GroIMP wiki

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== Exercise: Apple scab and canker on a virtual plant ======

//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.

===== Learning objectives =====

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.

===== Schedule =====

| Part | Content | Time |
| 0 | Setup: open and start the model | 10 min |
| 1 | The biology in brief | 15 min |
| 2 | The model at a glance | 25 min |
| 3 | Observe the reference run | 15 min |
| 4 | Virtual disease experiments | 40 min |
| 5 | Synthesis: control strategies | 15 min |
| (Bonus) | Fungicide sprays | for fast groups |

----

===== Part 0 – Setup (10 min) =====

  - Start GroIMP and open the scab project (//File → Open//). Open the code in the text editor.
  - 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.
  - 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.
  - 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:**

| //Chart// | //What it shows// |
| //Disease progression// | healthy leaves, leaves with latent infections only, sporulating leaves, leaves shed by scab (cumulative), cankered internodes, and whether leaves are wet (1) or dry (0) |
| //Leaf area: total and functional// | total leaf area, and the part that still photosynthesises (without lesions, with water) |
| //Fruit growth// | size of every fruit over time |

===== Part 1 – The biology in brief (15 min) =====

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

===== Part 2 – The model at a glance (25 min) =====

==== 2.1 Weather and infection periods ====

<code java>
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;
    }
}
</code>

''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?

==== 2.2 Who gets infected? ====

During an infection period, every leaf is infected with a probability:
<code java>
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));
}</code>
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**?

==== 2.3 What the disease does to the plant ====

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:
<code java>
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
</code>
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?

==== 2.4 Canker and transport ====

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:
<code java>
float g = D_PHLOEM * (1 - i_top[canker]) * (1 - i_bottom[canker]);   // phloem exchange between internodes
</code>
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?

===== Part 3 – Observe the reference run (15 min) =====

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

^ Observation ^ Your answer ^
| Number of infection periods | |
| Number of scab infections | |
| Leaves shed by scab / wilted above cankers | |
| Number of cankers | |
| Number of fruits, mean fruit size | |
| When do the first sporulating leaves appear? | |
| Where on the plant are the first lesions? The last ones? | |

  * **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?

===== Part 4 – Virtual disease experiments (40 min) =====

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.

^ Exp. ^ What to change ^ Question behind it ^
| C1 | ''PRIMARY_INOCULUM = 0'' **and** ''P_CANKER = 0'' | Healthy control: what does the plant produce without disease? |
| C2 | ''PRIMARY_INOCULUM = 0'' only | Canker alone (scars from natural leaf fall) |
| E1 | ''P_CANKER = 0'' only | Scab alone |
| E2 | ''RAIN_PROB'' halved | A drier spring |
| E3 | ''TEMP = 10'' | A cold spring |
| E4 | ''RESISTANCE = 0.7'' | A partially resistant cultivar |
| E5 | ''PRIMARY_INOCULUM'' divided by 5 | Sanitation: leaf litter removed or shredded in autumn |
| E6 | ''LATENT_PERIOD = 300'' | Realistic latent period (about 12 days) |
| E7 | ''SPLASH_DIST'' doubled | Longer splash dispersal (wind-driven rain) |
| E8 | ''SUSC_AGE = 300'' | Leaves stay susceptible longer (vigorous shoot growth) |

Results table:
^ Exp. ^ Prediction ^ Infection periods ^ Scab infections ^ Leaves shed (scab / wilt) ^ Cankers ^ Nb fruits ^ Mean fruit size ^
| C1 | | | | | | | |
| Ref | | | | | | | |
| | | | | | | | |
| | | | | | | | |

  * **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?

===== Part 5 – Synthesis: control strategies (15 min) =====

  * **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?

===== Bonus – Fungicide sprays =====

For fast groups. A **protective** fungicide stops new infections for a few days after spraying.
  - 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''.
  - In ''scabInfectionProb'', multiply the returned probability by ''(1 - SPRAY_EFFICACY)'' when ''time % SPRAY_INTERVAL < SPRAY_PROTECTION''.
  - Count the sprays (one every ''SPRAY_INTERVAL'' hours) and compare the result with the reference run.
  - Harder: instead of spraying on a calendar, spray at the beginning of every rainy day. Which strategy needs fewer sprays for the same protection?

===== Before the exam =====

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.