Master 2, specialty “Plant Health”. Duration: about 2 hours. Work in pairs.

This exercise builds on the model of the exercise Sources, sinks and assimilate transport in a virtual plant. The same plant now suffers from two apple diseases: apple scab (Venturia inaequalis) on the leaves, and European apple canker (Neonectria ditissima) on the stem.

At the end of this session you should be able to:

• describe the life cycles of apple scab and apple canker and the weather conditions that drive infection;
• identify, in model code, the elements of the disease triangle (host, pathogen, environment);
• explain how a leaf disease reduces yield (loss of functional leaf area, pathogen as a sink, defoliation), and how a stem disease blocks the transport of sugar and water;
• run disease experiments with a control, and quantify damage by comparing with a healthy plant;
• discuss control strategies (sanitation, resistant cultivars, fungicides) with the help of the model.

You do not need to be a programmer. Every change you are asked to make is a change of one value at the top of the file.

1. Start GroIMP and open the scab project (File → Open). Open the code in the text editor.
2. Each click on grow = one hour. Start a long run (run/loop button) and let it run to about step 600 (≈ 25 days) while you work on Parts 1 and 2. A run takes a few minutes.
3. The button harvestStats prints a summary of the situation: fruits, infection periods, scab infections, leaves shed, cankers, and the mean sugar content of the internodes. The model also prints this summary automatically at hour 600.
4. To start again: reset, or save the code (Ctrl+S).

What you see:

• 3D view: scab lesions appear as small brown spheres on the leaves, growing once they sporulate. Leaves with sporulating lesions turn yellow-brown; leaves above a canker wilt (pale olive). A cankered internode turns dark red.
• XL Console: rain days, infection periods, canker infections.
• Three charts:

Apple scab. The fungus overwinters in fallen leaves on the orchard floor. In spring, ascospores are released during rain and infect young leaves and fruits (primary infections). An infection succeeds only if the leaf surface stays wet long enough; the time needed depends on temperature (Mills table: about 9 h at 18–24 °C, but more than 20 h at 6 °C). After a latent period of 9–17 days, olive-brown lesions appear and produce conidia, which are splashed by rain to other leaves (secondary infections). Young leaves are very susceptible; they become resistant as they age (ontogenic resistance). Heavily infected leaves fall early.

Apple canker. Neonectria ditissima infects the wood through wounds: leaf scars at leaf fall, pruning cuts, cracks. Infection needs rain. The canker slowly girdles the branch; everything above it is cut off from the water and sugar supply, wilts and dies back.

Questions (answer in 2–3 lines each):

1. Q1. Draw the life cycle of apple scab, marking the primary and the secondary infection cycles.
2. Q2. Why do warning systems for scab use leaf wetness and temperature, and not rain alone?
3. Q3. How can a leaf disease (scab) increase the risk of a wood disease (canker)?

if (time % 24 == 0 && probability(RAIN_PROB)) {
    wetLeft = Math.max(wetLeft, irandom(WET_MIN, WET_MAX));   // rain: leaves are wet for some hours
}
if (wetLeft > 0) {
    wetLeft--;
    wetRun++;
    if (wetRun % wetHoursNeeded(TEMP) == 0) {   // enough wet hours at this temperature
        infectionEvent = true;
    }
}

wetHoursNeeded(TEMP) interpolates the Mills table (MILLS_T, MILLS_H at the top of the file).

• Q4. How many hours of leaf wetness are needed for an infection at 10 °C? At 24 °C? (Use the table in the code.)
• Q5. Which parameters describe the environment in this model?

During an infection period, every leaf is infected with a probability:

float scabInfectionProb(Leaf lf) {
	float sus = (lf[age] < SUSC_AGE) ? 1 - (float) lf[age] / SUSC_AGE : 0;   // young leaves are most susceptible
	if (sus <= 0) return 0;
	float zLeaf = (float) location(lf).z;
	float pressure = 0;
	// primary inoculum: ascospores from the leaf litter, decreasing with height
	if (time >= PRIMARY_START && time <= PRIMARY_END) {
		pressure += PRIMARY_INOCULUM * (float) Math.exp(-zLeaf / ASCO_HEIGHT);
	}
	// secondary inoculum: conidia splashed from sporulating lesions, mostly downwards
	for ((* src:Leaf, (src[nLesions] > 0) *)) {
		float d = (float) distance(lf, src);
		float dir = (location(src).z >= zLeaf) ? 1f : UPWARD_SPLASH;
		pressure += SECONDARY_INOCULUM * src[nLesions] * dir * (float) Math.exp(-d / SPLASH_DIST);
	}
	return (1 - RESISTANCE) * sus * (float) (1 - Math.exp(-pressure));
}

Ascospores are only released between PRIMARY_START and PRIMARY_END. direction is 1 if the source leaf is above the target leaf, and UPWARD_SPLASH (0.3) otherwise.

• Q6. Calculate the infection probability of a leaf aged 0 h at ground level, during the ascospore season, when no leaf is sporulating yet (PRIMARY_INOCULUM = 0.5, RESISTANCE = 0). Same for a leaf aged 75 h.
• Q7. Why is splash dispersal more efficient downwards?
• Q8. Which parameters describe the host, and which describe the pathogen?

Each infection is a small Lesion attached to the leaf. It stays latent for LATENT_PERIOD hours, then sporulates and grows. For every sporulating lesion:

float healthy = Math.max(0, 1 - lf[nLesions] * LESION_PS_LOSS);          // functional fraction of the leaf
lf[as] += calculatePS(...) * healthy * lf[water];                         // photosynthesis
lf[as] = Math.max(0, lf[as] - LESION_SINK * lf[nLesions]);               // the fungus uses leaf sugar

A leaf with SHED_LESIONS sporulating lesions falls, and so does a leaf older than 400 h. Every falling leaf leaves an open scar on its internode for WOUND_OPEN hours.

• Q9. List the three ways in which scab reduces the sugar available to the fruits.
• Q10. In the model, a shed leaf disappears with its lesions. What happens to it in a real orchard, and why does it matter for next year?

During an infection period, an open scar is infected with probability P_CANKER. The canker then grows by CANKER_GROWTH per hour, from 0 (healthy) to 1 (girdled). It reduces sugar transport through the internode:

float g = D_PHLOEM * (1 - i_top[canker]) * (1 - i_bottom[canker]);   // phloem exchange between internodes

and water supply: the water supply of a leaf is the product of (1 − canker) of all internodes between the leaf and the base. Photosynthesis is multiplied by this water supply, and a leaf whose supply falls below WILT_WATER (0.2) wilts and falls.

• Q11. What happens to a fruit above a girdled internode? To the leaves above it? To the leaves below it?
• Q12. How long does it take for a new canker to girdle the stem completely?

Look at your run (about 600 h), then click harvestStats.

• Q13. Describe the shape of the curve “sporulating leaves” in Disease progression. How is it related to the rain periods and to the latent period?
• Q14. Compare the total and the functional leaf area. At the end of the run, how much of the leaf area is still working?

A run takes a few minutes, so the experiments are shared out between the pairs; the results are pooled at the end. Every pair runs the control C1 and two other experiments.

Method: write your prediction first, change one value, save, run to 600 h, click harvestStats, record, and restore the original value. Because the weather is random, two runs with the same settings are never identical: compare the orders of magnitude, and discuss the variation between pairs.

Results table:

• Q15. Calculate the yield loss of the reference run compared with the healthy control (C1): (fruit number × mean size) of the reference, divided by that of C1. How much of the loss is due to scab, and how much to canker (C2, E1)?
• Q16. Which of E2–E8 reduced the disease most? Which reduced the yield loss most? Are they the same?
• Q17. In E6, why does a longer latent period reduce the epidemic so strongly? Think of the number of secondary cycles that fit into the season.
• Q18. Compare the healthy control (C1) with the reference run: did the disease reduce the number of fruits or their size more? At which moment of the season is each of them decided?

• Q19. Draw the disease triangle and place every parameter of the model in it (host, pathogen, environment).
• Q20. For each control measure, say which parameter(s) it changes in the model, and whether the model supports it: removing or shredding leaf litter; planting a resistant cultivar; spraying a fungicide before rain; pruning out cankers and protecting pruning wounds; avoiding excessive nitrogen (which prolongs shoot growth).
• Q21. List three limitations of the model (hint: fruit scab, temperature changes, the fate of shed leaves, resistance breakdown).
• Q22. In real orchards, leaf scab alone often costs relatively little yield in the current season. Why do growers nevertheless spray against scab many times a year?

For fast groups. A protective fungicide stops new infections for a few days after spraying.

1. Add three constants at the top of the file: SPRAY_INTERVAL = 168 (h between sprays), SPRAY_PROTECTION = 120 (h of protection) and SPRAY_EFFICACY = 0.9.
2. In scabInfectionProb, multiply the returned probability by (1 - SPRAY_EFFICACY) when time % SPRAY_INTERVAL < SPRAY_PROTECTION.
3. Count the sprays (one every SPRAY_INTERVAL hours) and compare the result with the reference run.
4. Harder: instead of spraying on a calendar, spray at the beginning of every rainy day. Which strategy needs fewer sprays for the same protection?

This exercise is not handed in, but the questions above cover the kind of reasoning expected in the exam: explaining disease cycles and the disease triangle, reading a simple model, quantifying damage with a control, and discussing control strategies. Keep your answers and tables as revision notes.