for flowering.

These are the major determinants of sowing time of any crop. Definitely there are some other environmental factors like humidity, sunshine, cloudiness etc., which might also play role in determining sowing time of a crop but they may not be the major factors. Overall, crop requirement of the environmental factors must be considered in determining sowing time of any crop.

Dear Dr. Sulik,

Hi. Our goal is to run model simulations and the model will simulate the emergence. So, we need to determine sowing date, instead. The only way I can think of is to use remote sensing (and I don't know how) to estimate the emergence date and use that data to calibrate sowing date. If I got this right, please elaborate more on your idea.

Regards.

Dear Dr. Satter,

Thank you for your comprehensive answer. The influential factors are not fully understood. As an example, I read a paper (I don't have access to my files right now, so I cannot add the reference!) in which the authors concluded there were weak correlations between farmers' choices on sowing dates and many meteorological factors, and even with yield. So, it is not easy to formulate these factors and this is exactly what we need to do, formulation. Many criteria have been proposed, like DEPTH by Raes et al. (2004) and Waha et al. (2012) others. But, I could not find any theoretical background for them and hence I cannot use them for other locations or crops.

Regards.

Sowing date for many crops has been determined by farmers as results of their experiences. The confirmation of the sowing dates have been verified by agronomical studies as well. The experience showed that the period between 15 October and 30 November is the most suitable sowing date for most of the winter sowing crops, wheat, barley, lentil etc. And 20 April and 20 May for warm climate plants, corn, cotton, soybean etc. in Southeastern of Turkey. Early and late sowing dates can be led to serious adverse effects.

Regards...

Usually there is no standard method of sowing based on the moisture content in the field of our region.  Even,We practiced the direct sowing of paddy seeds in the field assuming that the monsoon will come after some days. One of the most important criteria for sowing paddy are temperature, rainfall, photo-period sensitivity of the crop etc.

Dear Dr  Ababaei: .. Answering  your  questions. How can we estimate sowing date based on soil moisture and real farmer actions?

In my area based on the experience of Aymara farmers growing quinoa in the altiplano, they do the following: soil is sandy loam, soil water accumulate for two years, the soil moisture measured hand, there are various indicators, one is take a soil sample and no free water appears on the soil, but in the hand is a small spot.  Farmers do not sowing  if there is adequate soil moisture. Based on this we have has monitored the level of soil moisture in time and related to the rains, however there is a predominant factor in the highlands is high atmospheric demand and wind. The ideal would be to build a model based on these indicators and correlate it with the practical activity of farmers.

As discussed by some of the above answers and the original question, there are many factors at time of sowing that will affect successful germination, establishment and yield. The question about decision around soil moisture is a key one as this can sometimes mean the difference between yield and no yield (worse case scenario is seed is sown and starts germination but the soil dries too much before the coleoptile and roots have emerged from the seed) or even a decision to not plant that season if no rains are expected.

Soil moisture present can affect the sowing depth chosen, and the wheat variety chosen (if sowing deep, consider varieties with longer coleoptiles but accept that there is often a low yield potential). The topsoil needs to be moist enough and also the moisture should connect with moister soil in the profile. To further complicate the situation, sowing strategy can be to dry sow as long as the soil is not wet enough to start germination (wheat can start and stop germination but only to a certain extent) in anticipation of coming rain.

As Jose Delatorre-Herrera says, the model needs to incorporate all the parameters that will keep a soil moist or make it dry. Also risk of crusting, texture of soil, sowing depth, variety etr etc. There is plenty pf agronomic advice around (Australia, USA etc) about this but perhaps due to the complexity, there doesn't seem to be a published model around so it seems worthwhile for Dr Ababaei to construct a potential one for his area. Bon chance!

Edited the above for worse case scenario being if the seed is sown and starts germination (due to sufficient moisture) but then the soil dries too much before the coleoptile emerges from the seed.

The Aymara farmers sow quinoa in September, at a depth greater than recommended, cross the layer of dry soil up to the wet area, so as to ensure germination and emergence. Obviously this will depend on the depth of the moist layer. However they have a schedule, according to their ancestral wisdom, they know they should rain in December as fatal term to grow quinoa.

Regards

I find this thread really interesting. I am currently developing an algorithm to do just that within the Soil and Water Assessment Model. My current geographic area is one where sowing is controlled by temperature but delayed by excess moisture. One factor that does not seem to be part of the discussion here is that farmers have limited resources and cannot sow all their land on the same day. Reasons include anything from equipment that limits how much land can be sown at once, lack of labor, to other personal reasons such as health or financial reasons. Thus farmers do not always plant (and that is true for any field operation) at optimal times. At a watershed scale, one can take that into account by limiting the amount of land that can be sown on the same day.

My interest is to use long term rainfall data withe the help of INSTAT+ software 

because of rainfall distribution for specific location vary with years, but farmers in specific location usually adapt to sow when they believed to be optimum time, use specific variety and most uniform production system, 

Rainfall is the one that determine soil moisture, in my area one year the rain comes early another late , vary, use of long term rainfall gives you high probability of having possibility of determination of optimum sowing dates, even transplanting dates for vegetables for the rain fed agriculture 

Thus please have this softare + guideline and I recommend you to use INSTAT

Dear Dr. Baffaut

Thank you for sharing your experience. I total agree with you and this is a fact.

I have some experience with SWAT and was wondering what algorithm you are going to implement in SWAT? Are you going to incorporate sowing moisture status as part of the problem, or you are willing to keep it simple and just use precipitation and temperature?

I by now implemented two algorithms in R, the JRC (Savin et al. 2007) and Waha et al. (2012). If you are using SWAT in R, maybe I can share with you what I have done.

Regards.

In principle, it must be considered both soil moisture and temperature to set the date of planting a crop. In addition, it is also important to consider ground cover (stubble) since according to the type and amount of stubble there, both variables change. The tempartura is lower and humidity is higher, in general. As we know, each crop has its base temperature and thermal time to germinate, so if the soil moisture is not limiting is possible to estimate the length of time between planting and emergence. Another features to consider is the physyiological quality of seed, which affect their behaviour and final number of seedlings established.

Dear Brian

Unfortunately the precipitations are like the lottery.

The rainfall, in the long and medium term, can be estimated by the likelihood.

Through the likelihood, I can estimate what I expect in a time interval but I can not easily tell what will happen tomorrow or the next day and how many mm will rain.

In the very short term the weather forecasts are used, based on direct surveys.

So I think it would be a good result for a specific place:

-Through a theory:

• defining a time interval in which sow, rather than a date.

- Based on surveys of what happened:

• Define a time interval in which sow, using the occurrence of a rainfall event as a trigger

or

• Give a deadline within which to sow, trusting in the occurrence of what has not happened yet.

Greetings

Dear Behnam,

We have tried to generate the sowing dates using remote sensing NDVI data over India, It is a simple algorithm based on the weekly NDVI composite data. For more details you can  refer the  - "Extracting Regional Pattern of Wheat Sowing Dates Using Multi-spectral and High Temporal Observations from Indian Geostationary Satellite"

Dear all,

it is possible to measure soil moisture by remote sensing using a L-band radiometer. It measures surface soil moisture (5 to 15cm).

My former University had developed some of these instruments. Look for Prof. Adriano Camps papers and ARIEL radiometer instrument.

Greetings