Do trees senesce?

August 25, 2021 • 10:45 am

“Senescence” is defined as “deterioriation with age”, and, in biology, usually refers not to an accumulation of external injuries over one’s life, but to an inherent process of going downhill physically and physiologically, as many of us are experiencing now. In humans, things start going wrong, you get creaky, or your mind might go and diseases of age will occur. The evolutionary reason why animals aren’t immortal are not completely clear, but there are evolutionary theories. (One of them is that genes that make us reproduce early, but have the side effect of hurting us as we’re older, will be subject to positive natural selection.)

But what about plants? Do they senesce, too? Well clearly some plants are genetically programmed to live only a year or so, but the paper at hand is concerned with trees. Do they senesce, too, or do they have a limited life span simply because, over time, bugs, fires. lightning, climate change, and so on eventually cause them to die?

This new paper, which has sixty-one authors (!) says “yes”, but answering a bit more limited question: do larger trees have reduced fecundity (i.e., seed production)? Since we don’t know the age of a tree without counting its rings, the authors use size as a surrogate of age, though in many species the correlation between tree size and age is not tremendously strong. Another issue is that even in single trees, much less species, seed production varies tremendously from year to year, being huge in so-called “mast years.” Every squirrel knows this. So you can’t just look at seed production in one tree in one year, or even in an entire species in one year, to find out if it goes down as a tree ages.  (“Size”, by the way, is estimated as the diameter of the trunk.)

We’d like to know this for several reasons: ecological prediction, use of trees to produce fruit or nuts (do they need to be replaced at a given time? and, if so, when?), and for studies of what truncates life spans in various organisms.

Up to now the assumption has been that log of fecundity goes up with the log of a tree’s diameter, but the data from various species has been conflicting. This new paper in Proc. Nat. Acad. Sci. USA uses data from 597 species of trees, with measurements taken from 585,670 individual trees and 10,542,239 tree-years (this explains why there are so many authors). The conclusion? Yes, in general fecundity declines with tree size. (Fecundity is measured via standardized methods of seed sampling.

Click on the screenshot below to read the paper, or get the pdf here; the full reference is at the bottom.

The results are simple, and can be shown in one graph (below). Of all the tree species tested, 63% showed a decline in fecundity (relative to trunk diameter) as they age (actually, as they get bigger), while another 17% show an increase in seed production that slows down as the tree ages. The conclusion, then, is that “80% of the 597 species tested here show declining rates of increase in fecundity with diameter. . . and 63% of the total actually decrease.” They consider this “empirical evidence for declining fecundity with size”, ergo with age.  In other words, the reproductive effort of trees, like that of many animals, slows down as the organism ages. Trees get old and less functional.

Here are some figures showing that decline. Subfigures A-C are for temperate regions, and D-F are tropical regions. Each plot shows the standardized (by diameter) fecundity versus standardized diameter (see paper for how these were calculated), and each line represents one species of tree.

Plots A and D show a pattern of declining relative fecundity with diameter (age surrogate), and these have most of the data for both regions. Standardized fecundity is taken to be fecundity relative to maximum fecundity, which is why in A and D, it peaks at 1.0.

Plots B and E show a pattern whereby fecundity first increases with diameter and then, as the tree gets bigger, the rate of fecundity increase begins to level off (a “sigmoid” graph), showing that the increase in reproductive effort slows down as trees get bigger (and older).

Finally, plots C and F show a pattern in which standardized fecundity continually increases as the tree gets bigger. (It’s possible that if they kept measuring or found the very largest trees, the increase might slow down.)

Clearly, A and D represent most of the trees surveyed.

I’ve put the journal’s caption below the figure; click to enlarge it.

The relationship between fecundity and diameter for species in temperate (A–C) and tropical (D–F) regions, where diameter and fecundity are scaled as D/DoptD/Dopt and f(D)/f(Dopt)f(D)/f(Dopt), respectively. A and D exhibit type A species (fecundity eventually declines); B and E show type B species (sigmoid increase in fecundity); C and F represent type C species (continuous increase in fecundity). Line transparency is proportional to the 90% credible interval width across the diameter ranges, such that confident predictions are opaque, and vice versa. The percentages of species for each type of fecundity–diameter relationship are summarized in Table 1. DoptDopt is the diameter when maximum fecundity occurs.

So if anyone asks you if trees get old, you can tentatively answer: “Well, they appear to, at least insofar as older trees reduce their relative investment into seeds.”


Qiu, T., M.-C. Aravena, R. Andrus, D. Ascoli, Y. Bergeron, R. Berretti, M. Bogdziewicz, T. Boivin, R. Bonal, T. Caignard, R. Calama, J. Julio Camarero, C. J. Clark, B. Courbaud, S. Delzon, S. Donoso Calderon, W. Farfan-Rios, C. A. Gehring, G. S. Gilbert, C. H. Greenberg, Q. Guo, J. Hille Ris Lambers, K. Hoshizaki, I. Ibanez, V. Journé, C. L. Kilner, R. K. Kobe, W. D. Koenig, G. Kunstler, J. M. LaMontagne, M. Ledwon, J. A. Lutz, R. Motta, J. A. Myers, T. A. Nagel, C. L. Nuñez, I. S. Pearse, Ł. Piechnik, J. R. Poulsen, R. Poulton-Kamakura, M. D. Redmond, C. D. Reid, K. C. Rodman, C. L. Scher, H. Schmidt Van Marle, B. Seget, S. Sharma, M. Silman, J. J. Swenson, M. Swift, M. Uriarte, G. Vacchiano, T. T. Veblen, A. V. Whipple, T. G. Whitham, A. P. Wion, S. J. Wright, K. Zhu, J. K. Zimmerman, M. Żywiec, and J. S. Clark. 2021. Is there tree senescence? The fecundity evidence. Proceedings of the National Academy of Sciences 118:e2106130118.

22 thoughts on “Do trees senesce?

  1. So, less energy directed toward producing seeds and more energy directed toward the health of their wood? Makes sense to me.

  2. Comparing seed production versus age seems an appropriate way to measure senescence. But I am a bit suspicious about relying so heavily on size, since that brings into play several different variables that do not change at the same rate. I would like to see the results be cross-checked with something more independent to size. A tree produces new branches each year, and on each branch will be leaves and flowers. So one thing to compare in small (young) versus large (old) trees is leaf and later seed productivity of one-year-old branches in differently aged trees.
    That could be an interesting student science project.

  3. A biologically immortal species will eventually die off anyway, as the earth’s inevitable changes (climate, food, the emergence of other species etc.) will render it non-viable. And while that hypothetical species heads toward its doom, it will compete with its offspring, some of whom might be more fit, but who will die because the environment is already saturated with more experienced members of their kind. Species fitness therefore requires mortality.

    Unless we have life insurance and/or a will, dying is our last gift to our children.

    1. This doesn’t make sense to me either. There isn’t really any such thing as “species fitness”. There’s just individuals trying to survive and reproduce, and some are better at it than others. Immortal individuals would be better at that. So the problem is explaining why individuals don’t live longer than they do. The answer is not that there is some imperative for older individuals to get out of the way of younger individuals.

      1. Imagine a bird pair, both immortal, passing on that attribute to some/all of their offspring, who did the same. After a certain number of generations, the environment would be saturated with that species. A few might be lost to predation etc., but they’re not all dying within a generation or two, meaning new generations are competing with all the generations before. Now, imagine that this species all eat only one kind of fruit, and that fruit supply begins to dwindle, due to climate changes, a new species better at eating the fruit etc. Had the bird species continued to produce successful new replacement generations, some might have been born with a mutation enabling the consumption of a second food source, and they would have kept the species alive while the one-fruit variety died off. But because new, more fit generations are killed off by their forebears’ immortality, the species as a whole collapses.

        There might be other reasons that all creatures age and die. I’m just saying immortality wouldn’t be beneficial for any species in the long run.

        1. I think there’s some evidence for instances of group selection – which your hypothesis is an example of – but them tends to be fighting words around here.

          1. I’m not sure what I’m proposing qualifies as official group selection, which I understand as a situation where all members of a species possess a certain trait because it’s good for the species as a whole, even if said trait requires a sacrifice of individual fitness. What I’m saying instead is that immortality is a genetic deficiency on an individual basis. Yes, you could live and breed forever, and so will all your progeny, which sounds awesome, until you look at the mid-term consequences, specifically a devastating inability to adapt to a changing world.

  4. So… do fertilized seeds from older trees contain more genetic copying errors or fewer? I’ve no idea, nor how this would play out in the ‘contribution’ of older trees.

  5. Very interesting study. I wonder if fruit/nut-tree farmers have noticed that their older trees produce less fruit. Maybe the farmers aren’t around long enough to notice.

    1. I’m curious about that too. I’ve noticed that citrus farmers in my area remove existing trees and plant new ones on a relatively short time scale, when the old trees aren’t really all that big. It seems like they wouldn’t do that unless older trees were less productive, or maybe lower productivity/ area-investment in maintenance, care and harvesting.

      Seemingly contrary to that, I once rented a house that had a huge grapefruit tree in the back yard that produced an incredible amount of fruit. this tree was easily 40 feet tall and shaded about a 3rd of the back yard and the house. During season you had to go up on the roof and remove grapefruit for fear of overload. It was also some of the best grapefruit I’ve come across. I also knew of a mango tree that was similarly huge, bigger actually, and productive. The mangos were the best, like vanilla ice cream with exotic fruit.

      But those are just anecdotes.

      1. Thanks for the added context, anecdotes or not. It seems that the farmers would only go through that laborious process if it provided results. Maybe young orange trees are different producers than grapefruit or mango trees? Who knows?

        I miss having citrus around. I grew up in California, and we had a lemon and orange tree that produced like crazy- best orange juice or lemonade ever. Fresh lemons off the tree really do have a different (and better) taste than store bought. Having a grapefruit and/or mango tree that produced the fruit you describe would be heaven.

  6. I showed this to my wife, who has a considerable background in botany and molecular biology. She got quite exercised over the various possible problems. One being that at any given time, there will be branches in the tree that are years older than other branches. Its well known to fruit growers that trees lose productivity over time even though they are well tended and kept healthy. In that sense, tree senescence does not seem terribly new. She went on for quite a bit!

    1. Now I could be wrong, but my take on this is that they were attempting to show that aging explains, at least in part, the decline in fruit (or seed) production. There are many causes for declines in fecundity and they wanted to see if aging can explain some of it (indeed, if it occurs at all).

      I appreciate your wife’s consternation at a study of something that seems obvious, but it is not uncommon that scientists try to test, understand and quantify obvious things. It was also not clear if what fruit farmers see in domestic plants is reflected in the wild.

    2. Honestly, I am surprised by the whole thing. My observation was to the contrary, and I thought it was common knowledge, that in its full maturity a tree will devote all its photosynthetic output to reproduction. After having reserved it for growth during its earlier years, in order to get to canopy height.

      I’m specifically thinking about the locally native maple, which is a short lived, fast growing large forest tree that I have seen doing exactly that. What fruit trees may or may not do seems more to do with the breeders who select them.

  7. For trees with regular tree rings one can determine the age by boring a small cylindrical core out of the tree. I’m not sure if this harms or stresses the tree. Stressed trees tend to produce great amounts of seed.
    I gathered from George C Williams that if an individual can die, and hence sooner or later does, senescence is inevitable. and the greater the chance of dying, the earlier the senescence will kick in.
    [It is borne out by eg. some birds, that can escape predation, which can live pretty long (such as psittacines or fulmars), or trees with a low death rate, while animals with a high death rate (predation) age quickly, such as rats or opossums)]

  8. “This new paper, which has sixty-one authors (!)” – well, that’s another tree gone just printing their names…

    A few years back, the Times Higher Education printed an anecdote by someone who received an anguished email from the co-author of a CERN physics paper, which read in part along the lines of

    No, I’m not upset about being the 37th author, I’m upset about being the 37th author called Wong!

  9. My favorite senescent tree is to be found beside the Yellow Brick Road: “Are you implying my apples aren’t what they should be?”

  10. I haven’t read the article but, perhaps unwisely, will comment anyway. I think questions of growth geometry could explain some of the patterns shown as well or better than “senescence”. When a tree grows, tree volume and the mass of metabolizing tissue increase in proportion to the volume occupied by the plant. On the other hand, photosynthesis should be proportional to surface area and limited to half or less of the volume occupied by roots, trunk, limbs, twigs, etc. The surface-volume relationship leads us to expect that, as a tree grows bigger, there is less excess photosynthetic production per unit of volume available for investing in reproduction. If trees could get big enough, there should be a point at which everything photosynthesis produced would be necessary just to keep the plant alive through the night and no reproduction would be possible.
    Also, I suspect that the larger trees that were measured tended to live in high-biomass forests. These sites might be expected to have soil nutrients relatively depleted compared to sites where smaller plants (in clearings?) were measured. The reduced relative reproductive effort of large plants may simply be an ecological artifact of diminished soil quality.

  11. As expected, when cocoa trees (Theobroma cacao L.) age, “bean” yields decline. Old trees are often replaced with new plantings. The problem here, though, is that the farmer must wait for perhaps five years for full production again. That is why the Malaysian Cocoa Board recommends side grafting branches from young trees onto the lower trunks of old trees. In a few years, when the new branches are producing well, the main trunks are cut down. This way the farmer has no down-time for production. Rejuvenated trees are apparently just as young and healthy as its grafted branches. Thus it appears that the trunk and branches age faster than the roots.

Leave a Reply