00:03 - Good morning. Steve.

00:04 - Sorry if you were expecting to see a soybean presentation,

00:08 - and we had to make a slight adjustment in the program.

00:12 - So the soybean presentation will be tomorrow.

00:15 - If you need to get those lies, you're not able to come in here tomorrow.

00:20 - Just reach out to me, and we'll be more than happy to share that with you.

00:24 - And, with this change in the program, we're going to talk a little bit

00:28 - about precision nutrition, which is somewhat related to soybeans.

00:33 - And how important is for the dairy men to manage the nutrients

00:38 - that he's feeding to dairy cows and trying to optimize nutrient

00:42 - deliver to the cows and make them more productive or more efficient?

00:47 - So today's agenda will be, a brief overview

00:52 - of the intersection between precision and dairy cow nutrition.

00:56 - What does that precision ward mean?

00:59 - It's been a buzzword, and we've listened to this word, a lot lately.

01:05 - So what does this really mean?

01:07 - And how we can, precisely feed cows in our dairy farms.

01:12 - What are the the common source of variation in dairy farms?

01:16 - And how can we try to manage those source of variation.

01:21 - What are the current technologies, applications,

01:24 - opportunities and limitations that we we may have

01:27 - with the current technologies and the path forward.

01:31 - Given all the, the the other topics that we're going to explain through

01:35 - all the presentation.

01:37 - So just to begin,

01:40 - introducing these concepts and making sure everybody's

01:44 - on the same page when we're talking about precision, their nutrition.

01:48 - I like to use this shooting range example.

01:50 - So we're going to get our rifles in and go out in the shooting range

01:55 - to get our sights ready for the, the next, deer season.

02:00 - And we're going to shoot five shots in the target.

02:03 - Right.

02:04 - And what does precision mean in this context?

02:07 - It means that when I'm shooting my my rifle,

02:10 - I'm getting now my targets close to each other.

02:14 - And right on that bullseye.

02:16 - So in this situation, I can say that I have both accuracy and precision.

02:22 - On the other hand, when I go out and I five the same five shots

02:26 - and I get a group of shots pretty close to each other, but far from the bullseye.

02:32 - In this situation I can refer to as being precise

02:36 - because my shots are pretty close to each other, but not accurate at all.

02:42 - On a third situation.

02:44 - My shots seem to be all over the place,

02:47 - but on average, if I cancel out these far points

02:52 - to one another, I'm going to be on on the bullseye on average, right?

02:56 - So in this situation, I can't say I have accuracy, but no precision.

03:02 - And the worst case scenario is that actually my shots are all over the place

03:06 - and I have neither accuracy nor precision because, when I cancel them out,

03:12 - they're not in the bullseye and they are too far apart from each other.

03:16 - So we can summarize this as precision being how closely results agree

03:21 - with one another, and the accuracy,

03:25 - with how close and the results agree with the standard value.

03:29 - And what are the standard values in a dairy farm.

03:32 - So we can start talking about precision nutrition.

03:35 - Well, the targets for a dairy

03:36 - man can be completely different depending on the farm.

03:39 - It may be increased milk production.

03:42 - It may be milk composition.

03:44 - I want to have higher butterfat and higher protein in my milk.

03:49 - It may be feed efficiency.

03:50 - My cows are eating too much and they are not producing as expected.

03:55 - So the targets may differ from system to system.

03:59 - And and it's our job as farmers.

04:01 - Dairy means and nutrition is to kind of understand what a target

04:05 - our or the targets are and then act upon the

04:10 - AI. I

04:11 - believe that if you are a dairy man and you have a dairy farm, and look at

04:15 - the dairy production data in your farm, you will see that there is quite,

04:19 - a high variation on the cow to cow productivity.

04:24 - So this is an example of a farm, 400 dairy cows in Indiana.

04:30 - Each of these dots represent the milk production of cows

04:33 - in different days in milk.

04:35 - So you can see that it's quite scattered, right?

04:38 - We have cows producing around 40 pounds, but

04:41 - we have cows producing 140 pounds.

04:44 - It's quite a big difference.

04:46 - And when we see these animals, we are most of the times formulating one single diet

04:51 - and feeding them to that group of cows as if the cows have the same requirement.

04:57 - And the reality is that they have not,

05:00 - they don't have the same requirements.

05:03 - So milk production can be quite different, in a full lactation.

05:08 - Milk components can also be different.

05:10 - So each individual cow has its own requirement.

05:14 - And when we talk about precision feeding, the idea that comes to mind is

05:19 - how can we feed these individual cows better?

05:23 - How can I target those?

05:25 - How producing animals

05:26 - and deliver the nutrients that that high producing animal need.

05:31 - So we

05:31 - can use technology to to help us with this.

05:34 - All right.

05:35 - We all have these champion cows in our farms.

05:39 - They are the ones producing the most milk.

05:42 - They are the ones that, logically

05:45 - deserve more nutrients than the low producing cows.

05:49 - But what kind of information would help us feed these cow better?

05:54 - The cow can talk.

05:55 - We can't ask her, how much nutrients she's needing,

05:59 - but we can use some precision tech knowledge tools to give us data

06:04 - that help us to interpret if they're causing a chart.

06:08 - Or an excess nutrient, availability.

06:14 - This is a survey back in 2015 that as dairy farmers,

06:19 - which precision technologies they were using in their dairies

06:23 - and their responses were quite variable.

06:27 - 50% of the farms were tracking daily milk yield.

06:31 - And daily milk yield helps me to understand

06:34 - the nutrient requirements of the cow.

06:36 - All right.

06:36 - We know how much milk she's putting out.

06:39 - We can estimate how much energy she's going to use to produce that milk.

06:44 - But from a nutrition perspective,

06:48 - how this sensors that would give me insights

06:50 - about the requirements of these cows are not being used in farms.

06:54 - Only 11% of the farms are measuring body weight.

06:58 - 10% of the farms are measuring rumination.

07:01 - So as a rule of thumb, every time that I am looking for data,

07:05 - that would help me, interpret and understand how to better feed the cows.

07:11 - That data is not available

07:13 - because most of the times, the sensors we have at the dairy farm

07:17 - is, is tailored towards interpreting reproductive,

07:21 - data or health issues.

07:25 - And there is a, another factor to precision nutrition.

07:29 - That precision nutrition goes beyond just the cow information.

07:34 - It would be nice if, if I had a QR code on the cow

07:37 - that I could scan and know exactly what she needs.

07:41 - That's one piece of the information I need to to feed that cow better.

07:46 - But there are so many other factors in a farm that may affect the precision

07:52 - of that ration that we need to take into consideration.

07:56 - And that's where or when things become very complicated.

08:01 - So I used to say that

08:03 - in a, in a farm we may have up to four different diets.

08:07 - Ideally all these different diets,

08:11 - or all these differences would be minimized with better management.

08:16 - But if we are not doing things right,

08:19 - we may have a diet that was formulated that's going to be different from the diet

08:24 - that God makes in the wagon, the diet that was delivered

08:28 - in front of the cow, and the diet that the cow actually consumed.

08:32 - So ideally we minimize variation across all these diets,

08:37 - but the reality is that it's gonna be variable.

08:40 - One farms.

08:43 - These are these

08:44 - are some of the examples right.

08:47 - Everybody's mixing diets on a daily basis.

08:50 - But it's very common to see worn augers.

08:56 - All right.

08:58 - It doesn't mix well.

08:59 - So the video on the right is the Warren Auger.

09:03 - Compared with the mixed, the mix on the left

09:07 - where we have adequate or a new twin auger.

09:13 - The mixtures are completely different.

09:15 - So we need to pay attention to these details because they're important

09:19 - for the implementation of precision nutrition.

09:22 - Another extreme example.

09:25 - I don't need to explain what is wrong about this.

09:28 - Right.

09:28 - But of course these diets not again is not going to get well mixed

09:33 - and the nutrients delivered in front of the cow will not be adequate.

09:39 - Here we have some sort of technology.

09:42 - Liquid feeds at some extent can be considered as a technology.

09:46 - But the way we are, including this liquid feeding the wagon,

09:51 - may change how these liquid feed is actually operating into the whole TMR.

09:57 - So do we have reasons to believe that one

10:00 - single pipe dumping liquid in the back of the wagon

10:04 - will make this liquid be completely mixed with the feed?

10:07 - Probably not. All right.

10:12 - Another example

10:13 - when I have forage in the wagon, but I don't know how to work with them.

10:19 - Or maybe the feed ingredient loading sequence is not right.

10:23 - That's what happens.

10:24 - The old fossils all gathered,

10:26 - and then the side or the edge of the wagon and it doesn't get mixed at all.

10:31 - All right.

10:32 - So these are all issues that we face on a daily basis,

10:35 - and that we can act to correct in order to increase

10:39 - precision feeding other common things, basic

10:43 - things that we face on a daily basis in dairy farms.

10:47 - If I go to the face of this silo

10:50 - and then I come closer to it, I see that there's

10:53 - a layer of of, seed

10:56 - that needs to be removed because it was, expired.

11:00 - Now, what's the economic impact of this is spoiled silage?

11:05 - We can estimate that.

11:06 - And in this particular case, it was enough to increase

11:11 - the cost of the corn silage in $3 per ton.

11:15 - If I'm feeding 70 pounds per day,

11:18 - that represents a 13 cent increase in the cost of the corn silage per

11:23 - cow per day, or a $50 increase in the cost of that corn silage

11:28 - for cow per year in a a thousand, cow

11:32 - dairy, herd that that's going to be huge in a 500 cow dairy herd.

11:38 - That's going to be huge as well.

11:39 - So these are things that, needs to come

11:44 - before the implementation of precision technology and precision feeding.

11:49 - And not to mention

11:51 - that the quality of the material is going to also be affected.

11:55 - I'm removing the spoiled silage,

11:57 - but everything that is left in the silos is still compromised.

12:01 - The nutrient composition is not going to be the same.

12:04 - All right.

12:05 - So this is just an example of how variable it can be.

12:09 - The nutrient composition of, the same forage in the same silo

12:15 - but collected in different spots.

12:18 - All right.

12:19 - This is screwed protein, in alfalfa silage.

12:23 - So we may have a crude protein that is 23.4% here, 24.5% here,

12:29 - but 19 19.4% on top.

12:34 - All right.

12:34 - So this is all related to how well we preserve our silage

12:38 - and how well we manage our forage.

12:42 - There are

12:43 - some equipment or new technology that we could use

12:47 - to try to track those changes and those, nutrient composition variabilities.

12:52 - So, for example, what is called the anterior technologies

12:56 - that can be installed in the harvester

12:59 - or in, in the, in the feed mixer when you're unloading silage,

13:05 - and putting it in a wagon, you could get a measurement of how much

13:10 - dry matter, crude protein, fiber, carbohydrates you would have on that,

13:16 - on that feed on that specific day in time.

13:19 - The problem is that diesel works very well

13:23 - in the research setting that everything is controlled.

13:26 - And we we, we, we account for all sources of variability.

13:30 - But when you put

13:31 - these out there in the field, there are so many different corn silage,

13:35 - alfalfa silents with different mixtures, with different nutrient composition,

13:41 - that the technology itself cannot be generalized

13:45 - accurately to all kinds of situations, so that the technologies available.

13:50 - But this is a perfect example of having a technology

13:54 - that may not be, helpful.

13:57 - In the field at least not yet.

14:02 - When we come to feeding cows.

14:05 - Most of the time we are feeding the in of cows.

14:10 - All right. Using a total mix ration.

14:13 - And we now have some ways

14:17 - to estimate or predict dry matter intake of these scouts.

14:22 - So if I know the amount of feed that I'm.

14:24 - And I'm dropping in front of the cow in the morning and next morning,

14:29 - I collect whatever refusals or weight bags

14:32 - I have from from the past 24 hours.

14:35 - I can estimate that then dry matter eating.

14:38 - The problem is this is labor.

14:41 - And, you know, most farms will not do this

14:44 - because they don't see a lot of value in, in collecting that type of information.

14:49 - The reality is that this is probably the first limiting data that we need

14:54 - as nutrition is to actually start fine tuning precision nutrition in their farms.

15:00 - So understanding dry matter intake and feed and feeding variability,

15:06 - is something that can really help the implementation of precision nutrition.

15:11 - This is just an example.

15:13 - Don't worry about these numbers.

15:15 - There is more to read, but this is a study

15:18 - that looked into the deviations between what was the recommended diet.

15:23 - So it's the same as your diet and saying you should eat 1 pound of steak per day.

15:29 - And in the end your week 2 pounds of steak per day.

15:33 - So that's the deviation you are deviating at 1 pound

15:36 - relative to what was recommended.

15:39 - They measured these in in commercial farms.

15:42 - And and they observed that deviate portions in the,

15:46 - in the predict or not predicted, but the recommended amounts

15:51 - and the actual amounts that were makes for each of these feed ingredients.

15:56 - Non grain silage. Grain silage hays.

15:59 - They were always deviating from the recommended amount.

16:04 - So the zero line would be the recommended.

16:06 - Whatever is going up is being overfed.

16:09 - Whatever is going down is being underfit.

16:12 - And thus it has implication,

16:15 - with milk production and performance of the cows.

16:19 - They found that, yes, not always feeding more than than

16:23 - the expected amount will result in improved milk production.

16:28 - Actually, the best milk production was right on the spot of 100%,

16:33 - meaning that exactly the amount I I was recommended to feed.

16:37 - I mixed and delivered to the cow.

16:40 - And there is a huge economic impact on this type of situation,

16:46 - because most of the times we are feeding cows, to provide

16:50 - a little excess of nutrients so we can guarantee that

16:53 - those high producing animals are getting all the nutrients that they need.

16:58 - So more than, a milk production effect, we may have a huge economic impact,

17:05 - similar data, but looking into nutrients

17:09 - and not always deviations in starch, fat and the for protein

17:14 - were, positive in terms of dry matter

17:17 - intake responses or milk or milk.

17:21 - Again, not always.

17:23 - When we are overfeeding nutrients, we're going to have positive responses

17:28 - like how can we measure individual intake?

17:32 - We still have not develop

17:35 - any sort of accurate technology to measure that.

17:38 - There are some attempts now with, rumination collars.

17:43 - So these are collars we put in the cows and we can track how how long they spent

17:49 - ruminating, eating, and tuning on, on their cud per day.

17:54 - And then we can use that information to try to predict their total dry

17:58 - matter intake.

17:59 - Right now, these rumination collars are a very useful tools

18:04 - to track health and reproduction.

18:07 - And we're trying to develop this to track also dry matter intake

18:13 - and rumination,

18:14 - something very important that correlates well with performance.

18:17 - So for example, cows that increase their rumination times from 0

18:23 - to 120 minutes per day from day

18:26 - 1 to 6, and milk, they had higher milk production.

18:30 - So cows that ruminate more, they probably are eating more,

18:33 - they probably are producing more milk.

18:35 - And that's what these data is showing us.

18:38 - So it's important to have those numbers and use them

18:41 - in order to refine precision nutrition.

18:43 - And this is very important.

18:45 - Absolutely no precision technology

18:48 - can compensate for fundamental nutrition and management issues.

18:52 - Cows running out of feed that cannot be compensated by precision technology.

18:58 - So it's the basics done well that will guarantee

19:03 - the implementation of better precision nutrition.

19:06 - All right. And we need to measure to manage.

19:10 - So coming into an end of the presentation of

19:14 - how can we optimize nutrient delivery to our cows,

19:18 - I've mentioned that, most of the times we are feeding group of group of cows

19:23 - and not really focusing on the individual, but could we feed cows individually?

19:29 - That may be an easy task when we have a taste of a dairy

19:33 - where I know the cows by name and I know exactly how much milk they're producing,

19:37 - so I just dump a little more concentrate

19:40 - to these cows to support their energy demands.

19:44 - Nowadays, again with technology, we may have robots

19:48 - that will deliver nutrients to cows based on their milk production.

19:52 - So we can use, statistics and computer models

19:56 - to identify cows and give them whatever they, they need.

20:01 - So, in other words, can we target

20:03 - individual cows within a beef herd?

20:06 - And that's what we've been trying

20:09 - to do, here at Penn State, using robotic feeders

20:14 - like these, where the cows come to, to eat some concentrates.

20:18 - And each of these cows have their own diet programed in a computer.

20:24 - But not everything works as we planned.

20:28 - All right.

20:29 - This is, footage of cows visiting those robots.

20:34 - At 11 a.m.,

20:36 - a couple of hours after being milked.

20:39 - And this is in the evening, right after coming back from the milking farm.

20:43 - So we we, in theory, have everything set up for these cows

20:47 - to receive the nutrients that they need, but in reality is that these cows

20:52 - will change behavior based on what we are doing with their feeding management.

20:57 - And now, instead of spending time laying down,

21:00 - ruminating and making milk, they are fighting

21:04 - is standing in front of these robots and not getting rest at all.

21:09 - And the implications on it may be, may be negative

21:14 - because these cows will likely decrease milk production and performance.

21:19 - So changes in feed feeding behavior needs to be something

21:23 - that we incorporate in our algorithms, in our technologies.

21:27 - Because these represent a limit, a limitation

21:31 - for the implementation of precision feeding.

21:36 - What is the,

21:37 - overarching goal, of all these precision nutrition and where

21:43 - we want to take this moving forward, we need to be able to integrate data.

21:48 - Right now, this is probably the most, important point

21:53 - that we need to develop when talking about precision nutrition.

21:57 - I need to be able to understand what is the individual

22:01 - cow needs through measuring milk production.

22:05 - Milk components, dry matter intake.

22:07 - I need to integrate that in a computer model

22:11 - that is considering now variability we have in in in the mixture of diets,

22:16 - in seed composition variability,

22:19 - I need to integrate feeding behavior because we are going to, change that

22:25 - feeding behavior at some extent as, as we manipulate the diet of these cows.

22:31 - And these needs to be a loop where everything feeds into a model

22:36 - that can accurately predict and the needs of the cows,

22:40 - and we can accurately deliver, the nutrients that these cows need.

22:45 - So couple take home message for you today.

22:48 - Deviations in dietary nutrients, ingredient composition and mixing

22:53 - may negatively affect performance and health of dairy cows.

22:57 - Again, overfeeding nutrients will not always result in better performance.

23:03 - Measuring dry matter intake is key.

23:06 - That's something we haven't been doing

23:08 - enough in farms, and that's probably number one.

23:12 - Information that we need in order to to implement precision feeding.

23:17 - And we need to adjust expectations for precision nutrition.

23:20 - Sometimes the results will not come come

23:23 - as increased performance but most likely will be observed

23:28 - as increasing come over feed costs, more profitability.

23:32 - And again there is absolutely non precision technology

23:36 - that can make up for years basic errors in the farm.

23:41 - So the basics done well with consistency is the first step

23:45 - towards improved precision nutrition.

23:48 - If you have any questions I'll be happy to take them now.

23:52 - But you can also reach me out at Leon martinez@psu.edu.

23:57 - And I would like to acknowledge, Penn State College of

24:01 - Penn State extension, rest of lab and USDA.

24:06 - Thank you so much.

24:10 - For. Questions.

24:15 - So we will see us again.

24:18 - Sorry.

24:18 - We had to adjust the the program a little bit

24:23 - because the guy was going to present about soybeans didn't come.

24:27 - So, that presentation will be tomorrow.

24:31 - If you're not able to come tomorrow, please send me an email.

24:34 - And we will be happy to share that presentation with you.

24:38 - Or I think you can watch the recording later.

24:41 - That will be available.

24:47 - All your questions.

24:54 - If not, I'll thank you for your attention.

24:57 - Thank you.