Why there are more species in several small patches vs few large patches: a multi-species modeling approach

.title[
# Why there are more species in several small patches vs few large patches: a multi-species modeling approach
]
.subtitle[
## <em>British Ecological Society Macroecology Meeting, 5-7 July 2021, online.</em>
]
.author[
### Leonardo A. Saravia, [Universidad Nacional De General Sarmiento]
]
.date[
### 2021/07/06
]

---

# SLOSS

## is it better to have Several Small or Single Large habitat for conservation?

---

## On one side

---

## On the other

---
background-image: url(Figures/Saravia2018.png)
background-size: contain

## A Neutral - hierarchical model

---

## The Neutral side

---

## The Neutral side

---
background-image: url(Figures/NeutralHier02.jpeg)
background-size: contain

## The Neutral side

---
background-image: url(Figures/NeutralHier03.jpeg)
background-size: contain

## The Hierarchical side

---
background-image: url(Figures/NeutralHier04.jpeg)
background-size: contain

## An additional Neutral flavor

---

## What happens without habitat loss

???

The general pattern is that with more dispersal there are less species 
for smaller lambda (1.5 is near the extinction threshold) the increase in
dispersal produces and increase in survival

So with more dispersal there is a bigger probabilty for one species to dominate

---

## A Fragmentation experiment

---
## Fragmentation - Number of species

---
class:  center, middle

## The minimum patch size

---

## Fragmentation - Shannon Diversity

---

## Model results agree with Fahrig

--
### There is a minimum patch size that depends on lambda

--
### After that threshold fragmentation increase diversity

--
# why?

## Because fragmentation acts as a barrier to dispersal

---

## Model predictions

--
### For communities with a smaller average dispersal distance framgmentation will have strong positive effect

--
### For communities with a bigger average dispersal distance fragmentation will have no effect

--
### To my knowledge there is no experiment relating fragmentation to dispersal distance.

???

Prediction can be compared with experiments

---
## How to translate this to complex patch configurations

### The mean free habitat path (HP)

The mean distance from a random place inside the habitat to reach the degraded area

### The mean free degraded path (DP)

The mean distance from a random place inside the degraded area to reach the habitat

### HP and DP in relation to mean dispersal distance (DD)

???

These fragmentation patterns are delivertely simple so in real cases we need to have more operational 
parameters

---
background-image: url(Figures/FreePathDispersal01.jpeg)
background-size: contain

## A connected landscape - lower no. species

---
background-image: url(Figures/FreePathDispersal02.jpeg)
background-size: contain

## A less connected landscape - Higher no. species

---
background-image: url(Figures/FreePathDispersal03.jpeg)
background-size: contain

## The ideal fragmented landscape

---

# THE END