Hunting dog behaviour is a key driver impacting harvest quantity and quality of truffles

Truffle seasonality and eco-physiological interaction

The average weight of 31.1 g observed from a total of 3180 harvested truffles (Fig. 1), is startling close to the 33 g recorded in another large dataset reported by Büntgen et al.⁷,. Moreover, a dataset of 1003 collected truffles of the related species, Tuber melanosporum, displayed an average weight of 34 g with no difference between two tested irrigation regimes¹⁹. Despite our expectation that truffle size is influenced by environmental variables such as precipitation, the convergence of datasets across different countries and irrigation regimes suggests that this may not be the case. Indeed, we see here that average truffle weight over the harvesting season does not change (Supplementary Fig. 1B) despite significant seasonal shifts such as average temperatures of 18 °C in March to 33 °C in July (Fig. 2C; Fig. 8). This de-coupling of truffle weight with climate counters expectation and is interesting from an ecophysiology perspective, but also suggests that cultivators may not be able to use irrigation to alter truffle-weight within orchards to meet market expectations.

We also explored if truffle weight is related to maturity, i.e. are bigger truffles also more mature? Previous authors showed no correlation between weight and maturity for T. aestivum⁷ nor the related species T. melanosporum²⁰ but within this dataset, we see that although all truffle weights can correspond to all maturity grades, there is a relationship. Lower truffle weights were most prevalent in the highest maturity categories and further, depth of collection positively correlates with maturity and negatively with weight. Here, we hypothesise that it is not weight and maturity that are entangled, but rather that fruiting/collection depth is the key driver of this relationship, and this is explored further in 4.2.

In terms of mycophagy, truffle damage was greater at higher maturity grades, and this is likely explained by the dependency of truffles on mycophagy for the distribution of colonies in time and space, as well as successful fruiting through mating-type interaction^4,5,6. Truffles use aroma to attract predation and the concentration of many volatiles increase with maturation²¹ as do the levels of possible non-aromatic chemoattractants²². As such, higher levels of truffle damage are expected with increased maturation.

Finally, it is widely accepted by truffle hunters/cultivators that the maturity of collected truffles increases as the season progresses, and here we provide evidence that clearly validates this observation. However, although the size of recovered truffles doesn’t seem to show a seasonal pattern, the quantity of recovered truffle does. This latter observation presents as the season ‘warming-up’ in terms of numbers of fruit bodies recovered, before peaking and again slowly declining: this displays an expected progression in-line with previous studies^7,19. Finally, despite the interaction of mycophagy and truffle maturity we don’t see a seasonal progression of truffle damage and it’s important to note that this is a proportional value and so the amount of mycophagy/damage is increasing in quantitative terms but the proportion of the harvest that is damaged remains the same. These are all important points for truffle hunters/cultivators and will help to inform behaviour ‘on the ground’.

Fruiting depth correlates with truffle quality – dog ethology

Garcia-Barreda et al.²⁰, using a total of seven hunt days per site, reported that fruiting depth does not correlate with truffle maturity. However, using a larger and broader dataset we report here, and for the first time, that fruiting/harvesting depth indeed has a significant causal influence on truffle quality (Fig. 8). Within this we see a conflict, deeper truffles were more mature (Fig. 3) and therefore of higher value, but they were also more likely to be subject to higher levels of mycophagy, a factor which reduces market acceptance²³. First, we explore why these relationships exist. Čejka et al.³, showed that on consecutive day truffle hunting within the same area, the second day’s crop was found at a greater depth. From a dog perspective scent penetration is impacted by soil type²⁴, but also soil depth²⁵ and therefore, shallower truffles are likely easier for a truffle-dog to detect. Framed in this context, the relationship we see here between truffle maturity and depth of fruiting can be explained by deeper truffles being less easily detected and therefore harvested – the dogs quickly detect the ‘low hanging fruit’ of shallow truffles whilst deeper truffles require more focus and time. The deeper truffles may not be located as frequently as the dogs fail to invest enough time and effort to their detection, focussing on a quick and easy reward or it may be the handler losing patience with the longer time needed to detect truffles and moving the hunting team along. Whether the driver is the dog or handler, we suggest that those deeper truffles remain in the ground longer with a greater chance of reaching an advanced stage of maturity. This greater ‘soil persistence’ time also explains the higher degree of damage by mycophagy we observed in deeper forming truffles. These two measures of truffle quality are therefore a likely artefact not of ecophysiology but of truffle dog or handler ethology. Beyond some proposed hypotheses by Čejka et al.³, dog and handler ethology has never been associated with truffle quality metrics and here we present a theory that likely links the two. This raises the practical consideration that if a hunter wishes to harvest deeper truffles more frequently, perhaps to reduce the recovered proportion that are damaged or indeed to reduce the quantity of fruitbodies that remain undetected⁸, dog-training can be targeted accordingly. These insights may facilitate an informed approach by practitioners, which could be widely beneficial and further suggest that the percentage of truffles that are collected from deeper in the soil profile may be a good proxy for the efficacy of a hunt team. Moreover, these results highlight the significant and serious need to address dog ethology in studies that depend on truffle collection for data analysis. However, although dog ethology is a likely significant driver of the relationships we observe here, there are other explanations that may also contribute. As discussed by Thomas and Thomas⁴ invertebrates that predate truffles occur throughout the soil profile and may concentrate in deeper zones to escape desiccation in arid environments. The study area is within one of the dryer regions of Europe and this may be a driver of increased mycophagy at depth, although countering this is the fact that we see no proportional change in damage across the whole fruiting period of March − July whilst temperatures increase. It should also be noted that most of the recorded mycophagy damage will likely have been caused by invertebrates, as vertebrates such as mammals are more likely to remove the target truffle from a location, collect from higher in the soil profile, or occasionally consume in its entirety^6,26,27.

Finally, even though we suggest that targeted dog training can help to recover deeper truffles before they are damaged and increase collection quantity, caution should be applied. For example, it may be the case that deeper truffles are more impactful from a spore distribution perspective and therefore increased retrieval may be detrimental to future harvests (see 4.1) although there are interventions to address this, such as adding spores in a targeted manner²⁸.

From a hunter perspective—hunt intervals and optimisation

We hypothesised that a longer hunt interval would result in a greater truffle harvest per hunt, and although we found that this was not the case (see Fig. 3) we do see that a longer hunt interval is correlated with greater truffle maturity. This latter point is explained by a longer un-disturbed maturation period for the truffle within the soil environment.

The relationship between hunt interval and site was strongly significant (Fig. 3). Here we see that the hunters intuitively dedicate more hunting days to the sites that are producing the highest quantity of truffles (see Supplementary Fig. 4) and thus display a good degree of hunt-optimisation in terms of spatial distribution of inputs. Finally, towards the end of the season we see that although truffles collected per hunt event are still high, they do begin to decline before the final collection which occurs with an abrupt end to the fruiting season (see methods). This holistic and detailed exploration across a broad spatiotemporal scale, allows us to present causal relationships of truffle eco-seasonal variation and the interaction of dog/hunter ethology (see Fig. 8).